Remote dipole effects as a means to accelerate [Ru(amino alcohol)]catalyzed transfer hydrogenation of ketones

Article abstract of DOI:10.1002/1521-3765(20010401)7:7<1431::AID-CHEM1431>3.0.CO;2-L

A new generation of 2-azanorbornyl amino alcohol ligands for the catalytic transfer hydrogenation reaction of aromatic ketones was synthesized. Extremely active catalysts were formed by introducing a ketal functionality at the rear end of the ligand. Acetophenone was reduced in 96% ee at low catalyst loading, substrate to catalyst ratio, S/C 5000, within 90 minutes with isopropyl alcohol as the hydrogen donor. It was found that the dioxolane substituent in the ligand increased the turnover frequency, TOF₅₀, from 1050 h^-1 to 3000 h^-1 at an S/C ratio of 1000. Introduction of a methyl group at the carbinol carbon resulted in TOF₅₀ as high as 8500 h^-1. Transfer hydrogenation of a range of aromatic ketones was evaluated and found to reach completion within 30 minutes at room temperature, and excellent enantioselectivity, up to 99% ee, was obtained. A possible explanation for the enhanced activity was provided by density functional calculations, which showed that the presence of a remote dipole in the ligand lowered the transition state energy.

Full text of DOI:10.1002/1521-3765(20010401)7:7<1431::AID-CHEM1431>3.0.CO;2-L

FULL PAPER
Remote Dipole Effects as a Means to Accelerate [Ru(amino alcohol)]-
Catalyzed Transfer Hydrogenation of Ketones
Sofia J. M. Nordin,[a] Peter Roth,[a] Tibor Tarnai,[a] Diego A. Alonso,[b]
Peter Brandt,[c] and Pher G. Andersson*[a]
1
Abstract: A new generation of 2-aza-
norbornyl amino alcohol ligands for the
catalytic transfer hydrogenation reac-
tion of aromatic ketones was synthe-
sized. Extremely active catalysts were
formed by introducing a ketal function-
ality at the rear end of the ligand.
Acetophenone was reduced in 96% ee
at low catalyst loading, substrate to
catalyst ratio, S/C 5000, within 90 min-
utes with isopropyl alcohol as the hydro-
gen donor. It was found that the dioxo-
lane substituent in the ligand increased
the turnover frequency, TOF50, from
1050 h to 3000 h at an S/C ratio of
1000. Introduction of a methyl group at
the carbinol carbon resulted in TOF50 as
high as 8500 h . Transfer hydrogenation
of a range of aromatic ketones was
evaluated and found to reach comple-
tion within 30 minutes at room temper-
ature, and excellent enantioselectivity,
up to 99% ee, was obtained. A possible
explanation for the enhanced activity
was provided by density functional cal-
culations, which showed that the pres-
ence of a remote dipole in the ligand
lowered the transition state energy.
1
1
Keywords: amino alcohols ´ asym-
metric catalysis ´ reduction ´ ruthe-
nium
We reported earlier on the use of (1S,3R,4R)-2-azanor-
bornyl methanol and some analogues (1a ± c, Figure 1) in the
Ru-catalyzed transfer hydrogenation of aromatic ketones.[5]
Introduction
Asymmetric transfer hydrogenation of multiple bonds has
received great attention during the last two decades because it
is an important complement to asymmetric hydrogenation. A
number of Ru-, Ir-, Rh-, and Sm-based catalysts have been
reported[1] and high asymmetric inductions have been ob-
tained by using chiral N, O, and P ligands[1] with isopropyl
alcohol or formic acid/triethylamine as the hydrogen source.[2]
However, there is still room for improvement when it comes
to reaction rate and catalyst loading. The activity of the
catalyst is often just as important as high enantioselectivity,
especially for industrial applications.[3] Many of the previously
published catalysts for transfer hydrogenation that reduce
aromatic ketones at high rates are associated with elevated
reaction temperatures and high catalyst loading or low
enantioselectivity.[4]
Figure 1. Amino alcohol ligands evaluated in transfer hydrogenation.
The mechanism of the asymmetric transfer hydrogenation
with [Ru(arene)(aminoalcohol)] as catalyst has been inves-
tigated by others and us.[5b, 6] We now report on further
improvements concerning the activity and selectivity of
[Ru(p-cymene)(azanorbornyl methanol)]-based catalysts
with isopropyl alcohol as the hydrogen donor. To the best of
our knowledge, this is the first catalyst to effect highly
enantioselective transfer hydrogenation at low catalyst load-
ing and high rate. This was accomplished by a modification of
the azanorbornyl ligand backbone in a position remote from
the catalyst active site. By employing DFT, density functional
theory, we have been able to present a possible explanation
for the increased activity of the new catalysts.
[a] Prof. P. G. Andersson, S. J. M. Nordin, P. Roth, Dr. T. Tarnai
Department of Organic Chemistry, Uppsala University
Box 531, 751 21 Uppsala (Sweden)
Fax : (‡46)18-4713812
E-mail: phera@akka.kemi.uu.se
[b] Dr. D. A. Alonso
Departamento de Quimica Organica, Universidad de Alicante
Apartado 99, 03080 Alicante (Spain)
[c] Dr. P. Brandt
Karo Bio AB, Department of Medicinal and Computational Chemistry
14157 Huddinge (Sweden)
Chem. Eur. J. 2001, 7, No. 7
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0707-1431 $ 17.50+.50/0
1431

P. G. Andersson et al.
FULL PAPER
catalyst, ligands 2b ± d were synthesized by protecting the
hydroxyl groups with diethyl ketone, acetophenone and
dimethoxy methane, respectively. The enantioselectivity in
the transfer hydrogenations with ligands 2b ± d was almost
identical to that found with 2a, whereas the activity of the
catalyst was slightly lower for 2b and 2d (Table 1, entries
2 ± 5).
Results and Discussion
Synthesis of ligand 2a ± d. The bicyclic structure of ligand 1a
can be easily modified, making this ligand well suited for
optimization. The synthesis of ligand 2a ± d (Scheme 1) starts
Synthesis of ligand 5: We showed earlier that a (R)-methyl
substituent at the carbinol carbon on 2-azanorbornyl meth-
anol causes a significant increase in the activity of the
catalyst.[5b,c] If the positive effect of this modification was
transferable to a ligand with a remote dimethyl dioxolane
structure, we would obtain an extremely efficient catalyst.
The synthesis of ligand 5 (Scheme 2) starts with the
reduction of the ester 4a with LiAlH4 followed by Swern
oxidation to afford aldehyde 6 in 84% yield. Grignard
Scheme 1. Synthesis of ligands 2a ± d: i) 1 mol% OsO4, NMO, tBuOH/
H2O, RT, 12 h; ii) p-TsOH, 4a: (CH3O)2C(CH3)2, MeOH, 508C, 12 h, 89%
(steps i and ii), 4b: (C2H5)2CO, MeOH, 508C, 20 h, 68% (steps i and ii), 4c:
PhCOCH3, Dean ± Stark trap, benzene, reflux, 12 h, 70% (steps i and ii),
4d: (CH3O)2CH2, P2O5, CH2Cl2, 508C, 12 h, 51% (steps i and ii); iii) 30
wt% [Pd(OH)2], H2 (1 atm), MeOH, 508C; iv) LiAlH4, THF, 08C then RT,
67 ± 90% (steps iii and iv).
with a highly exo- and diastereoselective aza-Diels ± Alder
reaction between cyclopentadiene and the imine formed from
methyl glyoxylate and (S)-1-phenylethylamine, affording the
Diels ± Alder adduct 3.[7] This intermediate, which contains a
double bond, was dihydroxylated by using OsO4 and NMO in
a tBuOH/H2O mixture.[8] Protection of the diol with a ketone,
or a corresponding dimethyl ketal, in the presence of p-TsOH
afforded dioxolanes 4a ± d. Debenzylation with [Pd(OH)2]
under atmospheric hydrogen pressure and subsequent reduc-
tion of the ester with LiAlH4 afforded the corresponding b-
amino alcohols, 2a ± d, in 46 ± 80% overall yield from 3.
Scheme 2. Synthesis of ligand 5: i) LiAlH4, THF, 08C, 1 h; ii) DMSO,
oxalyl chloride, TEA, CH2Cl2, 788C, 84% (steps i and ii); iii) MeMgBr,
CeCl3, THF, 788C, 89%; iv) DMSO, oxalyl chloride, TEA, CH2Cl2,
788C, 86%; v) LiAlH4, THF, 08C, 1 h, and separation of the diaste-
reomers; vi) 30 wt% [Pd(OH)2], H2 (1 atm), MeOH, 508C, 12 h, 38%
(steps v and vi).
Transfer hydrogenation with ligands 1a and 2a ± d: The
reduction of acetophenone with [Ru(p-cymene)(2a)] as
catalyst showed a threefold increase in reaction rate com-
bined with an increase in enantioselectivity relative to [Ru(p-
cymene)(1a)] (Table 1, entries 1 and 2). To determine how
different substituents on the ketal affect the activity of the
reaction with MeMgBr and CeCl3 affords the two diastereo-
meric alcohols in 95:5 mixture with the desired isomer 5 as the
minor product. Therefore, it was necessary to oxidize the
diastereomeric mixture to the corresponding ketone 7 and
then reduce it back to a 1:1
Table 1. Asymmetric transfer hydrogenation of acetophenone.[a]
mixture of diastereomers, by
using LiAlH4 as the reducing
agent. This route allows recy-
cling of the undesired diaste-
reomer. Debenzylation of the
Product
TOF [h
alcohols affords ligand 5 in
38% yield.
1
[c]
Entry
1
Chiral ligand (L*)
t [h]
Conv. [%][b]
90
]
ee [%][d]
1a
3
1050
94
Transfer hydrogenation with li-
gand 5: When reducing aceto-
phenone with 5 at an S/C ratio
of 200 (Ru/ligand/base/sub-
strate 1:4:5:200), the reaction
was completed within six mi-
nutes (Table 2, entry 1) with
96% ee. This result encouraged
us to decrease the catalyst load-
ing and it was found that at an
2
3
4
5
2a: R1 ˆ R2 ˆ CH3
1
1
1
1
92
72
90
73
3000
1900
2800
1500
96
95
96
96
2b: R1 ˆ R2 ˆ C2H5
2c: R1 ˆ CH3, R2 ˆ Ph
2d: R1 ˆ R2 ˆ H
6
5
0.25
97
8500
96
[a] S/C ˆ1000. [b] Determined by 1H NMR spectroscopy. [c] Turnover frequencies [(mol product/mol catalyst)/h]
were calculated at 50% conversion. [d] Determined by chiral HPLC analysis.
1432
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0707-1432 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 7

Transfer Hydrogenation of Ketones
1431±1436
Table 2. Catalyst-loading study of asymmetric transfer hydrogenation of
acetophenone with [Ru(p-cymene)(5)].[a]
Table 3. Asymmetric transfer hydrogenation of aromatic ketones cata-
lyzed by [Ru(p-cymene)(5)].[a]
Product
Conv. [%][b]
Product
Entry
S/C
200
1000
3000
4000
5000
7000
t [min]
ee [%][c]
Entry Ketone
S/C t [min] Conv. [%][b] ee [%][c]
1
2
3
4
5
6
6
15
45
70
90
96
97
96
95
96
85
96
96
96
96
96
96
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
isobutyrophenone
a,a,a-trimethyl acetophenone 200 30
200 30
93
83
100
98
100
90
100
98
100
100
98
90
100
98
90
85
94
95
96
96
98
99
91
90
91
92
> 99
> 99
93
91
89
2-methyl acetophenone
2-bromo acetophenone
3-methyl acetophenone
3-methyl acetophenone
3-methoxy acetophenone
3-amino acetophenone
3-nitro acetophenone
3-nitro acetophenone
4-bromo acetophenone
4-chloro acetophenone
1-acetonaphthone
1-acetonaphthone
4-methyl acetophenone
2-acetyl pyridine
3-acetyl pyridine
4-acetyl pyridine
1000 15
1000 10
200
4
110
1000 15
200
200
200
4
4
4
[a] See Experimental Section for the transfer hydrogenation procedure.
[b] Determined by 1H NMR spectroscopy. [c] Determined by chiral GC
analysis.
1000 15
200
1000 15
200
1000 15
200
200 15
3
4
S/C ratio of 1000 the reaction was finished in less than
15 minutes, with a TOF50 as high as 8500 h 1 (Table 1, entry 6).
Even at an S/C ratio of 5000 the reaction proceeded to full
conversion after 90 minutes, but at an S/C ratio of 7000 the
reaction stopped at 85% conversion. The enantioselectivity
was unaffected by lowering the amount of catalyst, and no de-
crease in ee was detected as a result of prolonged reaction times.
Increasing the substrate concentration to 0.4m solution,
however, resulted in 3% lower ee (93%) and only 85% con-
version after one hour reaction time at an S/C ratio of 1000.
The reason for the apparent deactivation of the catalyst is
not fully understood, but formation of styrene as a side
reaction has been detected when reducing acetophenone, an
indication of the possible formation of water. To remove any
water formed, different drying agents such as molecular sieves
and MgSO4, were added to the reaction. However, the
presence of drying agents caused the reaction to stop at lower
conversion, probably due to interactions between the catalyst
and the drying agent. To find out whether styrene itself was
deactivating the catalyst, styrene was initially added to the
reaction, which continued unaffected. One possible explan-
ation for the low yields is dissociation of the [Ru(ligand)-
(arene)] complex. This could be excluded, since stirring the
catalyst in isopropyl alcohol for two hours before adding the
substrate gave the same conversion as when the reaction is
carried out according to the normal procedure. The possibility
of product inhibition was investigated by adding (R)- and (S)-
1-phenylethanol to two separate reaction mixtures prior to the
catalyst. The enantiomerically pure alcohols did not affect the
reaction. Since the reaction is known to be both moisture and
air sensitive it is possible that water is the cause of catalyst
deactivation, but deactivation by other by-products formed in
the catalytic cycle can not be excluded.
6
92
90
98
97
200
200
4
3
91
[a] See Experimental Section for the transfer hydrogenation procedure.
[b] Determined by 1H NMR spectroscopy. [c] Determined by chiral GC
analysis.
intermediates by fixed pre-oriented dipoles. In ligand 2a we
replaced two remote C H bonds in 1a with oppositely
polarized C O bonds, by introducing a dioxolane structure in
the ligand. This functionalization does not change the
geometry of the Ru ± hydride complex[9] and there is no
van der Waals contact between the dioxolane and the sub-
strate in the transition state (TS). Nevertheless, catalyst 2a
proved to be about three times faster than the parent
compound 1a. Possible explanations for the rate enhance-
ment include through-bond electronic effects, changes in
long-range electrostatics, or solvent effects. To distinguish
between these factors, we did a series of calculations.
The size of the system forced us to devise a scheme where
we started from B3PW91/LANL2DZ structures of the non-
functionalized system (1a) and optimized the dioxolane
structure using PM3(tm) with frozen co-ordinates of other
atoms (part C in Figure 2).[10] This generated a set of reactant
The substrate study (Table 3) shows that it is possible to
perform transfer hydrogenation on a range of different
aromatic ketones with a catalyst loading as low as S/C ˆ
1000 (Table 3, entries 1, 2, 4, 8, 10, and 12). Lowering the
amount of catalyst does not affect the enantioselectivity and
the reaction rates are still high. This system is capable of
reducing aromatic ketones that contain both electron-donat-
ing and -withdrawing substituents in ortho, meta, and para
positions with excellent enantioselectivity.
Density functional calculations: Enzyme catalysis partly relies
on the stabilization of transition states and high-energy
Figure 2. Computational scheme used in the evaluation of the dioxolane
effects.
Chem. Eur. J. 2001, 7, No. 7
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0707-1433 $ 17.50+.50/0
1433

P. G. Andersson et al.
FULL PAPER
iPrOH (48 mL), and the base (25 mL, 1m iPrOK in iPrOH), was added at
and TS structures suitable for an analysis of the cause of the
increased rate. By using these structures, single-point calcu-
lations at B3PW91[11] gave an almost perfect fit to the
expected reaction barrier difference between the two catalysts
RT, followed by the pre-catalyst. The resulting solution was stirred at RT,
1
and the reaction was monitored by H NMR.
Compound 3: Compound
procedure.[7]
3 was prepared according to a literature
1
(0.8 kcalmol vs. the experimental estimate of 0.7 kcal
(1R,2R,6S,7R,9R)-4,4-Dimethyl-8-[(S)-1-phenylethyl]-3,5-dioxa-8-azatri-
cyclo[5.2.1.02,6]decane-9-carboxylic acid methyl ester (4a): Compound 3
(6.6 g, 26 mmol) was dissolved in tBuOH (72 mL) and water (9.8 mL).
NMO (40 mL, 60% solution in water, 230 mmol) was added at RT,
followed by addition of OsO4 (0.2 g, 0.6 mmol). The reaction was stirred
overnight, diluted with H2O (50 mL), and quenched with Na2S2O5. The
organic solvent was removed under reduced pressure, and the residue was
washed with CH2Cl2 (3 Â 100 mL). Drying (MgSO4) and evaporation
afforded the diol in quantitative yield (7.4 g). The diol, was used without
further purification. 1H NMR (400 MHz, CDCl3): d ˆ 1.40 (d, J ˆ 6.4 Hz,
3H), 1.77 (d, J ˆ 10.8 Hz, 1H), 1.92 (d, J ˆ 10.8 Hz, 1H), 2.21 (brs, 1H) 2.44
(brs, 1H), 3.45 (s, 3H), 3.54 (q, J ˆ 6.4 Hz, 1H), 3.55 (brs, 1H), 3.79 (d, J ˆ
5.2 Hz, 1H), 4.25 (d, J ˆ 5.2 Hz, 1H), 7.12 ± 7.27 (m, 5H); 13C NMR
(100 MHz, CDCl3): d ˆ 22.2, 29.6, 48.8, 51.5, 60.1, 61.6, 65.6, 67.2, 73.2, 127.4,
127.9, 128.0, 143.8, 173.9.
1
mol ). This suggests that the effect of the dioxolane
substituent is described well by the computational method
employed. To differentiate between through-bond effects and
electrostatic effects, we calculated the interaction energy
between the truncated subsystems A and B (Figure 2). In this
case, system A was end-capped with hydrogens. Single-point
calculations at the B3LYP/6-311 ‡ G** level for the separated
systems and for the interacting systems of the TS geometry
indicated a repulsion of 0.4 kcalmol 1 in the TS of catalyst 1a
1
and an attraction of 0.9 kcalmol in the TS of catalyst 2a.
Adding these effects, a contribution to the net lowering of the
TS energy of catalyst 2a could be estimated at
1.3 kcalmol 1.[12] Therefore, it is most likely that the observed
rate enhancement is caused by the interaction of the new bond
dipoles, introduced at the rear end of the azanorbornyl
skeleton, with the substrate dipole.[13] By presenting this
analysis, we would like to encourage the use of van der Waals
attractions in the design of new catalysts.
The diol (7.4 g, 26 mmol) was dissolved in MeOH and p-toluenesulfonic
acid (5.4 g, 29 mmol) was added, followed by addition of 2,2-dimethoxy
propane (7.9 mL, 65 mmol). The reaction mixture was warmed to 40 ± 508C
and stirred overnight. The solvent was removed under reduced pressure,
and the residue was washed with aqueous NaOH (50 mL, 2m) and CH2Cl2
(3 Â 100 mL). Drying (MgSO4), evaporation, and purification by flash
chromatography (deactivated silica) afforded pure 4a in 89% yield (8.4 g).
Rf ˆ 0.25 (EtOAc/ pentane 1:9); m.p. 82 ± 838C; [a]3D0 ˆ ‡1.9 (c ˆ 1.0 in
1
CH2Cl2); IR (CH2Cl2): nÄ ˆ 2976, 1746, 1494, 1455, 1375 cm
;
1H NMR
(400 MHz, CDCl3): d ˆ 1.33 (s, 3H), 1.46 (d, J ˆ 6.4 Hz, 3H), 1.46 (s, 3H),
1.74 (d, J ˆ 10.4 Hz, 1H), 1.91 (d, J ˆ 10.4 Hz, 1H), 2.34 (s, 1H), 2.42 (s,
1H), 3.25 (s, 3H), 3.60 (q, J ˆ 6.4 Hz, 1H), 3.71 (s, 1H), 4.15 (d, J ˆ 5.6 Hz,
1H), 4.55 (d, J ˆ 5.6 Hz, 1H), 7.18 ± 7.23 (m, 5H); 13C NMR (100 MHz,
CDCl3): d ˆ 22.4, 24.3, 25.4, 29.3, 45.9, 51.4, 59.0, 60.1, 64.2, 75.6, 80.5, 109.6,
Conclusion
In conclusion, we have shown that the introduction of a
dioxolane ring in the amino alcohol ligand lowers the energy
in the transition state owing to van der Waals attractions
between the dipole in the dioxolane ring and the dipole in the
substrate. This results in a transfer hydrogenation catalyst that
can be used at low catalyst loading with high TOF50 and which
induces excellent enantioselectivity. To the best of our
knowledge, no other asymmetric catalyst for transfer hydro-
genation is capable of reducing acetophenone in a S/C ratio
exceeding 5000.
‡
127.4, 127.8, 128.0, 144.0, 173.6; MS (EI): m/z (%): 332 (59) [M] , 299 (22),
272 (25), 230 (63), 126 (47), 105 (100), 73 (32); elemental analysis calcd (%)
for C19H25NO4 ´ 0.4H2O: C 67.39, H 7.68, N 4.14; found C 67.28, H 7.33, N
4.34.
(1R,2R,6S,7R,9R)-4,4-Dimethyl-3,5-dioxa-8-azatricyclo[5.2.1.02,6]dec-9-yl-
methanol (2a): Compound 4a (8.4 g, 25 mmol) was dissolved in MeOH and
added to dried [Pd(OH)2] (2.5 g, 30 wt%). The flask was placed under
hydrogen atmosphere, and the reaction mixture was heated to 508C and
stirred overnight. The suspension was filtered through a pad of Celite, dried
(MgSO4), and evaporated to afford the ester in quantitative yield (5.8 g).
1H NMR (400 MHz, CDCl3): d ˆ 1.25 (d, J ˆ 10.8 Hz, 3H), 1.26 (s, 3H), 1.71
(d, J ˆ 10.8 Hz, 1H), 2.04 (brs, 1H) 2.63 (brs, 1H), 3.05 (brs, 1H) 3.41 (brs,
1H), 3.72 (s, 3H), 4.07 (d, J ˆ 5.6 Hz, 1H), 4.07 (d, J ˆ 5.6 Hz, 1H);
13C NMR (100 MHz, CDCl3): d ˆ 24.2, 25.5, 28.8, 44.4, 52.5, 57.24, 57.47,
80.6, 81.6, 110.1, 174.2.
Experimental Section
The ester (5.8 g, 25 mmol) dissolved in dry THF (50 mL) was added drop-
wise to a suspension of LiAlH4 (6.6 g, 0.18 mol) in THF at 08C. The
reaction mixture was allowed to reach RT, and after 25 min the reaction
was quenched by slow addition of H2O (6.6 mL), aqueous NaOH (6.6 mL,
1m), and H2O (20 mL). The reaction mixture was filtered, dried (MgSO4),
and the solvent was removed under reduced pressure. Precipitation of the
For general experimental information see reference [14]. iPrOH was dried
over CaH2 and freshly distilled under nitrogen prior to use. Acetophenone
was distilled and stored over activated molecular sieves. iPrOK (1m) was
prepared prior to use from freshly distilled iPrOH and potassium. Flash
chromatography was performed on silica gel (Matrex 60A, 37 ± 70 mm).
Deactivated silica gel means that it was treated with 5% Et3N in pentane,
and the column was eluted with the same solvent mixture until the eluent
was basic when tested with pH indicator paper. HPLC analysis was carried
out with a chiral column (ChiralCel OD-H) and a diode-array detector with
oil in EtOAc/pentane afforded 2a in 90% yield (4.7 g). M.p. 98 ± 1018C;
1
[a]2D4
ˆ
2.9 (c ˆ 0.5 in CH2Cl2); IR: nÄ ˆ 3357, 2974, 1458, 1168 cm
;
1H NMR (400 MHz, CDCl3): d ˆ 1.31 (s, 3H), 1.44 (d, J ˆ 10.4 Hz, 1H) 1.46
(s, 3H) 1.70 (d, J ˆ 10.4 Hz, 1H), 2.28 (s, 1H), 2.61 (dd, J ˆ 8.0, 5.6 Hz, 1H ),
3.28 (dd, J ˆ 10.4, 8.0 Hz, 1H), 3.35 (brs, 1H), 3.50 (dd, J ˆ 10.4, 6.0 Hz,
1H), 4.00 (d, J ˆ 5.6 Hz, 1H), 4.15 (d, J ˆ 5.6 Hz, 1H); 13C NMR (100 MHz,
CDCl3): d ˆ 24.2, 25.5, 27.9, 42.0, 55.6, 57.0, 65.1, 80.8, 82.5, 109.6; MS (EI):
1
a flow rate of 0.5 mLmin with 5% iPrOH in hexane as solvent. GC
analysis was performed on a Varian 3400 capillary gas chromatograph with
a CP-Chirasil-Dex CB column (25 m with 0.25 mm inner diameter),
nitrogen as carrier gas, and a flame ionization detector. MS-analysis was
carried out on a Finnigan MAT GCQ PLUS system. Infrared spectra were
recorded on a Perkin ± Elmer 1760 FT-IR spectrometer.
‡
m/z (%): 200 (71) [M] , 184 (30), 124 (17), 110 (17), 98 (30), 82 (36), 68
(100); elemental analysis calcd (%) for C10H17NO3: C 60.28, H 8.60, N 7.03;
found C 60.06, H 8.59, N 7.00.
General procedure for transfer hydrogenation of aromatic ketones: Amino
alcohol ligand (20 mmol) and [{RuCl2(p-cymene)}2] (1.53 mg, 2.5 mmol)
were added to a round-bottomed flask. Moisture was azeotropically
removed by evaporation with benzene (3 Â 4 mL) at reduced pressure.
Stirring the ligand and [{RuCl2(p-cymene)}2] at RT for 10 minutes in iPrOH
(2 mL) generated the pre-catalyst. The substrate (5 mmol) was dissolved in
(1R,2R,6S,7R,9R)-4,4-Diethyl-8-[(S)-1-phenylethyl]-3,5-dioxa-8-azatricy-
clo[5.2.1.02,6]decane-9-carboxylic acid methyl ester (4b): Compound 4b
was prepared by following the same procedure as described for 4a, except
for the protection of the diol. The diol (0.70 g, 2.4 mmol) was dissolved in
MeOH (50 mL), and p-toluenesulfonic acid (0.51 g, 2.6 mmol) was added
1434
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0707-1434 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 7

Transfer Hydrogenation of Ketones
1431±1436
followed by diethyl ketone (0.60 mL, 9.6 mmol). The reaction mixture was
heated under reflux overnight. The solvent was removed, and the residue
was washed with aqueous NaOH (5 mL, 2M) and CH2Cl2 (3 Â 20 mL).
Drying (MgSO4), evaporation, and purification by flash chromatography
(deactivated silica) afforded the pure product in 68% yield (0.59 g). Rf ˆ
0.4 (pentane/EtOAc 1:10); m.p. 82 ± 838C; [a]2D4 ˆ ‡0.4 (c ˆ 1.9 in CH2Cl2);
IR: nÄ ˆ 2971, 1743, 1492, 1458, 1379 cm 1; 1H NMR (400 MHz, CDCl3): d ˆ
0.90 (t, J ˆ 7.6 Hz, 3H), 0.95 (t, J ˆ 7.6 Hz, 3H) 1.46 (d, J ˆ 6.4 Hz, 3H), 1.59
(q, J ˆ 7.6 Hz, 2H), 1.70 (q, J ˆ 7.6 Hz, 2H), 1.81 (d, J ˆ 10.4 Hz, 1H), 1.88
(d, J ˆ 10.4 Hz, 1H), 2.37 (s, 1H), 2.41 (s, 1H), 3.25 (s, 3H), 3.58 (q, J ˆ
6.4 Hz, 1H), 3.74 (s, 1H), 4.12 (d, J ˆ 5.6 Hz, 1H), 4.51 (d, J ˆ 5.6 Hz, 1H),
7.18 ± 7.26 (m, 5H); 13C NMR (100 MHz, CDCl3): d ˆ 8.2, 8.2, 22.4, 27.3,
27.9, 29.4, 46.0, 51.5, 59.0, 60.1, 64.3, 75.5, 80.4, 113.7, 127.4, 127.8, 128.0,
4.15 (d, J ˆ 5.2 Hz, 1H), 4.31 (d, J ˆ 5.2 Hz, 1H), 7.25 ± 7.60 (m, 5H);
13C NMR (100 MHz, CDCl3): d ˆ 25.4, 28.0, 41.8, 56.0, 56.9, 64.9, 80.9, 82.1,
‡
110.0, 124.7, 125.2, 127.9, 128.1, 141.5; MS (EI): m/z (%): 262 (14) [M] , 220
(48), 205 (27), 152 (14), 1124 (21), 105 (100), 80 (55); elemental analysis
calcd (%) for C15H19NO3: C 68.94, H 7.33, N 5.36; found C 69.06, H 39, N
5.28.
(1R,2R,6S,7R,9R)-8-[(S)-1-Phenylethyl]-3,5-dioxa-8-azatricyclo-
[5.2.1.02,6]decane-9-carboxylic acid methyl ester (4d): The diol (5.0 g,
17 mmol) was dissolved in a solution of CH2Cl2 (80 mL) and dimethoxy-
methane (80 mL), followed by addition of P2O5 (40 g). The resulting
suspension was stirred at RT overnight and partitioned in a mixture of
saturated aqueous NaHCO3 (100 mL) and CH2Cl2 (100 mL) at 08C. The
organic phase was separated and washed with saturated aqueous NaHCO3
and brine. Drying (MgSO4), evaporatiion, and purification by flash
chromatography (deactivated silica, EtOAc/pentane; 1/10) afforded pure
4d in 51% yield (2.67 g). Rf 0.46 (EtOAc/pentane 1:4), m.p. 127 ± 1288C;
[a]2D5 ˆ ‡11.1 (c ˆ 1.0 in CH2Cl2); IR (KBr): nÄ ˆ 2970, 2866, 1742, 1438,
‡
144.0, 173.6; MS (EI): m/z (%): 358 (20) [M] , 329 (32), 300 (18), 229 (100),
225 (43), 126 (60), 105 (79), 79 (17); elemental analysis calcd (%) for
C21H29NO4: C 70.16, H 8.13, N 3.90; found C 70.00, H 8.21, N 3.97.
(1R,2R,6S,7R,9R)-4,4-Diethyl-3,5-dioxa-8-azatricyclo[5.2.1.02,6]dec-9-yl-
methanol (2b): Debenzylation of 4b by the procedure given for 2a
afforded the corresponding product in quantitative yield. 1H NMR
(400 MHz, CDCl3): d ˆ 0.86 (t, J ˆ 8 Hz, 3H), 0.93 (t, J ˆ 8.0 Hz, 3H),
1.27 (d, J ˆ 10.8 Hz, 1H), 1.55 (q, J ˆ 8.0 Hz, 2H) 1.66 (q, J ˆ 8.0 Hz, 2H),
1.81 (d, J ˆ 10.8 Hz, 1H), 2.69 (s, 1H), 3.08 (s, 1H), 3.48 (s, 1H), 3.74 (s,
3H), 4.09 (d, J ˆ 5.2 Hz, 1H), 4.17 (d, J ˆ 5.2 Hz, 1H); 13C NMR (100 MHz,
CDCl3): d ˆ 8.0, 8.6, 27.3, 27.8, 28.9, 44.4, 52.4, 57.2, 57.4, 80.4, 81.4, 114.1 and
174.1. Reduction of the ester according to the procedure given for 2a
1
1369 cm 1; H NMR (400 MHz, CDCl3): d ˆ 1.47 (d, J ˆ 6.4 Hz, 3H), 1.72
(d, J ˆ 10.8 Hz, 1H), 1.93 (d, J ˆ 10.8 Hz, 1H), 2.42 (s, 1H), 2.44 (s, 1H),
3.58 (q, J ˆ 6.4 Hz, 1H), 3.79 (s, 1H), 4.06 (d, J ˆ 5.6 Hz, 1H), 4.44 (d, J ˆ
5.6 Hz, 1H), 4.62 (s, 1H), 5.07 (s, 1H), 7.17 ± 7.28 (m, 5H); 13C NMR
(100 MHz, CDCl3): d ˆ 22.7, 29.4, 46.0, 51.8, 59.2, 60.5, 64.4, 76.3, 80.8, 95.3,
‡
127.7, 128.1, 128.3, 144.2, 173.9; MS (EI): m/z (%): 303 (4) [M] , 272 (38),
171 (19), 172 (40), 105 (100); elemental analysis calcd (%) for C17H21NO4:
C 67.31, H 6.98, N 4.62; found C 67.35, H 7.10, N 4.72.
afforded ligand 2b in 68% yield. M.p. 53 ± 548C; [a]D24
ˆ
23.5 (c ˆ 0.4 in
(1R,2R,6S,7R,9R)-3,5-Dioxa-8-azatricyclo[5.2.1.02,6]dec-9-ylmethanol
(2d): Debenzylation of 4d according to the procedure given for 2a afforded
the corresponding product in quantitative yield. 1H NMR (400 MHz,
CDCl3): d ˆ 1.29 (d, J ˆ 10.8 Hz, 1H), 1.71 (d, J ˆ 10.8 Hz, 1H), 2.28 (brs,
1H), 2.75 (s, 1H), 3.09 (s, 1H), 3.53 (s, 1H), 3.75 (s, 1H), 4.02 (d, J ˆ 5.6 Hz,
1H), 4.10 (d, J ˆ 5.6 Hz, 1H), 4.60 (s, 1H), 5.04 (s, 1H); 13C NMR
(100 MHz, CDCl3): d ˆ 29.0, 44.5, 52.9, 57.4, 57.7, 80.9, 81.9, 95.8, 174.2.
1
CH2Cl2); IR: nÄ ˆ 3360, 2975, 1458, and 1169 cm
;
1H NMR (400 MHz,
CDCl3): d ˆ 0.80 (t, J ˆ 7.6 Hz, 3H), 0.89 (t, J ˆ 7.6 Hz, 3H), 1.32 (d, J ˆ
10.4 Hz, 1H), 1.50 (q, J ˆ 7.6 Hz, 2H), 1.60 (q, J ˆ 7.6 Hz, 2H), 1.68 (d, J ˆ
10.4 Hz, 1H), 2.24 (s, 1H), 2.49 (dd, J ˆ 8.0, 5.4 Hz, 1H), 3.22 (dd, J ˆ 10.2,
8.0 Hz, 1H), 3.27 (brs, 1H), 3.39 (dd, J ˆ 10.2, 5.4 Hz, 1H), 3.89 (d, J ˆ
5.2 Hz, 1H), 4.02 (d, J ˆ 5.2 Hz, 1H); 13C NMR (100 MHz, CDCl3): d ˆ 8.0,
8.6, 27.3, 27.9, 28.0, 41.8, 56.1, 60.0, 64.9, 80.6, 82.0, 113.6; MS (EI): m/z (%):
‡
Reduction of the ester according to the procedure given for 2a afforded
228 (48) [M] , 198 (67), 124 (63), 110 (20), 106 (24), 97 (38), 80 (92), 68
ligand 2d in 75% yield. M.p. ˆ 110 ± 1118C; [a]D25
ˆ
31.8 (c ˆ 1.0 in
(100); elemental analysis calcd (%) for C12H21NO3 ´ 0.5H2O: C 60.99, H
9.38, N 5.93; found C 60.76, H 9.39, N 5.93.
1
CH2Cl2); IR (KBr): nÄ ˆ 3299, 2929, 1408, 1161 cm
;
1H NMR (400 MHz,
CDCl3): d ˆ 1.44 (d, J ˆ 10.8 Hz, 1H), 1.64 (d, J ˆ 10.8 Hz, 1H), 1.85 (brs,
2H), 2.37 (s, 1H), 2.59 (dd, J ˆ 6.8, 5.6 Hz, 1H), 3.28 (dd, J ˆ 10.4, 8.0 Hz,
1H), 3.41 (s, 1H), 3.49 (dd, J ˆ 10.4, 6.0 Hz, 1H), 3.88 (d, J ˆ 5.6 Hz, 1H),
4.01 (d, J ˆ 5.6 Hz, 1H), 4.59 (s, 1H), 5.03 (s, 1H); 13C NMR (100 MHz,
CDCl3): d ˆ 28.0, 42.1, 55.7, 57.0, 65.3, 81.1, 82.7, 95.3; MS (EI): m/z (%): 172
(1R,2R,4S,6S,7R,9R)-4-Methyl-4-phenyl-8-[(S)-1-phenylethyl]-3,5-dioxa-
8-azatricyclo[5.2.1.02,6]decane-9-carboxylic acid methyl ester (4c): Com-
pound 4c was prepared by following the same procedure as described for
4a, except for the protection of the diol. The diol (0.70 g, 2.4 mmol) was
dissolved in benzene (50 mL), and p-toluenesulfonic acid (0.51 g,
2.6 mmol) was added followed by acetophenone (0.80 mL, 6.8 mmol).
The reaction was heated under reflux overnight with azeotropic removal of
water by using a Dean ± Stark apparatus. The solvent was removed under
reduced pressure, and the residue was washed with aqueous NaOH (5 mL,
2m) and CH2Cl2 (3 Â 20 mL). Drying (MgSO4), evaporation, and purifica-
tion by flash chromatography (deactivated silica) afforded the pure product
‡
(29) [M] , 140 (14), 110 (15), 105 (85), 98 (12), 68 (100); elemental analysis
calcd (%) for C8H13NO3: C 56.13, H 7.65, N 8.18; found C 56.03, H 7.77, N
8.15.
(1R,2R,6S,7R,9R)-4,4-Dimethyl-8-[(S)-2-phenylethyl]-3,5-dioxa-8-azatri-
cyclo[5.2.1.02,6]decane-9-carbaldehyde (6): Compound 4a (11 g, 35 mmol)
was dissolved in dry THF (80 mL) and added drop-wise to a suspension of
LiAlH4 (1.3 g, 35 mmol) in THF (20 mL) at 08C. The reaction mixture was
stirred for 1 h at 08C and then quenched by slow addition of H2O (1.3 mL),
aqueous NaOH (1.3 mL, 1m), and H2O (3.9 mL). The reaction mixture was
filtered and dried (MgSO4), and the solvent was removed under reduced
pressure to afford the product in 94% crude yield (10 g). The product was
oxidized without further purification. 1H NMR (400 MHz, CDCl3): d ˆ 1.35
(s, 3H), 1.45 (d, J ˆ 6.4 Hz, 3H), 1.46 (s, 3H), 1.64 (m, 2H) 1.93 (dd, J ˆ 6.4,
2.4 Hz, 1H), 2.27 (s, 1H), 2.43 (dd, J ˆ 10.8, 6.4 Hz, 1H), 2.76 (d, J ˆ
10.8 Hz, 1H) 3.58 (q, J ˆ 6.4 Hz, 1H), 3.61 (s, 1H), 4.14 (d, J ˆ 5.4 Hz,
1H), 4.56 (d, J ˆ 5.4 Hz, 1H), 7.20 ± 7.35 (m, 5H); 13C NMR (100 MHz,
CDCl3): d ˆ 22.4, 24.3, 25.5, 29.1, 45.2, 59.7, 60.0, 63.6, 63.8, 75.7, 80.6, 109.3,
127.3, 127.7, 128.5, 145.6.
in 70% yield (0.77 g): Rf ˆ 0.38 (pentane/EtOAc 1:10); m.p. 1188C; [a]D24
ˆ
1
‡7.1 (c ˆ 1.4 in CH2Cl2); IR: nÄ ˆ 3425, 2972, 1746, 1639 cm
;
1H NMR
(400 MHz, CDCl3): d ˆ 1.49 (d, J ˆ 6.4 Hz, 3H), 1.60 (d, J ˆ 9.2 Hz, 1H),
1.64 (s, 3H), 1.84 (d, J ˆ 9.2 Hz, 1H), 2.44 (s, 1H), 2.48 (s, 1H), 3.25 (s, 3H),
3.65 (q, J ˆ 6.4 Hz, 1H), 3.78 (s, 1H) 4.34 (d, J ˆ 5.2, 1H), 4.74 (d, J ˆ
5.2 Hz, 1H) 7.20 ± 7.38 (m, 8H), 7.51 ± 7.54 (m, 2H); 13C NMR (100 MHz,
CDCl3): d ˆ 22.3, 25.5, 29.3, 46.0, 51.4, 59.0, 60.1, 64.1, 75.8, 80.6, 110.1,
124.7, 127.4, 127.8, 127.97, 128.0, 141.4, 143.8, 174.1; MS (EI): m/z (%): 393
‡
(1) [M] , 180 (16), 172 (18), 162 (17), 149 (68), 136 (17), 105 (34), 95 (55), 91
(54), 81 (67), 67 (66), 55 (100); elemental analysis calcd (%) for C24H27NO4:
C 73.25, H 6.92, N 3.56; found C 73.06, H 6.81, N 3.49.
(1R,2R,4S,6S,7R,9R)-4-Methyl-4-phenyl-3,5-dioxa-8-aza-tricyclo-
[5.2.1.02,6]dec-9-ylmethanol (2c): Debenzylation of 4c by the procedure
given for 2a afforded the corresponding product in quantitative yield.
1H NMR (400 MHz, CDCl3): d ˆ 1.20 (d, J ˆ 10.8 Hz, 1H), 1.57 (s, 3H), 1.60
(d, J ˆ 10.8 Hz, 1H), 2.13 (brs, 1H), 2.74 (s, 1H), 3.12 (s, 1H), 3.50 (s, 1H),
3.72 (s, 3H), 4.28 (d, J ˆ 5.2 Hz, 1H), 4.38 (d, J ˆ 5.2 Hz, 1H), 7.24 ± 7.36 (m,
8H), 7.49 ± 7.52 (m, 2H); 13C NMR (100 MHz, CDCl3): d ˆ 25.5, 28.9, 44.5,
52.5, 57.3, 57.4, 80.8, 81.8, 110.7, 124.7, 128.1, 141.6, 174.1.
DMSO (4.2 mL, 79 mmol) was added to a solution of oxalylchloride
(3.1 mL, 36 mmol) in dry CH2Cl2 (200 mL) for a period of 10 min at 788C.
The reaction mixture was stirred for 15 min, and the alcohol (10 g,
33 mmol), dissolved in dry CH2Cl2 (50 mL), was added over a period of
10 min. The reaction mixture was stirred for 15 min and triethylamine
(16 mL, 120 mmol) was subsequently added over a period of 10 min. The
reaction mixture was allowed to reach RT and then washed with brine
(100 mL) and CH2Cl2 (3 Â 200 mL). The combined organic layer was dried
(MgSO4) and evaporated, and the residue was purified by flash chroma-
tography (deactivated silica) affording 6 in 89% yield (8.9 g). Rf ˆ 0.2
(pentane/EtOAc 1:9); [a]3D0 ˆ ‡10.2 (c ˆ 2.3, CH2Cl2); IR (neat): nÄ ˆ 2980,
1723, 1382, 1207 cm 1; 1H NMR (400 MHz, CDCl3): d ˆ 1.32 (s, 3H), 1.43 (s,
Reduction of the ester according to the given procedure for 2a afforded
ligand 2c in 67% yield. [a]D25 ˆ ‡71.4 (c ˆ 1.4 in CH2Cl2); IR (KBr): nÄ ˆ
2927, 1138, 1043 cm 1; 1H NMR (400 MHz, CDCl3): d ˆ 1.35 (d, J ˆ 10.8 Hz,
1H), 1.56 (m, 1H), 1.58 (s, 3H), 2.34 (s, 1H) 2.62 (dd, J ˆ 8.0, 5.6 Hz, 1H),
3.26 (dd, J ˆ 10.8, 8.0 Hz, 1H), 3.39 (s, 1H), 3.47 (dd, J ˆ 10.8, 5.6 Hz, 1H)
Chem. Eur. J. 2001, 7, No. 7
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0707-1435 $ 17.50+.50/0
1435

P. G. Andersson et al.
FULL PAPER
3H), 1.45 (m, 1H) 1.47 (d, J ˆ 6.4 Hz, 3H), 1.75 (d, J ˆ 10.8 Hz, 1H), 2.26
(d, J ˆ 2.8 Hz, 1H), 2.44 (s, 1H), 3.58 (q, J ˆ 6.6 Hz, 1H), 3.72 (s, 1H), 4.16
(d, J ˆ 5.4 Hz, 1H) 4.56 (d, J ˆ 5.4 Hz, 1H), 7.12 ± 7.30 (m, 5H), 8.97 (s, 1H);
13C NMR (100 MHz, CDCl3): d ˆ 22.0, 24.1, 25.3, 29.6, 44.8, 59.2, 59.7, 69.6,
75.8, 80.2, 109.4, 127.5, 127.8, 128.4, 144.0, 203.2; MS (EI): m/z (%): 302 (7)
Acknowledgement
We would like to thank the Swedish Natural Science Research Council
(NFR), the Swedish Research Council for Engineering Science (TFR) and
the Swedish Foundation for Strategic Research (SSF) for generous
financial support. D. A. Alonso and T. Tarnai are also grateful to the
Wenner-Gren Foundation for grants.
‡
[M] , 272 (100), 172 (59), 105 (91), 68 (84); elemental analysis calcd (%) for
C18H25NO3 ´ 0.3H2O: C 70.47, H 7.75, N 4.57; found C 70.35, H 7.93, N 4.94.
1-{(1R,2R,6S,7R,9R)-4,4-Dimethyl-8-[(S)-1-phenylethyl]-3,5-dioxa-8-aza-
tricyclo[5.2.02,6]dec-9-yl}ethanone (7): MeMgI (29 mL, 3m in Et2O) was
added to a suspension of dry CeCl3 (22 g, 88 mmol) in Et2O at 788C.
After stirring for 1 h at 788C, compound 6 (8.9 g, 29 mmol) in Et2O
(50 mL) was added. The reaction mixture was allowed to reach RT
overnight. The solvent was removed under reduced pressure, and the
residue was dissolved in CH2Cl2 (400 mL) and washed with H2O (200 mL)
and CH2Cl2 (3 Â 400 mL). Drying (MgSO4) and evaporation of the solvent
afforded the alcohol (95:5 diastereomeric mixture) in 98% crude yield
(9.3 g). The product was used without further purification. The diastereo-
meric mixture of alcohols was transformed to the methyl ketone by Swern
oxidation according to the same procedure as described above for
preparation of compound 6. Purification by flash chromatography (deac-
tivated silica) afforded the methyl ketone 7, in 86% yield. Rf ˆ 0.18
[1] For reviews see a) ± c): a) G. Zassinovich, G. Mestroni, S. Gladiali,
Chem. Rev. 1992, 92, 1051 ± 1069; b) M. J. Palmer, M. Wills, Tetrahe-
dron: Asymmetry 1999, 10, 2045 ± 2061; c) R. Noyori, S. Hashiguchi,
Acc. Chem. Res. 1997, 30, 97 ± 102; d) Y. Nishibayashi, I. Takei, S.
Uemura, M. Hidai, Organometallics 1999, 18, 2291 ± 2293; e) K.
Murata, T. Ikariya, J. Org. Chem. 1999, 64, 2186 ± 2187; K. Murata, K.
Okano, M. Miyagi, H. Iwane, R. Noyori, T. Ikariya, Org. Lett. 1999, 1,
1119 ± 1121.
[2] a) Y. Sasson, J. Blum, Tetrahedron Lett. 1971, 24, 2167 ± 2170; b) Y.
Sasson, J. Blum, J. Org. Chem. 1975, 40, 1887 ± 1896; c) D. E. Linn, J.
Halpern, J. Am. Chem. Soc. 1987, 109, 2969 ± 2974; d) R. Noyori, H.
Takaya, Acc. Chem. Res. 1990, 23, 345 ± 350; e) R. L. Chowdhury, J. E.
Bäckvall, J. Chem. Soc. Chem. Commun. 1991, 1063 ± 1064.
[3] H.-U. Blaser, M. Studer, Chirality 1999, 11, 459 ± 464.
[4] a) T. Langer, G. Helmchen, Tetrahedron Lett. 1996, 37, 1381 ± 1384;
b) Y. Jiang, Q. Jiang, X. Zhang, J. Am. Chem. Soc. 1998, 120, 3817 ±
3818.
(pentane/EtOAc 1:9); m.p. 98 ± 998C; [a]2D5 ˆ ‡3.8 (c ˆ 0.8 in CH2Cl2); IR
1
(KBr): nÄ ˆ 2978, 1702, 1382, 1206 cm
;
1H NMR (400 MHz, CDCl3): d ˆ
1.35 (s, 3H), 1.45 (s, 3H), 1.47 (d, J ˆ 6.8 Hz, 3H), 1.53 (s, 3H), 1.68 ± 1.77
(m, 2H), 2.24 (s, 1H), 3.50 (s, 1H) 3.56 (q, J ˆ 6.6 Hz, 1H), 3.74 (s, 1H), 4.20
(d, J ˆ 5.6 Hz, 1H) 4.59 (d, J ˆ 5.6 Hz, 1H), 7.15 ± 7.28 (m, 5H); 13C NMR
(100 MHz, CDCl3): d ˆ 21.8, 24.3, 25.4, 27.3, 28.9, 45.3, 59.4, 60.0, 70.5, 75.6,
‡
80.9, 109.7, 127.7, 128.2, 128.3, 144.0, 209.3; MS (EI): m/z (%): 316 (9) [M] ,
[5] a) D. A. Alonso, D. Guijarro, P. Pinho, O. Temme, P. G. Andersson, J.
Org. Chem. 1998, 63, 2749 ± 2751; b) D. A. Alonso, P. Brandt, S. J. M.
Nordin, P. G. Andersson, J. Am. Chem. Soc. 1999, 121, 9580 ± 9588.
c) D. Alonso, S. J. M. Nordin, P. Roth, T. Tarnai, M. Thommen, U.
Pittelkow, P. G. Andersson, J. Org. Chem. 2000, 65, 3116 ± 3122.
[6] a) M. Yamakawa, H. Ito, R. Noyori, J. Am. Chem. Soc. 2000, 122,
1466 ± 1478; b) D. G. I. Petra, J. N. H. Reek, J.-W. Handgraaf, E. J.
Meijer, P. Dierkes, P. C. J. Kamer, J. Brussee, H. E. Schoemaker,
P. W. N. M. van Leeuwen, Chem. Eur. J. 2000, 2818 ± 2829.
[7] a) L. Stella, M. Abraham, J. Feneau-Dupont, B. Tinant, J. P. Declercq,
Tetrahedron Lett. 1990, 18, 2603 ± 2606; b) H. Waldmann, M. Braun,
Liebigs Ann. Chem. 1991, 1045 ± 1048; c) P. D. Bailey, G. R. Brown, F.
Korber, A. Reed, R. D. Wilson, Tetrahedron: Asymmetry 1991, 12,
1263 ± 1282.
272 (89), 172 (43), 105 (100), 68 (57); elemental analysis calcd (%) for
C19H25NO3: C 72.35, H 7.99, N 4.44; found C 72.24, H 8.06, N 4.40.
(R)-1-{(1R,2R,6S,7R,9R)-4,4-Dimethyl-3,5-dioxa-8-azatricyclo[5.2.1.02,6]-
dec-9-yl}ethanol (5): Reduction of 7 by LiAlH4 by the same procedure as
described for 6 afforded the alcohols in a 1:1 diastereomeric mixture. The
diastereomers were separated by flash chromatography (EtOAc/pentane
1:4, deactivated silica) to afford the (S)-methyl diastereomer (Rf ˆ 0.3) in
45% yield and the (R)-methyl diastereomer (Rf ˆ 0.2) in 46% yield.
(S)-Methyl diastereomer: 1H NMR (400 MHz, CDCl3): d ˆ 0.80 (d, J ˆ
6.6 Hz, 3H), 1.35 (s, 3H), 1.46 (s, 3H), 1.48 (d, J ˆ 6.6 Hz, 3H), 1.61 (d,
J ˆ 10.4 Hz, 1H), 1.69 (d, J ˆ 10.4 Hz, 1H) 1.72 (d, J ˆ 3.6 Hz, 1H), 2.29 (m,
1H), 2.41 (s, 1H), 3.61 (q, J ˆ 6.4 Hz, 1H), 3.66 (s, 1H) 4.08 (d, J ˆ 5.8 Hz,
1H), 4.58 (d, J ˆ 5.8 Hz, 1H), 7.20 ± 7.38 (m, 5H); 13C NMR (100 MHz,
CDCl3): d ˆ 18.0, 22.6, 24.4, 25.5, 29.6, 40.3, 59.8, 60.2, 65.3, 67.6, 76.0, 81.2,
109.3, 127.0, 127.8, 128.7, 145.6.
[8] P. Pinho, P. G. Andersson, Chem. Commun. 1999, 597 ± 598.
[9] Evaluated by B3PW91/LANL2DZ calculation.
[10] For computational details see reference [5b].
(R)-Methyl-diastereomer: 1H NMR (400 MHz, CDCl3): d ˆ 0.56 (d, J ˆ
6.6 Hz, 3H), 1.34 (s, 3H), 1.45 (s, 3H), 1.47 (d, J ˆ 6.6 Hz, 3H), 1.51 (d,
J ˆ 10.4 Hz, 1H), 1.67 (d, J ˆ 10.4 Hz, 1H,) 1.88 (d, J ˆ 4.4 Hz, 1H,), 2.23 (s,
1H), 2.83 (m, 1H), 3.58 (q, J ˆ 6.4 Hz, 1H), 3.61 (s, 1H) 4.15 (d, J ˆ 5.4 Hz,
1H), 4.58 (d, J ˆ 5.4 Hz, 1H), 7.20 ± 7.35 (m, 5H); 13C NMR (100 MHz,
CDCl3): d ˆ 19.9, 22.4, 24.3, 25.5, 29.0, 44.4, 59.9, 60.2, 67.4, 68.3, 76.0, 80.3,
109.3, 127.8, 128.0, 128.5, 145.3.
[11] SDD with an added set of f functions was used for Ru, whereas
6 ± 311 ‡ G was used for other atoms.
[12] MM3* in MacroModel 6.5 gives the difference in interaction energy as
0.8 kcalmol 1. BSSE in the B3PW91 calculation could cause an
overestimation of the interaction energy.
[13] Solvation free energies calculated in Gaussian 98 by using the PCM
method with parameters for ethanol together with B3PW91 and SDD
for ruthenium and 6 ± 31G* for other atoms indicate a possible rate
difference for ligand 2a. However, the small energy difference is
probably without the accuracy of this method.
Debenzylation of the (R)-methyl-diastereomer by the procedure given for
2a afforded 5 in 98% yield. The oil was precipitated in Et2O/pentane and
recrystallized in EtOH. [a]D25
ˆ
19.2 (c ˆ 0.8 in CH2Cl2); m.p. 538C; IR
1
(KBr): nÄ ˆ 3317, 2987, 2881, 1385, 1206, 1060 cm
;
1H NMR (400 MHz,
[14] S. K. Bertilsson, L. Tedenborg, D. A. Alonso, P. G. Andersson,
Organometallics 1999, 18, 1281.
CDCl3): d ˆ 1.12 (dd, J ˆ 6.0, 2.4 Hz, 3H), 1.28 (s, 3H), 1.41 (d, J ˆ 10.8 Hz,
1H), 1.42 (s, 3H), 1.66 (d, J ˆ 10.8 Hz, 1H), 2.13 (m, 1H), 2.24 (s, 1H), 3.26
(m, 1H) 3.32 (s, 1H), 3.97 (brs, 1H), 4.12 (d, J ˆ 5.6 Hz, 1H); 13C NMR
(100 MHz, CDCl3): d ˆ 20.1, 24.2, 25.5, 28.0, 42.2, 57.0, 61.3, 68.8, 80.7, 82.5,
‡
109.7; MS (EI): m/z (%): 214 (5) [M] , 198 (16), 138 (16), 110 (15), 82 (21),
68 (100); elemental analysis calcd (%) for C11H19NO3: C 61.94, H 8.98, N
6.57; found C 61.81, H 9.03, N 6.54.
Received: August 8, 2000 [F2658]
1436
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0707-1436 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 7