Fast racemisation of chiral amines and alcohols by using cationic half-sandwich ruthena- and iridacycle catalysts

Article abstract of DOI:10.1002/chem.200902103

The lipase-catalysed resolution of alcohols and amines yields only 50% of the desired enantiopure product. However, addition of a racemisation catalyst leads to 100% yield in what is called a dynamic kinetic resolution (DKR). There is a need for new racemisation catalysts that are fast and compatible with the conditions of the enzymatic reaction. We show that cationic half-sandwich ruthena- and iridacycle complexes are highly active and efficient in the racemisation of chiral alcohols and amines. Upon activation with base, these complexes are able to selectively racemise alcohols, whereas the non-activated complexes are selective catalysts for the racemisation of amines. We have applied the iridacycles in the DKR of racemic β-chloroalcohols to produce chiral epoxides in a biphasic system in good yields and high ee (ee = enantiomeric excess).

Full text of DOI:10.1002/chem.200902103

DOI: 10.1002/chem.200902103
Fast Racemisation of Chiral Amines and Alcohols by Using Cationic Half-
Sandwich Ruthena- and Iridacycle Catalysts
[
a]
[a]
[a]
[a]
Thomas Jerphagnon, Arnaud J. A. Gayet, Florian Berthiol, Vincent Ritleng,
[
a]
[a]
[b]
[a]
Nata sˇ a Mr sˇ i c´ , Auke Meetsma, Michel Pfeffer, Adriaan J. Minnaard,
[a]
[a, c]
Ben L. Feringa,* and Johannes G. de Vries*
Abstract: The lipase-catalysed resolu-
tion of alcohols and amines yields only
enzymatic reaction. We show that cat-
ionic half-sandwich ruthena- and irida-
cycle complexes are highly active and
efficient in the racemisation of chiral
alcohols and amines. Upon activation
with base, these complexes are able to
selectively racemise alcohols, whereas
the non-activated complexes are selec-
tive catalysts for the racemisation of
amines. We have applied the iridacycles
in the DKR of racemic b-chloroalco-
hols to produce chiral epoxides in a bi-
phasic system in good yields and high
ee (ee=enantiomeric excess).
50% of the desired enantiopure prod-
uct. However, addition of a racemisa-
tion catalyst leads to 100% yield in
what is called a dynamic kinetic resolu-
tion (DKR). There is a need for new
racemisation catalysts that are fast and
compatible with the conditions of the
Keywords: alcohols · amines · iridi-
um · metallacycles · racemisation
Introduction
sired enantiomer is produced; the other 50% of the non-de-
sired enantiomer being mere waste. Designing catalysts that
are capable of rapidly racemising the non-desired enantio-
mer and that are compatible with these enzymes enables a
dynamic kinetic resolution (DKR) that can give a 100%
Enantiomerically pure amines and alcohols are important
building blocks for pharmaceutical and agrochemical prod-
[1]
ucts. Although several catalytic methods exist for their
[2]
[5]
preparation in an enantiopure form, one of the most fre-
quently used production methods involves the classical reso-
yield of the desired enantiomer (Scheme 1). Relatively few
[3]
lution by crystallisation of diastereomeric salts. Enzymatic
[4]
resolution is also frequently used. The drawback of these
resolution methods lies in the fact that only 50% of the de-
[
a] Dr. T. Jerphagnon, Dr. A. J. A. Gayet, Dr. F. Berthiol, Dr. V. Ritleng,
N. Mr sˇ i c´ , A. Meetsma, Prof. Dr. A. J. Minnaard,
Prof. Dr. B. L. Feringa, Prof. Dr. J. G. de Vries
Stratingh Institute for Chemistry, University of Groningen
Nijenborgh 4, 9747 AG Groningen (The Netherlands)
Fax : (+31)50-363-4296
Scheme 1. DKR of alcohols and amines.
E-mail: J.G.de.Vries@rug.nl
B.L.Feringa@rug.nl
catalytic systems are available that can achieve this. The
[6]
[7]
[
b] Dr. M. Pfeffer
groups of Williams and Bꢀckvall developed the DKR of
alcohols by using a combination of lipases and ruthenium
catalysts. The latter used the dimeric Ru catalyst developed
Laboratoire de Synthꢁses Mꢂtallo-Induites, Universitꢂ de Strasbourg
4
rue Blaise Pascal, 67000 Strasbourg (France)
[
c] Prof. Dr. J. G. de Vries
[8]
by Shvo. These catalysts function through a dehydrogena-
DSM Pharmaceutical Products - Innovative Synthesis & Catalysis
P.O. Box 18, 6160 MD Geleen (The Netherlands)
Fax : (+31)46-476-7604
tion–hydrogenation sequence. DSM Pharmaceutical Prod-
ucts has further developed the Bꢀckvall system to allow in-
[9]
E-mail: Hans-JG.Vries-de@dsm.com
dustrial production of alcohols.
12780
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 12780 – 12790

FULL PAPER
The racemisation of amines is much more challenging.
Raney cobalt, Raney nickel or alkali metal hydroxide have
been used at high temperatures for the racemisation of
Results and Discussion
We started by investigating the behaviour of half-sandwich
metallacycles in the racemisation of (S)-1-phenylethanol
[10]
amines. In addition, aldehydes or thiols have been used as
[11]
catalysts at lower temperatures. The first DKR of amines
was achieved by Reetz et al. who managed to acylate (rac)-
a-methylbenzylamine by using a combination of Pd/C and
Candida antarctica lipase B (CALB, Novozym-435) in re-
(S1) and (R)-2-chloro-1-phenylethanol (S2). Cycloruthenat-
6
ed complexes of the type [Ru
A
H
U
G
R
N
U
G
A
H
U
G
R
N
U
G
A
T
N
T
E
U
G
6
6
3
AHCTUNGTNERNU(G PF ), in which C-N is a chiral primary or secondary amine
6
attached to the metallated aromatic ring through an alkyl
group, have been shown to be very effective for the asym-
[12]
fluxing triethylamine, although rather long reaction times
were needed for complete conversion. Jacobs and de Vos
metric transfer hydrogenation of aromatic ketones.
used a basic support, such as BaSO , for the palladium cata-
therefore decided to prepare an analogous complex, but
with an achiral amine, and to test its activity for the racemi-
sation of both alcohols (Scheme 2). Complex 1 was obtained
analytically pure according to the transformation shown in
Scheme 2.
4
lyst, which increased the catalyst activity and allowed the
racemisation to proceed under milder conditions. Thus, the
[13]
time of the DKR reaction was reduced to three days. In
addition, these authors found that application of a low pres-
sure of dihydrogen suppressed
the formation of secondary
amines in the racemisation of
primary amines. Palladium
nanoparticles developed by
Park et al. led to excellent
yields and enantioselectivities
in the DKR of primary amines
Scheme 2. Synthesis of ruthenacycle complex 1.
with a reaction time of three
[14]
days.
These nanoparticles
could be recycled without any
loss of activity for 10 consecutive runs. The Bꢀckvall group
It is noteworthy that, due to the presence of stereogenic
centres on the Ru and N atoms, 1 consists of two pairs of
enantiomers R S /S R and R R /S S , which show
well-separated signals in their H- and C NMR spectra and
are found in a diastereomeric ratio of 56:44 (de=12%; de=
[15]
made several contributions to this area.
They used the
and showed that better results were ob-
tained when using an electron-rich variant thereof in the
[15a]
Shvo catalyst
Ru
N
Ru
N
Ru
N
13
Ru N
1
[15b,c]
DKR of alcohols and amines.
Recently they reported
the use of the transfer hydrogenation catalysts developed by
Baratta for the racemisation of amines. No DKR could be
diastereomeric excess) in CD CN (Scheme 3). Considering
3
[15d]
achieved, though.
They applied this methodology in the
synthesis of Norsetraline.
In general, the sensitivity of the metal catalyst for the
enzyme seems to play an important role in the reactions
since the dimeric iridium complex [Ir ACHTUNGTRNEUNG( Cp*)(I )] (Cp*=pen-
2 2
tymethylcyclopentadiene) was successfully used by Blacker
as a fast racemisation catalyst for the dynamic kinetic reso-
lution of 6,7-dimethoxy-1-methyl-tetrahydroisoquinoline in
combination with immobilized Candida antarctica lipase,
whereas palladium and ruthenium complexes showed much
[16]
lower reactivity.
The development of fast racemisation
catalysts for alcohols and amines continues to be a major
challenge. As we have developed the use of ruthenacycles as
asymmetric transfer-hydrogenation catalysts, we were inter-
ested in studying these catalysts, and, in particular, the anal-
Scheme 3. Stereoisomers of 1.
[17]
ogous iridacycles, as racemisation catalysts. The synthesis
of iridacycles based on the reaction between [Ir
ACHTUNGTNERNUN(G Cl ) ACHTUNGTRENNNU(G Cp*)]
2
2
[
18]
and benzylamines has been described by Davies
and
the sterically less-constrained structure of the R S /S R
N
Ru
N
Ru
[19]
Pfeffer et al. Ikariya recently reported the use of iridacy-
pair (the NMe group is further away from the CH CN
3
[20]
cles as alcohol oxidation catalysts.
ligand than in the other pair) it is most probably the major
[21]
one. These diastereomers rapidly interconvert in solution.
The catalytic activity of 1 was examined in the racemisa-
tion of (S)-S1 and (R)-S2 in the presence of KOtBu as an ac-
tivator (Scheme 4). Complex 1 gave very poor results in di-
Chem. Eur. J. 2009, 15, 12780 – 12790
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12781

B. L. Feringa, J. G. de Vries et al.
This indicated that catalyst deactivation in the racemisation
attempts of (R)-S2 was due to the formation of small
amounts of 2-chloroacetophenone, the dehydrogenation
product of S2, which reacted with the active species. A simi-
lar behaviour had already been observed in the transfer hy-
drogenation of functionalised b-keto-esters catalysed by (b-
Scheme 4. Racemisation of chiral alcohols by using ruthenacycle catalyst
[22]
amino-alcohol) ACHUTNGRNENUG( arene)ruthenium(II) complexes.
1
.
The active species 1’, generated by the action of KOtBu
1
on 1, has been identified by H NMR spectroscopy in
CD CN (see the Experimental Section) as a neutral cyclo-
metallated ruthenium hydride complex in which the NH
proton is NMR silent, presumably due to rapid exchange
3
Table 1. Racemisation of alcohol (S)-S1 with ruthenacycle 1.
[
a]
À1
Entry
Solvent
c [molL
]
Reaction time [h]
ee [%]
1
2
CH Cl
PhMe
PhMe
iPrOH
iPrOH
iPrOH
iPrOH
2
2
0.25
0.1
0.1
0.05
0.25
1.0
1.0
0.5
0.5
19
7
3
20
7
1
0.5
46
5
70
8
2
50
6
2
0
8
6
with CD CN under basic conditions (Scheme 6). Like 1,
3
[
b]
3
4
5
6
7
8
9
[
[
[
c]
d]
H
H
2
O
O
d,e]
2
[
(
[
1
a] Reaction conditions: 0.20 mmol of enantiomerically pure substrate
ee>99%), 4 mol% of 1, 5 mol% of KOtBu, RT. [b] Reaction at 608C.
c] Reaction at 808C. [d] Reaction carried out with an additional
.0 equiv of acetophenone. [e] Reaction at 408C.
Scheme 6. Formation of active species 1’.
1
complex 1’ displays two sets of signals in its H NMR spec-
chloromethane with only 30% of racemisation of S1 after
9 h at RT (Table 1, entry 1). Even though much better re-
trum, which indicates the presence of two diastereomers.
The diastereoisomers are found in an approximate ratio of
1:1, and the ruthenium hydride signals are detected at d=
À7.13 and À7.85 ppm. The chemical shifts corresponding to
1
sults were observed in toluene at RT (8% ee after 7 h,
entry 2; ee=enantiomeric excess), it required 3 h at 608C to
observe almost complete racemisation of S1 (entry 3). In
iso-propanol, the reaction rate was found to be strongly de-
pendent on the substrate concentration.
6
the h -benzene units and the aryl protons are strongly
shielded in comparison to those of the cationic species 1,
which is characteristic of a neutral cycloruthenated com-
[23]
While the reaction was very slow at a concentration of
plex. Presumably, deprotonation of 1 leads to the forma-
tion of the neutral complex, which reacts with iso-propanol
to form 1’.
0
.05m (Table 1, entry 4), it was almost complete (ee=2%)
after 1 h at a concentration of 1.0m with 4 mol% of 1 and
.2 mol% of KOtBu (entry 6). When run at 808C under oth-
erwise similar conditions, racemisation was complete in
0 min (entry 7). In water, although poorly soluble, 1 racem-
5
2-Chloroacetophenone reacts almost instantaneously with
1’ (or with 1 in the presence of KOtBu) to give the inhibited
species. Unfortunately, the latter could not be isolated due
3
1
ised (S)-S1 with a reasonable rate at 408C in the presence of
one equivalent of acetophenone.
Unfortunately, 1 showed no activity for the racemisation
of (R)-S2 in the presence of KOtBu in various solvents. A
control experiment showed that a reaction conducted at RT
with a 1.0m solution of (S)-S1 in iPrOH in the presence of
to decomposition during the workup. The H NMR spectro-
scopic analysis of a crude reaction mixture points at the for-
mation of a new cycloruthenated species that is present in
solution in only one diastereomeric form. However, the
exact structure remains unclear.
We next focused on the synthesis of electron-rich half-
sandwich iridacycle complexes, as these complexes were ex-
pected to have a higher stability toward air and moisture
compared to ruthenacycles. A first set of cationic iridacycles
4
mol% of 1 and 5.2 mol% of KOtBu was inhibited by the
addition of 5 mol% of 2-chloroacetophenone (Scheme 5).
2
–4 (Scheme 7) was synthesized from [Ir ACHTNUGERTNNGNU( Cp*) ACTHUNGTRENNUNG( Cl )] and
2 2
two equivalents of benzylamine, N-methylbenzylamine or
N,N-dimethylbenzylamine, respectively, in the presence of
sodium hydroxide and potassium hexafluorophosphate in
acetonitrile at 458C for 16 h (Scheme 5). After filtration of
the reaction mixture through alumina, the complexes 2–4
were obtained as yellowish powders in high yields (up to
9
2%). Contrary to ruthenacycle 1 (Scheme 3), the two pairs
of enantiomers R S /S R and R R /S S were not ob-
served by H NMR spectroscopy at RT for iridacycle 3,
Scheme 5. Inhibition of the racemisation of (S)-S1 by 2-chloroacetophe-
none.
Ir
N
Ir
N
Ir
N
Ir N
1
12782
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 12780 – 12790

Fast Racemisation of Chiral Amines and Alcohols
FULL PAPER
Iridacycle complexes 2–4
were studied in the catalytic
racemisation of (S)-phenyletha-
nol (S1) and (R)-2-chloro-1-
phenylethanol (S2). We soon
observed a key role for the N-
substituent on the ligand in
complexes 2–4. First investiga-
tions were done by using
Scheme 7. Synthesis of iridacycle complexes.
5
mol% of 4 at 708C in toluene.
As observed with the ruthena-
cycles, complex 4 was not able to racemise chiral alcohols
without activation with base (Table 2, entries 1–2,). Activa-
which might be due to the faster inversion of configuration
at the iridium centre.
This hypothesis was confirmed by studying the behaviour
1
of 3 by using variable-temperature H NMR spectroscopy in
Table 2. Racemisation of alcohols S1 and S2 with iridacycles 2–4.
CD CN (Figure 1a,b). The N-methyl and Cp* moieties give
3
very broad signals at 208C. At 08C, the two diastereoisomer-
ic N-methyl groups are clearly observed as are the two Cp*-
methyl moieties (Figure 1a and b, respectively). This effect
was more pronounced at À10 and À208C. The coalescence
temperature of the Cp*-methyl signals is 38C.
[
a]
Entry
[Ir]
Substrate
T [8C]
ee [%]
Ketone [%]
[
b]
1
2
3
4
5
6
7
8
9
1
4
4
4
4
4
4
3
3
2
2
S1
S2
S1
S2
S1
S2
S1
S2
S1
S2
70
70
70
100
97
34
29
83
8
3
0
100
3
9
–
15
–
10
–
13
–
[
b]
70
RT
RT
RT
RT
RT
RT
12
–
0
[
(
2
a] Reaction conditions: 0.2 mmol of enantiomerically pure substrate
ee>99%), 5 mol% of [Ir], 6 mol% KOtBu, 2 mL of PhMe at RT for
4 h. [b] Without KOtBu.
tion with 1.2 equivalents of potassium-tert-butoxide led to a
highly efficient catalyst for the racemisation of S1 since
3
4% ee was obtained in 24 h at 708C in toluene (entry 3).
Acetophenone (15%) was also observed by GC analysis.
Under the same reaction conditions, 29% ee was obtained
in the racemisation of (S)-S2 with complete selectivity since
no corresponding ketone was found (entry 4). Lowering the
temperature to RT had the expected effect on the rate of
racemisation of (S)-S1 (83% ee, entry 5). Surprisingly, the
rate of racemisation of (R)-S2 was higher at RT compared
to 708C. Indeed, after 24 h, S2 was obtained quantitatively
in 8% ee (entry 6). Complexes 2 and 3 were then tested at
RT in the racemisation of (S)-S1 and (R)-S2. Complex 2 was
inactive in the presence of (S)-S1, but was able to racemise
(
S)-S2 almost to completion (100 and 3% ee, respectively,
entries 9–10). Mono-N-methyl-substituted iridacycle 3 was
the most efficient catalyst since complete racemisation of
both (S)-S1 and (R)-S2 was observed after 24 h at RT (3
and 0% ee, respectively, entries 7–8).
Subsequently, we studied the kinetics of the racemisation
of chiral a-aryl and alkyl alcohols S1–S8 by using the opti-
mised reaction conditions (5 mol% of catalyst 3, 6 mol% of
1
Figure 1. a) Variable-temperature H NMR spectra of the N-methyl sig-
nals of 3 in CD
3
CN. 1: +20, 2: +10, 3: 0, 4: À10, 5: À208C. b) Variable-
temperature H NMR spectra of the Cp*-methyl signals of 3 in CD CN.
: +20, 2: +10, 3: 0, 4: À10, 5: À208C.
1
3
1
Chem. Eur. J. 2009, 15, 12780 – 12790
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12783

B. L. Feringa, J. G. de Vries et al.
KOtBu, PhMe, RT; Table 3). With S1, 50% and complete
racemisation was achieved in and 7 h, respectively
Table 3, entry 1). A tremendous increase of reaction rate
(S)-1-phenylbutanol (S4) possessing a longer alkyl chain
(entry 7). Racemic 1-(2-naphthyl)ethanol (S5) was obtained
from (S)-S5 in 200 min with a t1/2 of racemisation of 17 min
1
(
(
entry 8). The rate of racemisation decreased with a more
rigid substrate structure, such as (R)-1-indanol (S6), since
complete racemisation was obtained after 24 h (entry 9). Iri-
dacycles also showed activity in the catalytic racemisation of
aliphatic alcohols. However lower reactivity was observed in
the racemisation of (S)-2-butanol (S7) and (R)-2-hexanol
Table 3. Racemisation of alcohols with iridacycle 3 at room temperature.
(
S8) with a t1/2 of racemisation of 150 and 240 min and com-
plete racemisation in 16 and 24 h, respectively (entries 10–
1). In all cases, when using a-methyl-substituted alcohols,
1
the formation of the corresponding ketone was observed,
showing that the rate-determining step is the reduction of
the ketone to the corresponding racemic alcohol. This is
also corroborated by the faster racemisation of S2. The re-
duction rate is enhanced by the presence of electron-with-
drawing groups. Based on these results, we suggest that the
substrate dissociates from the metal centre after dehydro-
genation. This was also shown by adding one equivalent of
[
a]
Entry
Substrate
t
1/2(rac). [min]
t
compl.(rac) [min]
Ketone [%]
1
2
S1
S2
S2
S2
S2
S3
S4
S5
S6
S7
S8
60
–
20
–
–
20
420
5
180
6
–
–
–
–
10
12
8
20
7
2
-bromoacetophenone to (S)-S1 in the presence of 5 mol%
of 3 and 5.2 mol% of base in toluene (Scheme 8). Only ace-
tophenone and the corresponding racemic bromoalcohol
were detected quantitatively by GC analysis.
The tolerance of half-sandwich iridacycle 3 towards aque-
ous media and its high efficiency in the racemisation of (R)-
[
[
[
b]
c]
d]
3
4
5
6
7
8
9
1
1
1080
60
240
720
200
1440
960
1440
120
17
360
150
240
2
-chloro-1-phenylethanol (S2) at RT led us to combine this
[
e]
racemisation reaction with an enzymatic kinetic resolution
process. Janssen et al. found that haloalcohol dehalogenase
Hhec is very efficient in the resolution of b-chloroalcohols
0
1
12
[
(
1
a] Reaction conditions: 0.75 mmol of enantiomerically pure substrate
ee>99%), 5 mol% of 3, 6 mol% KOtBu, 2.4 mL of PhMe. [b] With
mol% of 3. [c] After 16 h, 0.2 mmol of (S)-S2 was added and the reac-
tion mixture was stirred for an additional 2 h, 6% ee after 18 h. [d] Re-
[25]
to afford enantiopure epoxides.
By using the double
mutant Hhec C153S W249F and catalytic amounts of the ac-
tivated iridacycle 3, enantioenriched epoxides were obtained
2
action achieved in a mixture of PhMe/H O 1:1. [e] 10% ee after 24 h.
was observed with S2 since
complete racemisation was ach-
ieved within 5 min at RT
(
entry 2). Similar high rates
have only recently been report-
ed by Bꢀckvall and co-workers
for this substrate. Decreasing
the catalyst loading to 1 mol%
Scheme 8. Hydrogen-transfer reaction by using activated 3.
[24]
allowed complete racemisation to be reached in 3 h with a
half-life time of racemisation of 20 min (entry 3). The cata-
lyst was still active after 16 h. Indeed, when a new batch of
by dynamic kinetic resolution of racemic substituted 2-
chloro-1-phenylethanol in a one-pot reaction in high yields
and with excellent enantioselectivities (up to 90 and 98% ee,
[26]
(
R)-S2 was added only 6% ee was observed 2 h after the ad-
respectively; Scheme 9).
Unfortunately, only enzymatic
dition (entry 4). Furthermore, complete racemisation of (R)-
S2 was obtained in 1 h in a mixture of toluene/water as the
solvent (entry 5), which showed the high tolerance of 3 to
aqueous media.
Compared to substrate (S)-S1, the presence of an elec-
tron-withdrawing group at the para position of (S)-1-(4-fluo-
rophenyl)ethanol (S3) strongly decreased the time of com-
plete racemisation to 4 h, with a time of half-racemisation of
20 min (Table 3, entry 6). Compared to S1, a longer reaction
time was needed to perform the complete racemisation of
Scheme 9. Chemoenzymatic DKR of b-chloroalcohols to chiral epoxides.
12784
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 12780 – 12790

Fast Racemisation of Chiral Amines and Alcohols
FULL PAPER
resolution and no racemisation was observed when using ali-
phatic b-chloroalcohols.
Thus, iridacycle complexes 1–3 have shown their efficien-
cy in the catalytic racemisation of chiral alcohols. In addi-
tion, they can be combined with aqueous enzymes for a dy-
namic kinetic resolution.
Amine racemisation: We also investigated the behaviour of
these complexes in the catalytic racemisation of chiral
amines, such as (S)-a-methylbenzylamine (S9) and (S)-N,a-
dimethylbenzylamine (S10), as model substrates in toluene
(
(
0.25m) at 1008C when using 2 mol% of iridacycle
Scheme 10). In contrast to the corresponding reaction with
alcohols, no racemisation was observed after activating 1–3
with potassium-tert-butoxide.
Figure 2. Enantiomeric excess versus time in the racemisation of S10
(
0.75 mmol, 0.25m) with 2 mol% of [Ir] at 1008C in toluene. ^: 2;
&
: 3;
~
: 4.
Scheme 10. Racemisation of a-methylbenzylamine derivatives.
Without activation, preliminary tests with (S)-a-methyl-
benzylamine (S9) showed that only slow racemisation of this
substrate was observed and mainly formation of side prod-
ucts was detected. The side products are formed by the con-
densation reaction of the imine intermediate with the sub-
[27]
strate to form dimers. This result shows that, in the case
of substrate S9, the rate-determining step in the catalytic
cycle is the reduction of the imine to the racemic amine.
Iridacycles 2–4 showed good reactivity with (S)-a-methyl-
N-methylbenzylamine (S10) since a decrease of enantiomer-
ic excess down to 10% ee was obtained in toluene at 1008C
when using 2 mol% of catalyst after 16 h. Moreover, no side
products were observed. Kinetic studies showed that cata-
Figure 3. Enantiomeric excess versus time in the racemisation of S10
0.75 mmol, 0.25m) with 2 mol% of [Ir] at 1008C in chlorobenzene. ^: 2;
(
&
: 3; ~: 4.
lysts 2 and 4 have the same activity after 3 h (t1/2(rac)
=
200 min), but catalyst 2 is more active than 4 after 8 h (15
Catalysts 2–4 were efficient in the racemisation of (S)-a-
and 26% ee, respectively; Figure 2). A different behaviour
was observed with iridacycle 3 synthesised from N-methyl-
benzylamine: the racemisation rate is higher than with cata-
lysts 2 and 4 (41% ee after 1 h). After the initial fast racemi-
sation, catalyst 3 seems to be deactivated after 2 h since
only a decrease of less than 13% ee was detected in the next
methyl-N-methylbenzylamine since complete racemisation
was observed within 8 h in chlorobenzene and within 16 h in
toluene at 1008C when using 2 mol% of catalyst. However,
the most active catalyst 3 possessing a secondary amine as
the ligand again seems to be deactivated under the reaction
conditions. The deactivation of catalyst 3 might be due to
the oxidation of the N-methylbenzylamine ligand to N-ben-
zylidenemethylamine in solution, resulting in less-active cat-
alyst 5 (Scheme 11). This was confirmed by studying the sta-
6
h (34% ee after 2 h and 21% ee after 8 h).
The use of more polar solvents, such as chlorobenzene, al-
lowed an increase of the reaction rate (Figure 3). This effect
might be due to the enhanced solubility of these complexes
in this solvent. Complete racemisation of (S)-a-methyl-N-
methylbenzylamine (S10) was obtained with catalysts 3 and
4
in 8 h, (t1/2(rac) =30 and 90 min, respectively). Although cat-
alyst 3 seems to be deactivated after 2 h, the effect is less
prominent than in toluene. No improvement was observed
with catalyst 2 since the time of half-racemisation was about
180 min as in toluene and complete racemisation was ob-
tained after 16 h.
3
Scheme 11. Oxidation of iridacycle 3 into 5 in CDCl .
Chem. Eur. J. 2009, 15, 12780 – 12790
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12785

B. L. Feringa, J. G. de Vries et al.
bility of 3 in CDCl . It was found that complex 3 slowly oxi-
3
dized to imine-complex 5. Moreover, iridacycle 5 synthe-
sized from [Ir ACHTUNGTRENNUNG( Cp*) ACHTUNGRTENNUNG( Cl )] and N-benzylidenemethylamine
2 2
showed lower reactivity relative to complex 3 since only
55% ee was obtained in 16 h in the racemisation of S10.
The presence of a secondary amine moiety in the ligand,
as in 3, seems to be important in terms of reactivity during
the racemisation reaction since 3 is initially the most active.
To counter the oxidation of the ligand, iridacycles 6 and 7
Figure 5. Kinetic studies of the racemisation of (S)-S10 (0.75 mmol) in
chlorobenzene (0.25m) at 1008C by using 2 mol% of 6 and 7. ~: 6; &: 7.
2
were synthesized with ligands containing an sp carbon
centre at the benzylic position and a secondary amine, such
as phenylimidazoline and phenylimidazole. To compare with
S10 without any loss of activity. It is noteworthy that the
presence of a NH group in the ligand of iridacycle com-
plexes 6 and 7 is necessary to increase the racemisation rate
(compared to 8). This effect was already observed with com-
plexes 3 and 5 (Scheme 11).
The effect of the temperature on the reaction rate of the
racemisation of S10 in toluene and chlorobenzene was inves-
tigated with catalyst 6 (Table 4). In toluene at 1008C, the
6
and 7, iridacycle 8 possessing a 2-phenyloxazoline ligand
was also synthesized (Figure 4). These complexes were ob-
tained analytically pure in excellent yields (up to 95%) and
show a high stability towards air and moisture.
Table 4. Temperature effect in the racemisation of (S)-S10 with iridacycle
6
.
[
a]
Entry
Solvent
T [8C]
t
1/2(rac) [min]
t
compl.(rac) [min]
[
b]
1
2
3
4
5
6
7
8
PhMe
PhMe
PhMe
PhMe
PhCl
PhCl
PhCl
PhCl
40
60
80
100
40
60
–
–
160
45
20
–
90
25
4
960
360
180
–
960
280
40
[
c]
80
100
Figure 4. ORTEP style plot of cationic iridacycle 8. Thermal ellipsoids
are drawn at the 50% probability level and hydrogen atoms are omitted
for clarity.
[a] Reaction conditions: 0.25 mmol of substrate, 2 mol% of [Ir], 1 mL of
PhMe. [b] 63% ee observed after 16 h. [c] 68% ee observed after 16 h.
Iridacycle 8 is able to racemize (S)-S10 but no enhance-
ment of reactivity was observed (26% ee after 16 h at 1008C
when using 2 mol% of catalyst in toluene). On the contrary,
both catalysts 6 and 7 showed very high activities in the rac-
emisation of S10 with 2 mol% of iridacycle at 1008C. Com-
plete racemisation was obtained in chlorobenzene within 40
and 180 min with 6 and 7, respectively, with a remarkably
short time of half-racemisation of 4 min and less than
racemisation is complete in 3 h with a time of half-racemisa-
tion of 20 min. At 808C the time of half-racemisation is
about 45 min with a complete racemisation in 6 h. Decreas-
ing the temperature to 608C reduces the rate of the reaction
(t1/2(rac) =160 min and complete racemisation in 16 h) but
only a low turnover was observed at 408C since a loss of en-
antiomeric excess of less than 10% was observed after 4 h.
In chlorobenzene at 1008C, the rate of the reaction in-
creases relative to toluene since complete racemisation was
obtained in 40 min (t1/2(rac) =4 min; Table 1). At 808C, the
rate of the reaction is slightly higher than in toluene
20 min, respectively (Figure 5). As expected, no deactivation
of the catalysts was observed. By using the same conditions,
the reaction was scaled-up to 2 g of enantiopure substrate
12786
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 12780 – 12790

Fast Racemisation of Chiral Amines and Alcohols
FULL PAPER
(
t1/2(rac) =25 min). It is expected that the reactivity of the cat-
1008C with 2 mol% of 6 for 16 h ee values down to 34%
were reached, but the selectivity of the reaction was still low
(36%).
To extend the scope of the reaction, we investigated the
catalytic racemisation of a series of chiral amines including
primary, secondary and tertiary amines by using 2 mol% of
6 as the catalyst in toluene (Table 5).
alytic system decreases with the temperature, but it is still
efficient at 608C (t1/2(rac) =90 min). As in toluene, only slow
racemisation is obtained at 408C (less than 12% in 4 h).
However, iridacycle 6 has been shown to be very efficient in
racemising (S)-a-methyl-N-methylbenzylamine since race-
mic product S10 was obtained in 40 min at 1008C in chloro-
benzene.
To demonstrate the robustness of catalyst 6 in the racemi-
sation of (S)-S10, we added new batches of enantiopure S10
every 40 min (Figure 6). By starting from a reaction mixture
Table 5. Racemisation of amines S9–S15 with 2 mol% of iridacycle 6 in
toluene (0.25m).
[
a]
Entry
Substrate
T [8C]
t
1/2(rac) [min]
t
compl.(rac) [min]
[
b]
1
2
3
4
5
6
7
8
9
S9
100
80
dimers
45
–
S10
S10
S10
S11
S12
S13
S14
S15
360
180
40
100
100
100
100
100
100
100
15
4
[
c]
<5
–
120
<90
–
15
[
[
[
d]
b]
b]
960
300
600
960
Figure 6. Kinetic studies of the racemisation of (S)-S10 with 2 mol% of 6
by the sequential addition 0.25 mmol of (S)-S10 every 40 min.
of 0.25 mmol of S10 in chlorobenzene ([S10]=0.25m) in the
presence of 2 mol% of 6 at 1008C, complete racemisation
was obtained after 40 min. At this time, 0.25 mmol of (S)-
S10 was added to the reaction, increasing the concentration
of the reaction mixture to [S10]=0.5m and the substrate/cat-
alyst ratio to 100. Just after addition, the ee was 50%, as ex-
pected. After 2 min, the ee of the product already dropped
to 42%, which shows the high activity of the catalyst under
these conditions. Almost complete racemisation was ob-
tained after 80 min (1.5% ee). A new batch of (S)-S10 was
added at 80 min ([S10]=0.75m, s/c=150) and a sample ana-
lysed 2 min after the addition gave 44% ee for S10. At
[a] Reaction conditions: 0.75 mmol of substrate, 2 mol% of [Ir], 3 mL of
PhMe. [b] More than 60% of dimers were observed. [c] in PhCl. [d] ee of
(
S)-S12=86%.
With primary amine (S)-S9, only dimer formation was ob-
[27]
served (Table 5, entry 1).
Racemisation of (S)-a-methyl-
N-methylbenzylamine (S10) was fast in toluene at 1008C
with 2 mol% of 6 (t1/2(rac) =15 min, entry 3). By using chloro-
benzene instead of toluene, t1/2(rac) was decreased to 4 min
and complete racemisation was obtained after 40 min
(entry 4). (S)-N-benzyl-a-methylbenzylamine (S11) showed
the highest reactivity since complete racemisation was ob-
tained after 15 min (t1/2(rac) <5 min, entry 5). Quinoline was
also investigated as a substrate. Complete racemisation of
(S)-2-methyl-1,2,3,4-tetrahydroquinoline (S12) was obtained
1
20 min, the rate of racemisation slowed down since the ee
of S10 was 11%. Final addition of (S)-S10 ([S10]=1m, s/c=
00) highly influenced the catalytic system in terms of reac-
2
[28]
tivity. At 2 min, after addition, only 48% ee was observed
and 18% ee after 160 min. Complete racemisation was ob-
tained after 6 h.
overnight (entry 6).
Further studies on the racemisation
of cyclic benzylic primary amines, such as (S)-1-aminoindane
(S13) and (S)-1,2,3,4-tetrahydro-1-naphthylamine (S14),
gave good results with complete racemisation in 6 and 10 h,
respectively (t1/2(rac) =120 min, entry 7; t1/2(rac) <90 min,
entry 8). In these cases, major formation of dimers was also
observed. Finally, iridacycle 6 was able to racemise tertiary
amine (S)-a,N,N-trimethylbenzylamine (S15) overnight
(entry 9).
Iridacycles are very efficient in the catalytic racemisation
of chiral secondary and tertiary amines. However, dimer for-
mation was observed during the racemisation of primary
amines. This suggests that the imine resulting from the dehy-
We reinvestigated the racemisation of primary amine (S)-
a-methylbenzylamine (S9) by using 2 mol% of iridacycle 6
in toluene at 1008C. The decrease of enantioselectivity was
found to be low and corresponds to the formation of side
products. A way to overcome the formation of byproducts
might be the addition of dihydrogen during the reaction.
However, the hydrogen pressure must not be too high other-
wise it could inhibit the dehydrogenation step during the
catalysis. A significant improvement was indeed achieved.
Upon using a hydrogen atmosphere of 1 atm in toluene at
Chem. Eur. J. 2009, 15, 12780 – 12790
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12787

B. L. Feringa, J. G. de Vries et al.
drogenation of the chiral amine does not stay coordinated
to the iridium centre and that the rate-determining step
might be the hydrogenation of the imine derivative.
pension was filtered over Al
fraction was collected and concentrated under vacuum to 1–2 mL. Addi-
2 3 3
O by using CH CN as the eluent. A yellow
tion of Et
2
O (4 mL) gave 1 as a yellow solid that was washed with Et
2
O
(
3ꢄ2 mL) and pentane (3ꢄ2 mL). Yield: 200 mg, 0.30 mmol, 74%.
6
1
[
Ru
A
H
U
T
E
N
N
6
H
6
){2-
A
T
N
T
E
U
G
2
NHMe)-C
6
H
4
}
A
H
U
G
R
N
N
(NCMe)]
A
H
U
G
R
N
U
G
6
) (1): H NMR analysis
and
N N
revealed the presence of two pairs of enantiomers, RRuS /SRuR
Conclusion
R
R /S S , in a 56:44 ratio in CD CN. Enantiomer RRuS /S R :
Ru
N
Ru
N
3
N
Ru
N
1
3
H NMR (400.0 MHz, CD HÀH =7.6 Hz, 1H;
3
CN, 298 K): d=7.72 (d,
J
C
6
H
4
), 7.04–6.90 (m, 3H; C H ), 5.57 (s, 6H; C H ), 4.04 (m, 1H; NH),
6
4
6
6
We have synthesized and characterized a small library of
half-sandwich cationic ruthena- and iridacycle complexes
and studied their behaviour in the catalytic racemisation of
chiral amines and alcohols. We found that readily synthe-
sized electron-rich iridacycles 2–8, obtained from commer-
cial available metal precursors and aromatic secondary
amines, are able to racemize both chiral alcohols and
amines. These complexes are among the fastest racemisation
catalysts described. Moreover, they show an unexpected
switchable behaviour. Without activation of iridacycle com-
plexes by KOtBu, no activity towards chiral alcohols was ob-
served, whereas complete racemisation of amines occurred
in the best case in 15 min. The reverse effect was observed
as well since base-activated iridacycles were efficient only
with chiral alcohols; (R)-2-chloro-1-phenylethanol (S2) was
racemized within 5 min at RT, but no reaction occurred with
amines. Alkyl alcohols can also be racemized with longer re-
action times. Mechanistic aspects and applications of these
iridacycles in other chemo-enzymatic dynamic kinetic reso-
lutions are currently being investigated by our group.
2
3
2
3
J
.94 (dd,
J
HÀH =13.6,
J
HÀH =4.4 Hz, 1H; CH
2
N), 3.61 (dd,
J
HÀH =13.6,
),
CN, 298 K): 166.4,
CN), 87.9 (C ), 67.8
CN); enantiomer
3
3
HÀH =11.2 Hz, 1H; CH
2
N), 3.06 (d,
JHÀH =.6.0 Hz, 3H; NCH
3
13
2
.15 ppm (s; CH
3
CN); C NMR (100.6 MHz, CD
3
147.6, 140.2, 127.4, 124.2, 122.0 (C
CH N), 47.3 (NCH ), 1.3 ppm (CH
H NMR (400.0 MHz, CD
), 7.04–6.90 (m, 3H; C
6
H
4
), 118.3 (CH
3
6 6
H
(
2
3
3
R
Ru
R
N
/SRu
S
N
:
1
3
3
CN, 298 K): d=7.96 (d,
J
HÀH =7.6 Hz,1H;
C
6
H
4
6
H
4
), 5.61 (s, 6H; C
6
H
6
), 4.04 (m, 1H; NH),
2
3
2
3
J
.77 (dd,
J
HÀH =14.0,
J
HÀH =5.6 Hz, 1H; CH
=9.0 Hz, 1H; CH N), 2.84 (d, J
2
N), 3.47 (dd,
=6.0 Hz, 3H; NCH ), 2.15 ppm
HÀH 3
J
HÀH =14.0,
3
3
HÀH
2
1
3
(s; CH
39.9, 127.1, 124.5, 122.4 (C
CH N), 45.7 (NCH ), 1.3 ppm (CH
for C16 RuPF : C 39.59, H 3.95, N 5.77; found: C 39.91, H 3.93, N
.71.
3
CN); C NMR (100.6 MHz, CD
), 118.3 (CH
CN); elemental analysis calcd (%)
3
CN, 298 K): d=162.7, 148.3,
1
(
6
H
4
3
CN), 88.3 (C ), 63.2
6 6
H
2
3
3
H
19
N
2
6
5
6
[
Ru
was added to an orange suspension of 1 (25 mg, 0.052 mmol) in iPrOH
2.5 mL). After 20 min, of stirring at RT, all compounds were dissolved
leading to a dark-orange solution. An aliquot was then removed by sy-
ringe, immediately dried in vacuo, and redissolved in CD CN for
6 6 2 6 4
ACHTUNGRTNNGE(U h -C H ){2- ACTHUNGTRNENU(G CH NHMe)-C H }H] (1’): KOtBu (9 mg, 0.077 mmol)
(
3
1
H NMR spectroscopic analysis, which revealed the presence of two pairs
1
of enantiomers in an approximate ratio of 1:1. H NMR (400.0 MHz,
CD
3
3
CN, 298 K): d=7.48 (d, JHÀH =7.5 Hz, 1H; C
6
H
4
), 6.83–6.62 (m, 3H;
2
C
6
H
4
), 5.09 (s, 6H; C
.61 (s, 3H; NCH
CD
6
H
6
), 3.74, 3.09 (AB,
J
HÀH =12.6 Hz, 2H; CH
2
N),
1
2
3
), À7.13 ppm (s, 1H; RuH); H NMR (400.0 MHz,
3
3
CN, 298 K): d=7.35 (d, JHÀH =7.5 Hz, 1H; C
6
H
4
), 6.83–6.62 (m, 3H;
2
C
6
H
4
), 5.09 (s, 6H; C
.64 (s, 3H; NCH
), À7.85 ppm (s, 1H; RuH).
Synthesis and characterisation of cationic iridacycle complexes: The pro-
6 6 2
H ), 3.79, 3.46 (AB, JHÀH =12.0 Hz, 2H; CH N),
2
3
Experimental Section
[
19]
cedure described by Pfeffer et al. for analogous compounds was used.
In a typical experiment, a 50 mL Schlenk tube was thoroughly flame-
dried and put under an atmosphere of nitrogen, after which the following
General: Starting materials were purchased from Acros, Sigma–Aldrich,
Strem, Merck or Alfa Aesar, and were used as received unless stated oth-
erwise. All solvents were reagent grade and, if necessary, dried and dis-
tilled prior to use. Toluene and diethylether were distilled over Na/benzo-
phenone. Column chromatography was performed on silica gel (Aldrich
compounds were added, respectively, in acetonitrile (5 mL): [Ir
(Cp*)] (159 mg, 0.2 mmol), amine (0.40 mmol), NaOH (16 mg,
.4 mmol) and KPF (147 mg, 0.8 mmol). This mixture was stirred at
58C for 16 to 50 h. The mixture was then cooled down to RT, washed
2
ACHTUNGTRENNUNG( Cl )-
A
H
U
G
E
N
N
2
0
4
6
6
0, 230–400 mesh) or activated neutral aluminium oxide (Merck alumini-
um oxide 90 neutral activated). TLC was performed on silica gel 60/Kie-
1
13
with hexane and filtered over neutral aluminium oxide (eluent: MeCN).
The resulting solution was concentrated in vacuo. Subsequent stripping
with dry Et
selguhr F254 or neutral aluminium oxide 60 F254. H and C NMR spectra
1
were recorded in CDCl
3
on a Varian VXR300 (299.97 MHz for H,
a
1
3
1
2
O yielded the desired iridacycle complex.
7
1
5.48 MHz for C) or
Varian AMX400 (399.93 MHz for H,
1
3
5
00.59 MHz for C) spectrometer. Chemical shifts are reported in d
[Ir
A
H
U
G
R
N
U
G
5
Me
5
)(C -2-CH NH (NCCH )](PF ) (2): Complex 2 was ob-
6
H
4
2
2
)
A
H
U
G
R
N
N
3
A
H
N
T
E
U
G
6
1
values (ppm) relative to the solvent peak (CHCl
3
,
d
=7.26 ( H),
7.0 ppm ( C)). The following abbreviations are used to indicate multi-
plicity: s (singlet), d (doublet), t (triplet), q (quartet), m (multiplet), br
broad). Mass spectra (HRMS) were performed on a Jeol JMS-600H.
tained as a brown-solid precipitate (216 mg, 0.35 mmol, 87%) by using
1
3
1
7
benzylamine (43 mg). H NMR (400.0 MHz, CDCl
3
, 298 K): d=7.40 (dd,
HÀH =7.6 Hz, 1H; Ph-H), 6.98
HÀH =1.3, 7.2 Hz, 2H; Ph-H), 4.35 (s, 2H; NH ), 4.24 (m, 1H;
), 3.86 (m, 1H; CH ), 2.39 (s, 3H; NCCH ), 1.75 ppm (s, 15H; C
); C NMR (100.6 MHz, CDCl , 298 K): d=150.2, 149.1, 135.9,
127.4, 123.9, 121.3 (CH CN), 89.6, 56.1, 9.1 ppm (CH CN); HRMS (EI+
): m/z: calcd. for C17
3
3
J
HÀH3=1.4, 7.1 Hz, 1H; Ph-H), 7.12 (d,
J
(
(td,
CH
J
2
GCMS spectra were recorded on a Hewlett Packard HP6890 equipped
with a HP1 column and an HP 5973 Mass Selective Detector. GC analy-
sis were performed on a Shimadzu GC-17 A or a Hewlett Packard
HP6890 chromatograph equipped with the columns indicated for each
compound separately. HPLC analysis was performed on a Shimadzu
HPLC system equipped with two LC-10AD vp solvent delivery systems,
a DGU-14 A degasser, a SIL-10AD vp auto injector, an SPD-M10 A vp
diode array detector, a CTO-10 A vp column oven and an SCL-10A vp
system controller by using the columns indicated for each compound sep-
arately. Elemental analyses were carried out on a EuroVector Euro EA
elemental analyser.
2
2
3
5
-
1
3
A
H
U
G
R
N
U
G
3
)
5
3
3
3
+
+
H
23NIr : 434.1460 (ÀCH
3
CN/PF
H F N PRu: C 39.59, H
19 6 2
6
, [M] ); found:
434.1441; elemental analysis calcd (%) for C16
3.95, N 5.77; found: C 39.91, H 3.93, N 5.71.
5
[
Ir
A
H
U
G
R
N
U
G
5
Me
5
) (C
6
H
4
-2-CH
2
NHCH
3
)
A
H
N
T
N
U
G
3
)]
A
H
N
R
N
N
(PF
6
) (3): Complex 3 was
obtained as a yellow powder (233 mg, 0.37 mmol, 92%) by using N-meth-
1
ylbenzylamine (49 mg). H NMR (400.0 MHz, CDCl
3
, 298 K): d=7.36 (d,
HÀH =6.9 Hz, 1H; Ph-H), 7.00 (t,
HÀH =7.0 Hz, 1H; Ph-H), 4.40 (s,
3
3
J
J
HÀH =6.9 Hz, 1H; Ph-H), 7.07 (d,
HÀH =7.2 Hz, 1H; Ph-H), 6.92 (t,
J
3
3
J
Synthesis and characterisation of the cationic ruthenacycle complex: The
1H; NH), 4.17 (s, 1H; CH
(s, 3H; NCCH ), 1.68 ppm (s, 15H; C
CDCl , 298 K): d=150.2, 146.6, 135.2, 134.4, 127.4, 123.6, 121.0, 119.1
(CH CN), 89.8, 67.2, 44.7, 9.0 ppm (CH CN); HRMS (EI+): m/z: calcd.
3 6
for C18 CN/PF , [M] ); found: 448.1617; elemen-
2
), 3.68 (s, 1H; CH
2
), 3.11 (s, 3H; NCH
) ); C NMR (100.6 MHz,
3
), 2.38
[
19]
13
procedure described by Pfeffer et al. was used. A suspension of [Ru-
3
5
A
H
U
G
R
N
U
G
3 5
6
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(Cl
.56 mmol), NaOH (32 mg, 0.8 mmol) and KPF
CH CN (6 mL) was stirred at RT for 3 d. The resulting dark-yellow sus-
2
)
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(h -C
6
H
6
)] (200 mg, 0.4 mmol), N-methylbenzylamine (72 mL,
3
0
6
(294 mg, 1.6 mmol) in
3
3
+
+
3
H
25IrN : 448.1616 (ÀCH
12788
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 12780 – 12790

Fast Racemisation of Chiral Amines and Alcohols
FULL PAPER
tal analysis calcd (%) for C20
C 38.64, H 4.54, N 3.99.
H
28
F
6
N
2
PIr: C 37.91, H 4.45, N 4.42; found:
Schlenk flask under an atmosphere of nitrogen, 3 (6.3 mg, 10 mmol) and
KOtBu (1.3 mg, 12 mmol) were dissolved in freshly distilled toluene
(3 mL), after which the mixture was stirred for 15 min at RT. The result-
ing solution was slowly transferred by syringe into another Schlenk tube
containing water (3 mL) and a chiral alcohol (l200 mmol). Care was taken
not to disturb phase separation, since this resulted in deactivation of the
catalyst. The reaction was monitored by periodically taking 0.1 mL ali-
5
[
Ir
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(h -C
5
Me
5
)(C
6
H
4
-2-CH
2
N
A
H
U
G
E
N
N
(CH
3
)
2
)
A
H
U
G
R
N
U
G
3
A
H
U
G
E
N
N
6
obtained as a yellowish powder (212 mg, 0.33 mmol, 82%) by using N,N-
dimethylbenzylamine (54 mg). H NMR (400.0 MHz, CDCl , 298 K): d=
3
1
3
3
7
7
3
1
1
9
.43 (d,
.05 (t,
J
HÀH =7.6 Hz, 1H; Ph-H), 7.13 (d,
J
HÀH =6.8 Hz, 1H; Ph-H),
3
3
J
HÀH =6.4 Hz, 1H; Ph-H), 7.00 (t,
), 2.94 (brs, 6H; NC(CH
); C NMR (100.6 MHz, CDCl
49.3, 146.7, 134.7, 127.4, 123.8, 122.4, 120.1 (CH CN), 90.7, 76.2, 57.5,
.0 ppm (CH CN); HRMS (EI+): m/z: calcd for C19
ÀCH CN/PF
JHÀH =7.2 Hz, 1H; Ph-H),
2
quots from the mixture, filtering them over silica gel (eluent: Et O) and
.79 (s, 2H; CH
.67 ppm (s, 15H; C
2
A
H
U
G
R
N
U
G
3
)
2
), 2.55 (brs, 3H; NCCH
3
),
1
3
analysing the resulting samples by chiral GC. Chirasil-Dex CB column
5
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(CH
3
)
5
3
, 298 K): d=
À1
(
25 mꢄ0.25 mmꢄ0.25 mm), gas vector: helium, flow=1 mLmin , injec-
3
À1
+
tor: 2508C, program: 1008C for 3 min, 1208C (158Cmin ) for 15 min,
1408C (158Cmin ) for 15 min, 1008C (158Cmin ): 1-phenylethanol
3
H27IrN : 462.1773
À1
À1
+
(
3
6
, [M] ); found: 462.1751.
5
(S1):
t
(R) =12.8,
t
(S) =13.7 min; program: 1008C for 5 min, 1608C
[
Ir
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(h -C
5
Me
5
) (C -2-CHNCH (NCCH )] ACHTUNGETRNNUNG( PF ) (5): Complex 5 was ob-
6
H
4
3
)
A
H
U
G
R
N
U
G
3 6
À1
À1
(
128Cmin ) for 10 min, 1008C (108Cmin ): 1-(4-fluorophenyl)ethanol
tained as a yellow powder (227 mg, 0.36 mmol, 91%) by using N-benzyli-
1
(S3): t(R) =10.5, t(S) =10.8 min; program: isotherm 608C: 2-butanol (S7):
(S) =11.5, t(R) =13.1 min; program: isotherm 808C, 2-hexanol (S8): t(S)
3
denemethylamine (48 mg). H NMR (400.0 MHz, CDCl , 298 K): d=8.37
3
3
t
=
(
1
1
C
1
8
(
C
s, 1H; CH), 7.69 (d,
H; Ph-H), 7.21 (t,
J
HÀH =7.6 Hz, 1H; Ph-H), 7.58 (d,
J
J
HÀH =7.6 Hz,
HÀH =7.2 Hz,
3
3
12.7, t(R) =15.5 min; GTA column (30 mꢄ0.25 mmꢄ0.25 mm), gas vector:
J
HÀH =9.4 Hz, 1H; Ph-H), 7.11 (t,
), 2.37 (s, 3H; NCCH
); C NMR (100.6 MHz, CDCl , 298 K): d=178.4, 161.7, 146.8,
CN), 91.0, 51.2,
À1
helium, flow=1 mLmin , injector 2508C, program: 508C, for 3 min,
H; Ph-H), 3.94 (s, 3H; CH
3
3
), 1.77 ppm (s, 15H;
À1
À1
1
3
958C (108Cmin ) for 10 min, 1258C (108Cmin ) for 20 min, 508C
5
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(CH
3
)
5
3
À1
(
108Cmin ): 2-chloro-1-phenylethanol (S2):
t
(R) =12.8,
t
(S) =13.7 min;
39.1, 134.4, 132, 129.6, 128.9, 128.7, 123.5, 118.9 (CH
.8 ppm (CH CN); HRMS (EI+): m/z: calcd for C18
CN/PF
PIr: C 38.03, H 4.15, N 4.44; found: C 38.04, H 4.24, N 4.26.
Me )(C -2-C (NCCH )](PF ) (6): Complex 6 (237 mg,
.36 mmol, 91%) was obtained as a yellowish powder by using 2-phenyl-
3
+
program: isothem 1008C: 1-phenylbutanol (S4): t(R) =15.9, t
HPLC, OD-H column, hexane/isopropanol 95:5, flow=0.5 mLmin : 1-
S
=16.4 min;
3
H
23IrN : 446.1460
À1
+
ÀCH
3
6
, [M] ); found: 446.1455; elemental analysis calcd (%) for
(
(
2-naphthyl)ethanol (S5):
25 cm), hexane/isopropanol 98:2, flow=0.5 mLmin : 1-indanol (S6):
t
(S) =28.6,
t
(R) =29.8 min; OD-H column
20 26 6 2
H F N
À1
5
[
Ir
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(h -C
5
5
6
H
4
3
H
5
N
2
)
A
H
U
G
R
N
U
G
3
A
H
U
G
E
N
N
6
t
(S) =30.4, t(R) =34.2 min.
Typical procedure for catalytic racemisation of amines: In a thoroughly
flame-dried Schlenk flask under nitrogen atmosphere, amine
0
1
2
-imidazoline (59 mg). H NMR (400.0 MHz, CDCl
HÀH =7.6 Hz, 1H; Ph-H), 7.33 (d,
HÀH =7.4 Hz, 1H; Ph-H), 7.07 (t,
3
, 298 K): d=7.67 (d,
JHÀH =7.6 Hz, 1H; Ph-H), 7.20 (t,
3
3
a
J
J
3
3
(0.75 mmol) was added to a solution of iridacycle (0.015 mmol, 2 mol%)
and dodecane (100 mL) in dry and degassed toluene (3 mL) at the desired
temperature. The reaction mixture was then stirred at this temperature.
For GC analysis, samples were taken (100 mL), filtered over silica gel
(eluent: Et O) and diluted in dichloromethane (1.0 mL) after adding
2
2 drops of triethylamine and acetic anhydride and were analysed by
J
HÀH =7.6 Hz, 1H; Ph-H), 5.91 (s,
), 1.78 ppm (s, 15H;
, 298 K): d=177.7, 157.3, 135.3,
CN), 89.8, 51.8, 45.9, 9.2 ppm
1
C
H; NH), 3.97 (s, 4H; (CH
2
)
2
), 2.33 (s, 3H; NCCH
3
1
3
5
A( CH
C
H
T
U
N
G
T
R
E
N
N
U
N
G
3
)
5
); C NMR (100.6 MHz, CDCl
3
1
34.8, 131.8, 125.5, 123.2, 117.5 (CH
3
+
(
(
CH
ÀCH
3
CN); HRMS (EI+): m/z: calcd for
CN/PF
3 6
C
19
H
24IrN
2
:
473.1563
, [M] ); found: 473.1542; elemental analysis calcd (%) for
PIr: C 38.29, H 4.13, N 6.38; found: C 37.93, H 4.07, N 6.18.
Me )(C -2-C (NCCH )](PF ) (7): Complex 7 was ob-
tained as a yellowish powder (249 mg, 0.38 mmol, 95%) by using 2-phe-
+
chiral GC analysis. CP Chiralsil-Dex CB column (25 mꢄ0.25 mmꢄ
21 27 6 3
C H F N
À1
0
1
1
1
1
3
1
.25 mm), gas vector: helium, flow: 1 mLmin , injector: 2508C, program:
5
[
Ir
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(h -C
5
5
6
H
4
3
H
3
N
2
)
A
H
U
G
R
N
U
G
3
A
H
U
G
E
N
N
6
À1
À1
258C for 4 min, 1408C (38Cmin ), 1808C (108Cmin ) for 30 min,
258C (108Cmin ): acylated a-methylbenzylamine (S9): t(S) =12.3, t(R)
2.5 min; acylated a,N-dimethylbenzylamine (S10):
2.9 min; acylated N-benzyl-a-methylbenzylamine (S11): t(S) =33.4, t(R)
À1
=
=
=
1
3
nylimidazole (58 mg). H NMR (400.0 MHz, CDCl , 298 K): d=10.55 (s,
1
t(S) =12.8, t
(R)
3
3
H; NH), 7.65 (d, JHÀH =7.6 Hz, 1H; Ph-H), 7.56 (d, JHÀH =7.2 Hz, 1H;
3 5 3 5
Ph-H), 7.07 (m, 4H), 2.24 (s, 3H; NCCH ), 1.74 ppm (s, 15H; C AHCTUGNENRTNUG( CH ) );
3.9 min; acylated aminoindane (S13): t(S) =16.0, t(R) =16.3 min; acylated
1
3
C NMR (100.6 MHz, CDCl
23.8, 122.2, 118.2 (CH CN), 89.9, 8.9 ppm (CH
z: calcd for
3
, 298 K): 158.6, 152.9, 135.5, 135.0, 129.9,
,2,3,4-tetrahydronaphthylamine (S14):
t
(S) =18.8,
t
(R) =19.4 min; pro-
1
3
3
CN); HRMS (EI+): m/
À1
À1
gram: 958C for 15 min, 1808C (58Cmin ) for 10 min, 958C (108Cmin ):
+
+
C
19
H
22IrN
2
:
471.1412 (ÀCH
3
CN/PF
6
, [M] ); found:
1
H
-methyltetrahydroquinoline (S12): t(S) =25.7, t(R) =26.0 min; HPLC, OD-
4
71.1409.
À1
column (25 cm), hexane/isopropanol 99.8:0.2, flow=1.0 mLmin
:
5
[
Ir
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(h -C
5
Me
5
) (C
6
H
4
-2-C
3
H
4
NO)
A
H
U
G
E
N
N
(NCCH
3
)]
A
H
U
G
E
N
N
(PF
6
) (8): Complex 8 was ob-
N,N,a-trimethylbenzylamine (S15): t(R) =5.4, t(S) =6.2 min.
tained as a yellowish powder (171 mg, 0.26 mmol, 65%) by using 2-phe-
1
nyloxazoline (58 mg). H NMR (400.0 MHz, CDCl
3
, 298 K): d=7.70 (d,
HÀH =7.3 Hz, 1H; Ph-H), 7.29 (t,
HÀH =7.8 Hz, 1H; Ph-H), 5.03 (m,
), 4.29 (m, 1H; CH ), 4.04 (m, 1H; CH ),
), 1.81 ppm (s, 15H; C NMR
, 298 K): d=177.5, 153.4, 131.6, 129.4, 127.4, 123.3,
19.4, 117.2, 86.3, 68.7, 47.2, 5.5 ppm; elemental analysis calcd (%) for
OPIr: C 38.24, H 3.97, N 4.25; found: C 38.88, H 4.03, N 4.08.
Crystal data for 8: C21 IrN OP; M =659.53; monoclinic; space group
P2 /n; a=12.7656(13), b=13.8437(14),
3
3
1
H, JHÀH =7.5 Hz; Ph-H), 7.47 (d, J
3
3
J
HÀH =7.3 Hz, 1H; Ph-H), 7.13 (t,
H; CH ), 4.89 (m, 1H; CH
.41 (s, 3H; NCCH
J
1
2
2
2
2
2
Acknowledgements
1
3
3
5 3 5
C ACHTUNRTGEN(NUG CH ) );
(
100.6 MHz, CDCl
3
Part of the research was carried out in the framework of the Ultimate
Chiral Technologies project funded in part by FNN (Foundation North-
ern Netherlands) and EFRO (European Fund for Regional Develop-
ment). Financial support from the Netherlands Organisation for Scientific
Research (NWO-CW), the Dutch Ministry of Economic Affairs, Royal
DSM N.V., and N.V. Organon, administered through the IBOS program
is gratefully acknowledged.
1
21 26 6 2
C H F N
H
26
F
6
2
r
1
c=13.4841(14) ꢅ;
=1.888 gcm ; F
b=
3
À3
1
03.1777(13)8; V=2320.2(4) ꢅ ; Z=4; D
x
A
H
U
G
R
N
N
(000)=1280;
À1
m=58.87 cm
; l ACHTUNGTRENNUNG( MoKa)=0.71073 ꢅ; T=100(1) K; 17885 reflections
2
measured; GoF=1.064, wR(F )=0.0751 for 4910 unique reflections and
95 parameters, and R(F)=0.0308 for 4151 reflections obeying F
ꢀ 4.0
s(F ) criterion of observability.
2
o
o
[
1] M. Breuer, K. Ditrich, T. Habicher, B. Hauer, M. Keßeler, R.
Stꢆrmer, T. Zelinski, Angew. Chem. 2004, 116, 806–843; Angew.
Chem. Int. Ed. 2004, 43, 788–824.
Typical procedure for catalytic racemisation of alcohols: In a thoroughly
flame-dried Schlenk flask under an atmosphere of nitrogen, catalyst
(
37.5 mmol) and KOtBu (41.2 mmol) were dissolved in freshly distilled tol-
[2] a) B. Singaram, C. T. Goralski, in Transition Metals for Organic Syn-
thesis, Vol. 2 (Eds.: M. Beller, C. Bolm), Wiley-VCH, Weinheim,
1998, pp. 147–154; b) H. U. Blaser, F. Spindler in Comprehensive
Asymmetric Catalysis, Vol. 1 (Eds.: E. N. Jacobsen, A. Pfaltz, H. Ya-
mamoto), Springer, Berlin, 1999, pp. 247–265; c) F. Spindler, H. U.
Blaser in The Handbook of Homogeneous Hydrogenation, Vol. 3
uene (2.4 mL), after which chiral alcohol (0.75 mmol) was added. The re-
action was monitored by periodically taking 0.1 mL aliquots from the
mixture, filtering them over silica gel (eluent: Et O) and analysing the re-
2
sulting samples by chiral GC. For the reactions performed in the pres-
ence of water: in a typical experiment, in a thoroughly flame-dried
Chem. Eur. J. 2009, 15, 12780 – 12790
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12789

B. L. Feringa, J. G. de Vries et al.
(
Eds.: J. G. de Vries, C. J. Elsevier), Wiley-VCH, Weinheim, 2007,
[16] a) A. J. Blacker, M. Stirling, M. I. Page, Org. Process Res. Dev. 2007,
11, 642–648; b) M. Stirling, A. J. Blacker, M. I. Page, Tetrahedron
Lett. 2007, 48, 1247–1250; c) A. J. Blacker, M. Stirling, WO2006/
046059, 2006, to Avecia Pharmaceuticals Ltd.
pp. 1193–1214.
[
[
3] R. M. Kellogg, B. Kaptein, T. R. Vries, Top. Curr. Chem. 2007, 269,
1
59–197.
4] a) Enzyme Catalysis in Organic Synthesis: A Comprehensive Hand-
book, Eds: K. Drauz, H. Waldmann), Wiley-VCH, Weinheim, 1995;
b) Hydrolases in Organic Synthesis (Eds.: U. T. Bornscheuer, R. J.
Kaslauskas), Wiley-VCH, Weinheim, 1999; c) E. Fogassy, M.
Nꢇgrꢈdi, D. Kozma, G. Egri, E. Pꢈlovics, V. Kiss, Org. Biomol.
Chem. 2006, 4, 3011–3030.
[17] a) J. B. Sortais, V. Ritleng, A. Voelklin, A. Holuigue, H. Smail, L.
Barloy, C. Sirlin, G. K. M. Verzijl, J. A. F. Boogers, A. H. M. de V-
ries, J. G. de Vries, M. Pfeffer, Org. Lett. 2005, 7, 1247–1250; b) J. G.
de Vries, G. K. M. Verzijl, A. H. M. de Vries, V. Ritleng, A. M. G.
Voelklin, 2006, WO2006/050988 to DSM IP Assets bv; c) J.-B. Sor-
tais, L. Barloy, C. Sirlin, A. H. M. de Vries, J. G. de Vries, M. Pfeffer,
Pure Appl. Chem. 2006, 78, 457–462.
[
5] a) B. Martꢉn-Matute, J. E. Bꢀckvall, Curr. Opin. Chem. Biol. 2007,
[
18] D. L. Davies, O. Al-Duaij, J. Fawcett, M. Giardiello, S. T. Hilton,
D. R. Russell, Dalton Trans. 2003, 4132–4138.
1
1, 226–232; b) J. E. Bꢀckvall in Asymmetric Synthesis—The Essen-
tials, (Eds.: M. Christmann, S. Brꢀse), Wiley-VCH, Weinheim, 2007,
pp. 171–175; c) M.-J. Kim, Y. Ahn, J. Park in Biocatalysis in the
Pharmaceutical and Biotechnology Industries (Ed.: R. N. Patel),
CRC-Press, Boca Raton, 2007, pp. 249–272; d) Y. Ahn, S.-B. Ko,
M.-J. Kim, J. Park, Coord. Chem. Rev. 2008, 252, 647–658; e) A.
Kamnal, M. A. Azhar, T. Krishnaji, M. S. Malik, S. Azeeza, Coord.
Chem. Rev. 2008, 252, 569–592; f) H. Pellissier, Tetrahedron 2008,
[
19] a) J. B. Sortais, N. Pannetier, A. Holuigue, L. Barloy, C. Sirlin, M.
Pfeffer, N. Kyritsakas, Organometallics 2007, 26, 1856–1867; b) J. B.
Sortais, N. Pannetier, N. Clement, L. Barloy, C. Sirlin, M. Pfeffer, N.
Kyritsakas, Organometallics 2007, 26, 1868–1874; c) N. Pannetier, J.-
B. Sortais, P. S. Dieng, L. Barloy, C. Sirlin, M. Pfeffer, Organometal-
lics 2008, 27, 5852–5859.
[
20] a) S. Arita, T. Koike, Y. Kayaki, T. Ikariya, Angew. Chem. 2008, 120,
6
4, 1563–1601.
2
481–2483; Angew. Chem. Int. Ed. 2008, 47, 2447–2449; b) S. Arita,
[
[
[
[
6] P. M. Dinh, J. A. Howarth, A. R. Hudnott, J. M. J. Williams, H.
T. Koike, Y. Kayaki, T. Ikariya, Chem. Eur. J. 2008, 14, 1479–1485.
21] a) V. Ritleng„ P. Bertani, M. Pfeffer, C. Sirlin, J. Hirschinger, Inorg.
Chem. 2001, 40, 5117–5122; b) M. Robitzer, V. Ritleng, C. Sirlin, A.
Dedieu, M. Pfeffer, C. R. Chimie, 2002, 5, 467–472.
Harris, Tetrahedron Lett. 1996, 37, 7623–7626.
[
7] A. L. E. Larsson, B. A. Persson, J.-E. Bꢀckvall, Angew. Chem. 1997,
1
09, 1256–1258; Angew. Chem. Int. Ed. Engl. 1997, 36, 1211–1212.
8] Y. Shvo, D. Czarkie, Y. Rahamim, D. F. Chodosh, J. Am. Chem. Soc.
986, 108, 7400–7402.
9] G. K. M. Verzijl, J. G. de Vries, Q. B. Broxterman, Tetrahedron:
Asymmetry 2005, 16, 1603–1610.
[
[
22] K. Everaere, A. Mortreux, M. Bulliard, J. Brusse, A. van der Gen,
G. Nowogrocki, J.-F. Carpentier, Eur. J. Org. Chem. 2001, 275–291.
23] a) S. Fernandez, M. Pfeffer, V. Ritleng, C. Sirlin, Organometallics
1
1
999, 18, 2390–2394; b) V. Ritleng, Ph. D. Dissertation, 2001, Uni-
[
10] a) G. Vitt, H. Siegel, M. Schneider, DE2851039, 1980, to BASF
A.G; b) T. Inoue, Y. Hirayama, JP10072410, 1998, to Nagase & Co.
Ltd; c) A. N. Parvulescu, N. Andrei, P. A. Jacobs, D. E. de Vos, Adv.
Synth. Catal., 2008, 350, 113–121; d) J. M. Paul, G. A. Potter,
WO9721662, 1997, to Chirotech Technology Ltd.
11] a) S. Nagata, K. Hagiya, Y. Yamada, H. Goto, EP778260, 1997, to
Sumitomo Chemical Co; b) S. Escoubet, S. Gastaldi, N. Vanthuyne,
G. Gil, D. Siri, M. P. Bertrand, Eur. J. Org. Chem. 2006, 3242–3250.
12] M. T. Reetz, K. Schimossek, Chimia 1996, 50, 668–669.
versitꢂ Louis Pasteur, Strasbourg (France).
[
24] A. Trꢀff, K. Bogꢈr, M. Warner, J.-E. Bꢀckvall, Org. Lett. 2008, 10,
4
807–4810.
[
25] a) J. H. Lutje Spelberg, L. Tang, R. M. Kellogg, D. B. Janssen, Tetra-
hedron: Asymmetry 2004, 15, 1095–1102; b) L. Tang, D. E. Torres
Pazmino, M. W. Fraaije, R. M. de Jong, B. W. Dijkstra, D. B. Janssen,
Biochemistry 2005, 44, 6609–6618; c) R. M. Haak, C. Tarabiono,
D. B. Janssen, A. J. Minnaard, J. G. de Vries, B. L. Feringa, Org.
Biomol. Chem. 2007, 5, 318–323.
[
[
[
13] a) A. N. Parvulescu, D. E. de Vos, P. A. Jacobs, Chem. Commun.
[
26] R. M. Haak, F. Berthiol, T. Jerphagnon, A. J. A. Gayet, C. Tarabio-
no, C. C. Postema, V. Ritleng, M. Pfeffer, D. B. Janssen, A. J. Min-
naard, B. L. Feringa, J. G. de Vries, J. Am. Chem. Soc. 2008, 130,
13508–13509.
2
005, 5307–5309; b) A. N. Parvulescu, P. A. Jacobs, D. E. de Vos,
Chem. Eur. J. 2007, 13, 2034–2043.
[
[
14] M.-J. Kim, W.-H. Kim, K. Han, Y. K. Choi, J. Park, Org. Lett. 2007,
9
, 1157–1159.
[27] S.-I. Murahashi, N. Yoshimura, T. Tsumiyama, T. Kojima, J. Am.
Chem. Soc. 1983, 105, 5002–5011.
[28] N. Mr Sˇ ic, L. Lefort, J. A. F. Boogers, A. J. Minnaard, B. L. Feringa,
15] a) O. Pꢊmies, A. H. ꢋll, J. S. M. Samec, N. Hermanns, J.-E. Bꢀckvall,
Tetrahedron Lett. 2002, 43, 4699–4702; b) J. Pꢀtzold, J.-E. Bꢀckvall,
J. Am. Chem. Soc. 2005, 127, 17620–17621; c) C. E. Hoben, L.
Kanupp, J.-E. Bꢀckvall, Tetrahedron Lett. 2008, 49, 977–979;
d) L. K. Thalꢂn, D. Zhao, J.-B. Sortais, J. Paetzold, C. Hoben, J.-E.
Bꢀckvall, Chem. Eur. J. 2009, 15, 3403–3410.
J. G. de Vries, Adv. Synth. Catal. 2008, 350, 1081–1089.
Received: July 28, 2009
Published online: October 15, 2009
12790
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 12780 – 12790