New chiral lanthanide amide ate complexes for the catalysed synthesis of scalemic nitrogen-containing heterocycles

Article abstract of DOI:10.1002/chem.200701090

New chiral binaphthylamido yttrium and ytterbium ate complexes with lithium and potassium counterions have been synthesised and characterised. X-ray structures have been obtained for [Li(thf)₄][Ln{(R)-C ₂₀H₁₂-(NC₅H₉)₂} ₂] (Ln = Yb, Y) and [K(thf)₅] [Yb{(R)-C₂₀H ₁₂(NCH₂CMe₃)₂}₂] as isostructural complexes. The efficiency of these complexes for the enantioselective intramolecular hydroamination was examined. [Li(thf) ₄][Yb{(R)-C₂₀H₁₂(NC₅H ₉)₂}₂] afforded the highest enantiomeric excess (up to 87%) for the synthesis of a spiropyrrolidine, while [Li(thf) ₄][Y{(R)-C₂₀H₁₂-(NC₅H ₉)₂}₂] proved to be slightly more active. The role of the counter cation in the active catalytic species was evidenced by the comparison between lithium and potassium ate complexes. The most active catalyst of this series, [Li(thf)₄][Yb{(R)-C₂₀H ₁₂(NCH₂CMe₃)₂}₂], was successfully used for the cyclisation of aminopentenes with internal double bonds.

Full text of DOI:10.1002/chem.200701090

FULL PAPER
DOI: 10.1002/chem.200701090
New Chiral Lanthanide Amide Ate Complexes for the Catalysed Synthesis of
Scalemic Nitrogen-Containing Heterocycles
Isabelle Aillaud,[a] Jacqueline Collin,*[a] Carine Duhayon,[b] RØgis Guillot,[c]
Dmitrii Lyubov,[d] Emmanuelle Schulz,*[a] and Alexander Trifonov[d]
Abstract: New chiral binaphthylamido
yttrium and ytterbium ate complexes
with lithium and potassium counterions
have been synthesised and character-
ised. X-ray structures have been ob-
was
C20H12
examined.
[Li
N
active. The role of the counter cation
in the active catalytic species was evi-
denced by the comparison between
lithium and potassium ate complexes.
The most active catalyst of this series,
[Li(thf)4][Yb{(R)-C20H12(NCH2CMe3)2}2],
was successfully used for the cyclisation
of aminopentenes with internal double
bonds.
ACHTREUNG
enantiomeric excess (up to 87%) for
the synthesis of spiropyrrolidine,
while [Li(thf)4][Y{(R)-C20H12-
(NC5H9)2}2] proved to be slightly more
a
AHCTREUNG
tained for [Li
(NC5H9)2}2] (Ln=Yb, Y) and [K
[Yb{(R)-C20H12(NCH2CMe3)2}2] as iso-
N
ACHTREUNG
A
ACHTREUNG
ACHTREUNG
Keywords: asymmetric catalysis
heterocycles
lanthanides
structural complexes. The efficiency of
these complexes for the enantioselec-
tive intramolecular hydroamination
· hydroamination
Introduction
lisation of functionalised substrates. Until now, various
amino–alkene, amino–diene and amino–allene derivatives
were successfully enantioselectively transformed into the
corresponding heterocycles by using lanthanide complexes.[1]
This research field was initiated by the group of Marks at
the beginning of the nineties, who proved that scalemic pyr-
rolidines (up to 74% ee)[2] and a piperidine (up to 67% ee)[3]
could be obtained in the presence of chiral metallocenes.
Almost ten years later, the first use of chiral non-metallo-
cene lanthanide complexes was reported by Scott et al. for
promoting the cyclisation of 2,2-dimethylpent-4-enylamine
as a test substrate. They synthesised chiral bisarylamines[4]
and bisaryloxides[5] as ligands for the preparation of yttrium
and lanthanum complexes, respectively. The latter promoted
the formation of the corresponding pyrrolidine albeit with
low activity (reaction performed at 708C) and 61% enantio-
meric excess. At the same time, the group of Hultzsch pre-
pared various rare-earth bisphenolate and bisnaphtholate
derivatives.[6] The corresponding yttrium complex afforded
moderate activities in this transformation with enantiomeric
excesses up to 57% for the intramolecular hydroamination
of pent-4-enylamine. Hultzsch also described the kinetic res-
olution of 2-aminohex-5-ene with this catalyst. Simultane-
ously, we published the preparation of lanthanide amide ate
complexes and obtained up to 53% ee for the cyclisation of
a new substrate (C-(1-allylcyclohexyl)methylamine) at low
temperature by using an ytterbium catalyst.[7] In 2003,
The development of efficient methodologies for the synthe-
sis of scalemic nitrogen-containing heterocycles is of high
importance in the context of the economical and environ-
mental preparation of sophisticated substrates with biologi-
cal activities. Intramolecular asymmetric catalytic hydroami-
nation can readily answer this challenge by the efficient cyc-
[a] I. Aillaud, Dr. J. Collin, Dr. E. Schulz
Laboratoire de Catalyse MolØculaire, ICMMO
UMR CNRS 8182, UniversitØ Paris-Sud, 91405 Orsay (France)
Fax : (+33)169-154-680
E-mail: jacollin@icmo.u-psud.fr
emmaschulz@icmo.u-psud.fr
[b] Dr. C. Duhayon
Laboratoire de Chimie de Coordination
205 route de Narbonne, 31077 Toulouse (France)
[c] Dr. R. Guillot
ICMMO, UniversitØ Paris-Sud
91405 Orsay (France)
[d] D. Lyubov, Prof. A. Trifonov
G. A. Razuvaev Institute of Organometallic Chemistry of
Russian Academy of Sciences
Tropinia 49, 603600 Nizhny Novgorod GSP-445 (Russia)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author. NMR spectra for
ligand (R)-(+)-2,2’-bis(cyclopentylamino)-1,1’-binaphthyl and com-
plexes Li-5b and K-5b.
Chem. Eur. J. 2008, 14, 2189 – 2200
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2189

Marks reported the easy synthesis of enantioselective cata-
lysts by the reaction of lanthanide amides with commercially
available C2-symmetric bisoxazolines.[8] These complexes
proved to be active at room temperature (up to 67% ee for
the test substrate) and promoted the hydroamination/cycli-
sation of several aminopentenes, aminohexenes and amino-
dienes.
Based on these preliminary results, these authors and
other groups attempted both to find new active and enantio-
selective complexes to promote the asymmetric intramolecu-
lar hydroamination reaction, but also to extend their appli-
cation to the transformation of other substrates. We have
synthesised ytterbium and lutetium ionic complexes derived
from enantiopure, substituted (R)-binaphthylamine ligands
with isopropyl or cyclohexyl substituents; these complexes
proved to be efficient catalysts for the cyclisation of various
aminopentene derivatives (with up to 78% ee) and also to
promote the formation of a piperidine (45% ee at 608C).[9]
Kim and Livinghouse[10] have synthesised amido yttrium
tions. The group of Schafer[16] reported the synthesis of neu-
tral, chiral, zirconium amidate complexes and their efficient
use for the preparation of various pyrrolidines (up to
93% ee). Simultaneously, Scott et al. published very similar
zirconium complexes and reported up to 91% ee.[17] Finally,
Hultzsch et al. reported the first example of a chiral main-
group metal as an efficient catalyst for the preparation of
pyrrolidines.[18]
A
dimeric, proline-derived, diamidobi-
naphthyl dilithium salt was synthesised and promoted the
cyclisation of various primary aminopentenes at room tem-
perature with up to 75% ee. The development of new lan-
thanide catalysts in the last four years has thus allowed im-
portant improvement in terms of activity and enantioselec-
tivity for obtaining some pyrrolidines with up to 95% ee.
There is nevertheless always the need for the preparation of
catalysts able to promote the cyclisation of more demanding
substrates. The hydroamination procedure is indeed known
to be considerably slowed down, if di- (or more-) substituted
double bonds are involved, relative to the corresponding un-
substituted substrates. Hence, the higher temperature re-
quired to achieve a convenient conversion is sometimes det-
rimental to the enantioselectivity of the reaction if asymmet-
ric processes are considered. To the best of our knowledge,
only the group of Marks has studied the asymmetric cyclisa-
tion of 2,2-dimethylhex-4-enylamine in the presence of
chiral metallocenes and afforded the corresponding pyrroli-
dine in up to 28% ee by performing the reaction at 808C.[19]
An analogous piperidine was interestingly isolated with
68% ee.
complexes from [Y{N
N
N
gands that catalysed the cyclisation of various aminopen-
tenes and an aminohexene at 608C, but with enantiomeric
excesses above 80% for several substrates. Interestingly the
cyclisation of 2,2-dimethyl-5-phenylpent-4-enylamine was
considered and performed with success together with the
transformation of a secondary amine, which occurred albeit
with a longer reaction time. For the hydroamination/cyclisa-
tion of such methylaminoalkenes, much more efficient cata-
lysts were prepared by Scott et al. as chiral zirconium–alkyl
cations.[11] Chiral thiophosphinic amides were also proven by
Livinghouse and co-workers to promote the cyclisation of
the test substrate (up to 61% ee at room temperature) in
the presence of a yttrium–trisamide precatalyst.[12] A chiral-
bridged alkylaminotroponiminate complex of lutetium was
reported by Roesky et al. as active catalyst for the prepara-
tion of the usual pyrrolidines and a piperidine.[13] However,
the reactions had to be performed at high temperature to
afford complete conversion in reasonable time and the en-
antiomeric excesses remained moderate (up to 44% ee for
the piperidine). The best results in terms of enantioselectivi-
ty were very recently described by Hultzsch and co-work-
ers.[14] The preparation of type 3,3’-bis(trisarylsilyl)-substitut-
ed binaphthol ligands led to corresponding rare-earth-metal
complexes (Sc, Lu and Y) as active and enantioselective cat-
alysts for the cyclisation of amino-olefins towards several
pyrrolidines and piperidines; 2-methyl-4,4-diphenylpyrroli-
dine was prepared with up to 95% ee. The more demanding
methylpent-4-enyl-amine, as a secondary amine, was cyclised
with up to 53% ee with a scandium catalyst. Zirconium cata-
lysts were further prepared aimed at promoting the intramo-
lecular hydroamination reaction. The group of Bergman
In such a context, we report here the preparation, charac-
terisation of new lanthanide amide ate complexes and their
use for promoting various hydroaminations/cyclisations with
different aminopentene and aminohexene derivatives.
Results and Discussion
We have previously reported the preparation and the char-
acterisation of a new family of lanthanide ate complexes
[Li(thf)4][Ln{(R)-C20H12(NR)2}2] (Ln=Nd (Li-1), Sm (Li-2),
C
Yb (Li-3), Lu (Li-4), Y (Li-5) R=CH2CMe3, CH2CHMe2,
CH2Ph, iPr, Cy, Ph) derived from chiral disubstituted (R)-bi-
naphthylamido ligands.[7,9,20,21] These compounds consist of a
complex anion resulting from coordination of two (R)-bi-
naphthylamine ligands to the lanthanide atom and a discrete
synthesised zirconium–bis
ACHTREUNG
combination of diphosphinic amides and Zr
AHCTREUNG
Those catalysts were active (at high temperature) for the
transformation of the usual substrates with up to 80% ee. A
methyl-2,3-dihydro-1H-indole derivative was also isolated
with high enantioselectivity (70% ee) under those condi-
counterion [Li
A
tramolecular hydroamination.
2190
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2189 – 2200

Asymmetric Catalysis
FULL PAPER
Both the influence of the nature of the lanthanide and of
the alkyl substituent on the nitrogen atoms were evaluated
on the cyclisation of various aminopentene derivatives.[9,20]
The ytterbium and lutetium complexes with bulky substitu-
ents on the nitrogen atom (isopropyl, cyclohexyl) led to the
highest enantioselectivities (up to 78% ee for a spiropyrroli-
dine), while ytterbium showed higher activities than luteti-
um complexes. We further prepared a similar yttrium ate
complex Li-5 (R=iPr) and the corresponding neutral com-
plex coordinated with the same chiral ligand, which led to
different results in terms of activity and/or enantioselectivity
for several hydroamination reactions, indicating that differ-
ent active species were involved. With the neutral, chiral
complex, higher rates of reaction and/or higher asymmetric
inductions were generally observed for the formation of sev-
eral pyrrolidines.[21] Yet, the synthesis of ate complexes is
much more straightforward and we maintained our efforts
towards more enantioselective and more active catalysts for
widening the scope of substrates suitable for hydroamina-
tion reactions.
tained by slow condensation of hexane into the concentrated
solutions of the complexes in THF at room temperature.
Both complexes crystallise in the monoclinic space group
P21 (Z=2). Complex Li-5b crystallises as a THF solvate
with half a molecule of THF per molecule of complex
[Li
A
(NC5H9)2}2]·0.5
A
Complex
Li-5b crystallises with two crystallographically independent
molecules in the asymmetric unit.
The molecular structures of complexes Li-3b and Li-5b
are depicted in Figure 1; the crystal and structural refine-
New lanthanide amide ate complexes are presented here
with a special emphasis on the influence of the structure of
the ligand, and the nature of the lanthanide and the alkaline
cation on the rates and asymmetric inductions of hydroami-
nation reactions. Particularly, optimised complexes have
been tested for the intramolecular hydroamination of more
demanding aminoalkenes.
Preparation and characterisation oflanthanide lithium ate
complexes:
The
ate
complexes
[Li(thf)4][Ln{(R)-
A
C20H12(NR)2}2] (R=C6H11, Ln=Yb (Li-3a); R=C5H9, Ln=
Yb (Li-3b), Y (Li-5b)) were synthesised by metathesis reac-
tions of anhydrous LnCl3 with two equivalents of Li2{(R)-
C20H12(NR)2} in THF at ambient temperature (Scheme 1).
The preparation of complex Li-3a was previously report-
ed.[9] Complexes Li-3b and Li-5b were purified by extrac-
tion of the solid residue after evaporation of THF with tolu-
ene and subsequent recrystallisation from THF/hexane mix-
tures. Complexes Li-3b and Li-5b were isolated in 52 and
53% yields, respectively. Both complexes were obtained as
highly air- and moisture-sensitive crystalline solids that are
soluble in THF and toluene and insoluble in hexane.
Clear single-crystal samples of complexes Li-3b (green)
and Li-5b (orange) suitable for X-ray investigation were ob-
Figure 1. Top: ORTEP drawing of complex [Li
(NC5H9)2}2] (Li-3b) with the atom-labelling scheme. Ellipsoids are drawn
at the 30% probability level. The [Li
(thf)4]+ ion and the hydrogen atoms
are omitted for the sake of clarity. Selected bond lengths [Š] and angles
A
AHCTREUNG
AHCTREUNG
À
À
À
À
[8]: Yb N12.258(7), Yb N2 2.268(9), Yb N3 2.245(8), Yb N4 2.252(7),
À
À
À
À
Yb C12.692(9), Yb C2 2.775(10), Yb C3 2.820(9), Yb C4 2.755(8),
À
À
Yb C5 2.669(10), Yb C6 2.808(9), N1-Yb-N2 121.8(3), N3-Yb-N4 120.1
(3). Bottom: ORTEP drawing of complex [Li(thf)4][Y{(R)-C20H12-
(NC5H9)2}2] (Li-5b) with the atom-labelling scheme. Ellipsoids are drawn
at the 30% probability level. The [Li
(thf)4]+ ion and the hydrogen atoms
are omitted for the sake of clarity. Selected bond lengths [Š] and angles
AHCTREUNG
AHCTREUNG
AHCTREUNG
À
À
À
À
À
[8]: Y1 N12.28(2), Y1 N2 2.35(3), Y1 N3 2.30(2), Y1 N4 2.26(2), Y1
À
À
À
À
C12.77(2), Y1 C2 2.88(2), Y1 C3 2.83(3), Y1 C4 2.56(3), Y1 C5
À
À
À
À
2.65(2), Y1 C6 2.79(2), Y1 C7 2.84(3), Y1 C8 2.74(3), Y2 N5 2.27(2),
Y2 N6 2.28(2), Y2 N7 2.269(19), Y2 N8 2.290(19), Y2 C9 2.78(2), Y2
C10 2.93(2), Y2 C11 2.82(2), Y2 C12 2.73(3), Y2 C13 2.83(2), Y2 C14
2.82(2), Y2 C15 2.80(2), Y2 C16 2.79(3), N1-Y1-N2 119.8(8), N3-Y2-N4
119.1(8), N7-Y2-N8 117.6(7), N5-Y2-N6 116.8(8).
À
À
À
À
À
À
À
À
À
À
À
Scheme 1. Synthesis of ytterbium and yttrium N-substituted-(R)-binaph-
thylamine lithium ate complexes.
Chem. Eur. J. 2008, 14, 2189 – 2200
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2191

J. Collin, E. Schulz et al.
ment data are listed in Table 1. X-ray single crystal structure
analysis has revealed that the compounds are isostructural
um chloride by extraction in toluene and recrystallised from
THF/hexane mixtures before their use in catalytic hydro-
ionic complexes, containing a discrete cation [Li
A
amination reactions. We observed then that higher activities
a discrete complex anion [Ln{(R)-C20H12
A
were obtained for the cyclisation of C-(1-allyl-cyclohexyl)-
methylamine if the catalysts were used without purifica-
tion.[9] For the following results the catalysts were prepared
from binaphthylamine in an improved procedure with short-
er reaction times for the formation of both the dilithium salt
(in hexanes) and the lanthanide ate complex (in THF). The
hydroamination reactions were initiated immediately after
the evaporation of the solvent from the lanthanide ate com-
plex containing an admixture of LiCl. Such a procedure was
applied to the synthesis of catalyst Li-3a, bearing (R)-N-cy-
clohexyl binaphthyl amine substituents. Tested for the cycli-
sation of C-(1-allylcyclohexyl)methylamine, it afforded the
pyrrolidine 7a (Scheme 2) in 24 h at room temperature with
ing from coordination of two diamido ligands to the triva-
lent lanthanide atom. The main geometric parameters of
complexes Li-3b and Li-5b are very similar to those previ-
ously reported for related complexes.[7,20,21] The average
À
Yb N bond length (2.255(8) Š) and N-Yb-N bond angle
(121.0(3)8) in complex Li-3b fall into the region observed
for
ytterbium
ate
complexes
[Li
ACHTREUNG
C20H12(NR)2}2]:
R=CH2CMe3,
2.261(2) Š;[7]
CH2CHMe2, 2.252(4) Š, 121.87(17)8;[20] R=iPr, 2.246(4) Š,
119.87(17)8.[21] For complex Li-5b the average Y N bond
À
length (2.29(2) Š) and N-Y-N bond angle (118.3 (8)8) are
close to the values found for yttrium ate complexes
[Li
C
R=iPr,
2.284(5) Š,
Catalytic hydroamination tests with lanthanide lithium ate
complexes: In our initial studies on the synthesis of lantha-
nide ate complexes for the catalysis of hydroamination reac-
tions we used a two-step procedure. The lithium binaphthyl-
amine salts were first isolated and in a second step they af-
forded the lanthanide ate complexes by reaction with lan-
thanide chlorides. The complexes were separated from lithi-
Scheme 2. Synthesis of nitrogen heterocycles by catalysis with lanthanide
lithium ate complexes.
Table 1. Crystallographic data and structure refinement details for Li-3b, Li-5b, and K-3d.
78% ee, an increase in enan-
tioselectivity with respect to
our previous results (65% ee).
With the aim to increase the
enantiomeric excess we decid-
ed to examine another ligand,
(R)-N-cyclopentylbinaphthyl-
amine, as a possible sterically
Li-3b
Li-5b
K-3d
formula
Mr
C76H92LiN4O4Yb
1305.52
C156H192Li2N8O9Y2
2514.88
C80 H108 K N4 O5 Yb
1417.84
T [K]
180(1)
180(1)
180(1)
crystal system
space group
a [Š]
b [Š]
c [Š]
monoclinic
P21
monoclinic
P21
orthorhombic
P212121
11.059(5)
17.209(5)
39.121(5)
90
7445(4)
4
1.265
1.363
10.6012(9)
17.6720(15)
18.9000(13)
99.830(2)
3488.8(5)
2
1.243
1.390
1362
17.360(3)
10.6930(18)
37.347(4)
91.330(3)
6930.8(18)
2
1.205
0.894
2680
more-demanding
chelating
b [8]
group on the lanthanide. We
synthesised this ligand accord-
ing to the method described in
our previous reports and the
corresponding ytterbium ate
complex Li-3b was prepared
by the above-mentioned pro-
cedure.[9] Gratifyingly this cat-
alyst promoted the cyclisation
of C-(1-allylcyclohexyl)methyl-
amine into the corresponding
pyrrolidine 7a within 20 h at
V [Š3]
Z
1calcd [mgmÀ3
]
m [mmÀ1
(000)
]
F
A
2980
crystal size [mm]
q range [8]
index ranges
0.22”0.16”0.07
1.59–26.45
À13ꢀhꢀ13
À22ꢀkꢀ22
À16ꢀlꢀ23
35486
0.11”0.07”0.03
1.58–20.17
À9ꢀhꢀ14
À4ꢀkꢀ9
À22ꢀlꢀ23
7946
0.25”0.10”0.10
1.91–24.21
À12ꢀhꢀ12
À19ꢀkꢀ19
À44ꢀlꢀ45
48104
reflns collected
independent reflns
13975 [Rint =0.0469]
98.6 (26.45)
Sadabs
0.751/0.911
0.070(15)
13975/0/775
1.084
R1=0.0671
wR2=0.0868
R1=0.1653
wR2=0.1752
0.809/À0.436
5960 [Rint =0.0280]
65.1(20.17)
Sadabs
0.928/0.974
0.18(2)
11546 [Rint =0.1730]
98.5 (24.2)
multiscan
0.8508/0.8391
0.01(2)
completeness [%] (to q [8])
absorption correction
max/min transmission
flack parameter
data/restraints/parameters
goodness-of-fit on F2
final R indices [I>2s(I)]
room
temperature
with
87% ee. The study of hydroa-
mination reactions catalysed
by Li-3b was thus widened to
several amino-olefins and the
results are given in Table 2.
5960/1580/1559
1.019
11546/1015/760
0.898
R1=0.0969
wR2=0.0978
R1=0.2480
wR2=0.2486
0.664/À0.742
R1=0.0626
wR2=0.1113
R1=0.1566
wR2=0.1456
1.361/À3.056
R indices (all data)
The influence of the temper-
ature on the cyclisation of
À3
largest diff. peak/hole [eŠ
]
2192
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2189 – 2200

Asymmetric Catalysis
FULL PAPER
Table 2. Enantioselective hydroamination reactions catalysed by ytterbi-
um and yttrium N-substituted-(R)-binaphthylamine lithium ate com-
plexes.
are indeed cyclised with higher asymmetric inductions. Fur-
thermore, the temperature marginally influences the enan-
tiomeric excess of hydroamination reactions. All pyrroli-
dines could be synthesised with enantiomeric excesses over
70%.
Till now, there are very few examples of enantioselective
reactions involving sterically hindered substrates such as
amines with disubstituted double bonds. For performing the
cyclisation of such amino-olefins, very active catalysts are
needed and we aimed at testing the performance of the lan-
thanide amide ate complexes for such reactions. In the
course of our previous studies concerning the comparison of
neutral and ate complexes, we prepared the yttrium ate
complex coordinated by (R)-N-isopropylbinaphthylamine
Cata-
lyst
Sub-
T
Cat. ratio
[mol%]
t
Conver-
ee
strate [8C][a,b]
G
[h]
sion [%][c] [%]
1Li- 3b 6a
0
25
60
25
60
0
25
25
60
40
60
80
6
6
6
8
6
6
6
4
4
8
4
6
168
20
15
144
17
168
18
84
15
144
40
15
92
90
92
14
94
100
100
30
100
100
100
100
85
87
78
72
69
78
76
76
73
50
51
50
2
3
4
5
6
7
8
9
Li-3b 6a
Li-3b 6a
Li-3b 6b
Li-3b 6b
Li-3b 6c
Li-3b 6c
Li-3b 6d
Li-3b 6d
10 Li-3b 6e
11 Li-3b 6e
12 Li-3b 6e
ligand [Li
N
N
plex catalysed the hydroamination of 6a within 20 h with
67% ee. When the ytterbium ate complex prepared with the
13 Li-5b 6a
14 Li-5b 6a
15 Li-5b 6b
16 Li-5b 6b
17 Li-5b 6c
18 Li-5b 6c
19 Li-5b 6e
0
25
25
60
0
6
8
6
6
6
6
6
144
15
180
22
96
5
54
93
49
96
94
75
81
64
68
75
75
41
same ligand [Li
N
N
used in this reaction, an identical enantiomeric excess was
obtained but six days were necessary for a complete conver-
sion, indicating the higher activity of the yttrium ate com-
plex relative to the corresponding ytterbium one. This result
prompted us to compare the activity and enantioselectivity
of the ytterbium complex Li-3b with that of the yttrium
complex Li-5b, coordinated by (R)-N-cyclopentylbinaph-
thylamine. The yttrium catalyst Li-5b afforded the three
pyrrolidines 7a, 7b and 7c with good enantioselectivities
(Table 2, entries 13–18). The comparison with the corre-
sponding ytterbium catalyst Li-3b showed that at the same
temperature the activity is higher with yttrium. At room
temperature the cyclisation of 6b is promoted with yttrium
(entry 15), while a lower conversion is only observed with
ytterbium (entry 4). Similarly, reaction times are shorter
with substrate 6c using catalyst Li-5b (entries 17, 18) rather
than Li-3b (entries 6, 7). A slight decrease of the enantio-
meric excess with the temperature was observed for pyrroli-
dine 7a using yttrium catalyst Li-5b, while no significant
variation was observed for pyrrolidines 7b and 7c. The hy-
droamination of aminohexene 6e was achieved in a shorter
time with yttrium catalyst Li-5b, but with a decrease in the
enantiomeric excess relative to Li-3b (entries 19 and 11).
We previously described the cyclisation of 6a in 1h, cata-
lysed by an ytterbium ate complex Li-3d coordinated with
(R)-N-neopentylbinaphthylamine.[7] Thus complex Li-3d
proved to be more active for the cyclisation of 6a than the
other lanthanide ate complexes studied so far and the evalu-
ation of this catalyst was extended to the preparation of
other pyrrolidines and piperidines. To check its stability,
complex Li-3d was tested in the hydroamination reaction of
6a at 608C, which afforded the expected product with
40% ee, as at room temperature (Table 2, entries 21and 22).
A total conversion of aminopentenes 6b and 6d in the cor-
responding pyrrolidines was observed at room temperature
with catalyst Li-3d (entries 23 and 25), while with catalysts
Li-3b or Li-5b, coordinated with a more bulky ligand, a
temperature of 608C was necessary. The piperidine 7e was
prepared at 608C with the three ytterbium and yttrium cata-
25
60
100
100
24
20 Li-3d 6a
21Li- 3d 6a
22 Li-3d 6a
23 Li-3d 6b
24 Li-3d 6c
25 Li-3d 6d
26 Li-3d 6e
27 Li-3d 6e
0
25
60
25
25
25
25
60
6
7
6
8
6
6
6
4
17
81
41
90
84
47
191
0.3
17
1.3 100
17
40
10
À17[d]
95
100
100
10
72
16
27
30
[a] Reactions performed at 08C in C7H8. [b] Reactions performed at
258C, 60 8C and 808C in C6D6. [c] Conversion was measured by H NMR
spectroscopy. [d] The absolute configuration of product 7c was opposite
to that obtained with catalysts Li-3b and Li-5b.
1
C-(1-allylcyclohexyl)methylamine (6a) by catalyst Li-3b was
first examined (Table 2, entries 1–3). At 08C hydroamina-
tion was performed with 85% ee, a value close to that ob-
tained at room temperature, but needed a long reaction
time to be completed, while at 608C a small decrease in the
enantiomeric excess was observed. The cyclisation of 2,2-di-
methylpent-4-enylamine (6b) was interestingly realised with
high conversion after one night at 608C and only a very
slight decrease in enantioselectivity compared to the reac-
tion at room temperature (entries 4, 5). With the gem-diphe-
nylaminopentene (6c) a total conversion was observed after
one night with a high value for the enantiomeric excess
(76%), which is not modified when the reaction is run at
lower temperature (entries 6,7). The best conditions for the
preparation of the spiropyrrolidine 7d are to perform the re-
action at 608C for a total conversion and 73% ee (entries 8,
9). The variation of the temperature did not either modify
the asymmetric induction for the transformation of the
aminohexene 6e and the piperidine is formed with complete
G
conversion and 50% ee after one night at 808C (entries 10–
12). These results are significantly improved relative to
those previously reported for the ytterbium ate complex
Li-3a.[9] With catalyst Li-3b the pyrrolidines and piperidine
Chem. Eur. J. 2008, 14, 2189 – 2200
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2193

J. Collin, E. Schulz et al.
lysts, but the reaction was faster with Li-3d (entries 11, 19
and 27). Interestingly, catalyst Li-3d proved to be active
enough for promoting this transformation at 258C (entry 26)
and the spiropiperidine was prepared with 27% ee and a
complete conversion within three days. However, all hydro-
condensation of hexane into a concentrated solution of K-
3d in THF at room temperature. It crystallises in the ortho-
rhombic space group P212121 with four molecules in the unit
cell. The molecular structure of complex K-3d is depicted in
Figure 2; the crystal and structural refinement data are
A
poor enantioselectivity. From the results gathered in Table 2
it appears that the ytterbium ate complex coordinated by
(R)-N-neopentylbinaphthylamine (Li-3d) showed the high-
est activity, while ytterbium ate complex Li-3b with the
bulkier (R)-N-cyclopentyl binaphthyl amine ligand afforded
the highest enantiomeric excesses. However, the (R)-N
cyclopentylbinaphthylamine yttrium ate complex Li-5b al-
lowed us to prepare the pyrrolidines and piperidine with
good enantiomeric excesses within short times.
Preparation and characterisation oflanthanide potassium
ate complexes: So far, our efforts towards the improvement
of the activity and enantioselectivity of lanthanide ate com-
plexes for hydroamination reactions have been directed to-
wards the choice of the lanthanide and the optimisation of
the size of the nitrogen substituent. The differences in activi-
ty and enantioselectivity of hydroamination reactions cata-
lysed by an yttrium ate complex and a neutral yttrium com-
plex coordinated by (R)-N-isopropylbinaphthylamine ligand
have indicated that probably both catalysts are not trans-
formed in the same active species.[21] This suggested a possi-
ble participation of the alkaline cation in the catalytic cycle.
Lanthanide potassium ate complexes have thus been syn-
thesised and studied as intramolecular hydroamination cata-
lysts.
Figure 2. ORTEP drawing of complex [K
(NCH2CMe3)2}2] (K-3d) with the atom-labelling scheme. Ellipsoids are
drawn at the 30% probability level. The [K
(thf)5]+ ion and the hydrogen
atoms are omitted for the sake of clarity. Selected bond lengths [Š] and
(thf)5][Yb{(R)-C20H12-
AHCTREUNG
AHCTREUNG
À
À
À
angles [8]: Yb1 N12.234(01) , Yb1
N2 2.148(13), Yb1 N3 2.175(13),
À
À
À
À
Yb1 N4 2.158(12), Yb1 C111 2.671(17), Yb1 C112 2.798(14), Yb1
À
À
À
C211 2.749(13), Yb1 C212 2.834(14), Yb1 C311 2.689(15), Yb1 C312
À
À
2.794(11), Yb1 C411 2.806(14), Yb1 C412 2.980(15), N1-Yb1-N2
120.9(4), N3-Yb1-N4 116.8(4).
listed in Table 1. X-ray single crystal structure analysis re-
veals that compound K-3d is an ionic complex, containing
The ate complexes containing the potassium counterion
discrete [K
A
[K
A
C20H12
ACHTREUNG
Y (K-5b); R=CH2CMe3, Ln=Yb (K-3d)) have been syn-
thesised by the reaction of anhydrous LnCl3 with two equiv-
alents of K2{(R)-C20H12(NR)2} in THF at ambient tempera-
ture. The potassium derivatives K2{(R)-C20H12(NR)2} were
prepared by treatment of binaphthylamine ligands with two
equivalents of Ph2CHK in THF (208C) (Scheme 3).
Complexes K-3b, K-5b and K-3d were purified following
the standard procedure described above for compounds Li-
3b and Li-5b. They were isolated in 43, 57 and 64% yields,
respectively, as highly air- and moisture-sensitive crystalline
solids. They are soluble in THF and toluene and insoluble in
hexane.
two diamido ligands to the trivalent lanthanide atom. Sur-
prisingly despite their identical nature and composition and
similar geometry of the anionic fragment [Yb{(R)-C20H12-
À
(NCH2CMe3)2}2]À in complexes K-3d and Li-3d[7] the Yb N
bonds in complex K-3d are noticeably shorter (the average
À
Yb N bond length: 2.178(13) Š) compared to those in Li-
[7]
À
3d (the average Yb N bond length: 2.261(2) Š). The
À
À
N Yb N bond angles in complex K-3d (120.9(4), 116.8(4)8)
are also smaller than the related angles in complex Li-3d
(120.92(10), 122.13(11)8).[7] The potassium cation in K-3d is
coordinated by five THF molecules, which is consistent with
the bigger ionic radius of potassium relative to that of lithi-
um.[22] Continuous drying of complex K-3d in vacuo at am-
bient temperature resulted in the loss of one THF molecule
that was proved by the results of microanalysis.
Clear deep-red single-crystalline samples of the complex
K-3d suitable for X-ray investigation were obtained by slow
Catalytic hydroamination tests with lanthanide potassium
ate complexes: The activity and enantioselectivity of the
three potassium ate complexes K-3b, K-5b and K-3d were
further examined for the hydroamination reactions leading
to the pyrrolidines and the piperidine described above and
the results are gathered in Table 3. With the ytterbium po-
tassium ate complex K-3b, coordinated by (R)-N-cyclopen-
tylbinaphthylamine, the formation of pyrrolidines 7a and 7c
Scheme 3. Synthesis of ytterbium and yttrium N-substituted-(R)-binaph-
thylamine potassium ate complexes.
2194
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2189 – 2200

Asymmetric Catalysis
FULL PAPER
Table 3. Enantioselective hydroamination reactions catalysed by ytterbi-
um and yttrium N-substituted-(R)-binaphthylamine potassium ate com-
plexes.
plexes, cyclisations in pyrrolidines and piperidines with cata-
lyst K-3d were more rapid than with ytterbium and yttrium
potassium K-3b and K-5b coordinated by a bulkier ligand :
reaction time was shorter for spiropyrrolidine 7a (see en-
tries 1, 5, 9), while the formation of pyrrolidine 7b and pi-
peridine 7e were realised at room temperature with K-3d.
Surprisingly with this potassium ate complex for almost all
pyrrolidines (except 7c) the major enantiomer is different
from that obtained using the similar ate complex Li-3d.
Only moderate enantiomeric excesses are found, as for the
corresponding lithium ate complex.
This inversion of the sense of asymmetric induction in the
case of (R)-N-neopentylbinaphthylamine ligand clearly indi-
cates that the active species in hydroamination reactions cat-
alysed by ate complex is not a neutral species and the alka-
line metal atom is somehow involved in the catalytic cycle.
We have previously proposed structures for the intermediate
catalytic species involved in the asymmetric hydroamination
promoted by lanthanide ate complexes.[21] Our new results
are in accordance with those hypotheses, but further studies
are still needed for more precise explanations of the role of
the second chiral ligand and of the associated alkaline
cation.
Cata-
lyst
Sub-
strate
T
Cat. ratio
[mol%]
t
Conver-
sion [%]
ee
[8C][a]
G
[h]
[%][b]
1K-
2
3
3b
K-3b
K-3b
K-3b
6a
6b
6c
6e
25
40
25
40
4
7
6
7
63
108
63
92
80
100
87
75
76
55
9
4
336
5
6
7
8
K-5b
K-5b
K-5b
K-5b
6a
6b
6c
6e
25
40
25
40
7
6
6
6
40
40
16
64
95
96
100
92
63
59
56
7
9
K-3d
3dK- 6b
K-3d
6a
6c
25
25
25
25
25
2
6
6
6
7
3
144
17
68
288
92
71
100
96
À45
À9
1
0
11
1
À31
À32
0
2
3
3dK- 6d
3dK- 6e
1
100
[a] Reactions performed in C6D6. [b] In entries 9–12, the negative sign in-
dicates opposite configurations to those obtained with K-3b and K-5b.
could be realised at room temperature with 75% and
55% ee, respectively (entries 1and 3). However, both activi-
ty and enantioselectivity were lower than those furnished by
the corresponding lithium ate complex Li-3b. Substrates 6b
and 6e could not be cyclised at room temperature with cata-
lyst K-3b and the hydroamination reactions had to be per-
formed at 408C (entries 2 and 4). At higher temperature the
potassium ate complexes are not stable and very low conver-
sions are obtained. The potassium ate complex catalyst
K-3b afforded the gem-dimethylpyrrolidine 7b with higher
enantiomeric excess (76%) than that obtained with the simi-
lar lithium ate complex. In contrast, the formation of the pi-
peridine was slow and the asymmetric induction very low;
these results might be explained by a lack of stability of the
catalyst at this temperature. The comparison of the ytterbi-
um and yttrium ate potassium complexes coordinated with
the same ligand, (R)-N-cyclopentylbinaphthylamine, gave
the same trend as that observed with the analogous lithium
ate complexes. For the cyclisation of substrates 6a (entries 1
and 5), 6b (entries 2 and 6), 6c (entries 3 and 7) and 6e (en-
tries 4 and 8) the reaction times were shorter for the yttrium
complex K-5b and the enantiomeric excesses lower for pyr-
rolidines 7a and 7b, while no variation was noticed for pyr-
rolidine 7c and piperidine 7e. The comparison of the yttri-
um based catalyst Li-5b and K-5b (Table 2, entry 14 and
Table 3 entry 5; Table 2, entry 18 and Table 3 entry 7) indi-
cated an increase in the reaction time and a decrease in en-
antiomeric excesses from lithium to potassium. For yttrium
and ytterbium ate complexes coordinated by (R)-N-cyclo-
pentylbinaphthylamine, a similar variation of the catalytic
properties for hydroamination reactions were observed
when lithium is replaced by potassium (comparison of Li-5b
with K-5b and Li-3b with K-3b). Next the ytterbium potas-
sium ate complex K-3d coordinated by (R)-N-neopentylbi-
naphthylamine was tested in various hydroamination reac-
tions (entries 9–13). As indicated above for lithium ate com-
Catalytic hydroamination tests with more demanding sub-
strates: To widen the scope of the utilisation of our new
chiral ate amide complexes, we have chosen to test the hy-
droamination/cyclisation of disubstituted amino-olefins
(Scheme 4). We were furthermore interested in the prepara-
Scheme 4. Asymmetric intramolecular hydroamination of more demand-
ing substrates.
tion of 2,3-dihydro-1H-indole as an important substrate
structure variation. To perform these challenging reactions,
the more active catalyst of our series, the ytterbium ate
complex Li-3d, was selected to expect convenient conver-
sion. The results obtained are summarised in Table 4.
The intramolecular hydroamination of 2-(1-but-2-enylcy-
clohexyl)ethylamine (6 f) was first examined at different
temperatures, and the use of toluene at 1108C yielded the
expected ethyl-substituted pyrrolidine with 90% conversion
(58% isolated yield) albeit with a four-day reaction time
(Table 4, entry 1). Product 7 f was obtained with 27% ee,
which is one of the best values reported to date for such a
cyclisation. Analogously, and under the same conditions, the
Chem. Eur. J. 2008, 14, 2189 – 2200
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2195

J. Collin, E. Schulz et al.
Table 4. Hydroamination/cyclisation of more demanding substrates with
catalyst Li-3d.
far the more active complex of this series, since 2,2-dime-
thylpent-4-enylamine (6b) and the aminohexene derivative
6e were efficiently transformed at room temperature. The
enantiofacial discrimination with this catalyst was unfortu-
nately diminished for all the substrates.
The role of the countercation was examined by the syn-
thesis of analogous complexes K-3b, K-5b and K-3d. These
complexes proved to be less stable than their lithium coun-
terparts and crystals suitable for X-ray analysis were only
obtained for K-3d. All these complexes promoted hydro-
Substrate T [8C][a,b] Cat. ratio t [h] Conversion [%] ee [%]
[mol%]
1
2
3
4
6 f
6g
6h
8
110
110
25
4
6
6
4
96
96
90
40
27
5
0
0.7 100
96 68
110
0
[a] Reactions performed at 1108C in C7H8. [b] Reactions performed at
258C in C6D6.
amination cyclisations, albeit in a less efficient way than the
analogous lithium ate complexes. Longer reaction times
were required for achieving satisfying conversions and the
enantiomeric excesses were generally lower. Moreover for
several substrates, hydroamination transformations catalysed
by K-3d afforded the products with a change of configura-
tion in contrast to Li-3d. A similar observation was made
for the hydroamination/cyclisation catalysed by organophos-
phine oxide substituted lanthanides that a small structural
variation in the catalytic precursor, such as a slight decrease
in the ionic radius, could promote an inversion in the sense
of enantioselectivity.[23] In a previous report, we initiated a
study of the mechanism of hydroamination reactions cata-
lysed by lanthanide ate complexes.[21] The decoordination of
one binaphthylamine ligand leading to a neutral trisamide
species was envisaged. The comparison of the efficiency of
an yttrium lithium ate and a neutral trisamide yttrium com-
plex coordinated by the same N-isopropylbinaphthylamine
ligand in various hydroamination reactions led to the con-
clusion that different active species were involved. This im-
plies that the total decoordination of one binaphthylamine
ligand in ate-catalysed reactions cannot be considered. The
different results provided herein by lithium and potassium
ate complexes similarly indicate that the active catalytic spe-
cies is not the same in both cases. We thus assume that both
ligands in ate complexes are in some way involved in the
stereodifferentiating course of the transformation. The exact
role of the counter cation is still unclear, but is certainly of
major importance since it interestingly allows in some cases
(compare in particular the catalytic results obtained by
Li-3d and K-3d) to obtain the cyclised products with the op-
posite configurations. In this context, Shibasaki demonstrat-
ed that for enantioselective carbon–carbon bond formation
promoted by heterobimetallic catalysts both lanthanide and
alkali metals were involved in the reaction mechanism.[24]
The most active catalyst of our series, complex Li-3d,
which posseses less bulky substituents on the nitrogen
atoms, was involved in the transformation of more demand-
ing substrates such as internal alkenes. Unhindered catalysts
are more effective for the cyclisation of such alkenes as was
already demonstrated with non-chiral lanthanide hydroami-
nation catalysts by Molander et al.[25] and chiral catalysts by
Marks et al.[19] To our delight the ate complex Li-3d indeed
gem-diphenyl substituted derivative led to 2-ethyl-4,4-diphe-
nylpyrrolidine (7g) with only a 40% conversion within four
days. This result is in contrast with those previously de-
scribed (Table 2) in which the cyclisation of the gem-diphen-
yl substrate always occurred faster than that of the corre-
sponding cyclohexyl derivative. Moreover, in this case the
ate ytterbium catalyst furnished a nearly racemic compound
(Table 4, entry 2). The transformation of 2,2,5-triphenylpent-
4-enylamine (6h) was also considered. In this case, however,
the presence of the phenyl substituent on the double bond
promoted the reaction, which could thus be smoothly per-
formed at room temperature. A complete conversion was
observed within 40 min (Table 4, entry 3) but 2-benzyl-4,4-
diphenylpyrrolidine was unfortunately obtained as a racemic
mixture. Finally, 2-allyl-phenylamine was submitted to the
hydroamination procedure as a good model for bicyclic ni-
trogen heterocycles. The cyclisation occurred slowly at
1108C, and a conversion of 68% was achieved after four
days (Table 4, entry 4). The neopentyl ytterbium ate catalyst
nevertheless did not induce any selectivity for this transfor-
mation.
Even if evident improvements are needed in terms of
enantioselectivity for achieving these delicate transforma-
tions, their feasibility was here unambiguously demonstrated
with lanthanide amide ate complexes.
Conclusion
We have presented the in-situ preparation of new lanthanide
amide ate complexes bearing chiral N-cyclopentyl-substitut-
ed binaphthylamine ligands. The ytterbium complex Li-3b
proved to be the most enantioselective catalyst we have
published so far and the spiropyrrolidine 7a was isolated
with 87% ee after 20 h reaction at room temperature. As al-
ready noted in the isopropyl-substituted series,[21] the yttri-
um catalyst proved to be more active than its ytterbium
counterpart for all the substrates tested. This enhanced ac-
tivity is unfortunately accompanied with a slight decrease in
the enantioselectivity. Both catalysts allowed the cyclisation
for the synthesis of a piperidine, albeit by heating the reac-
tion mixture, and 3-methyl-2-azaspiro
G
allowed the synthesis of 3-ethyl-2-azaspiro[4,5]decane with
N
was isolated with up to 51% ee.
27% ee. The synthesis of this type of product has been
Complex Li-3d, an amide ate complex bearing the less
seldom described by intramolecular asymmetric hydro-
hindered neopentyl substituent at the nitrogen atom, was by
A
2196
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2189 – 2200

Asymmetric Catalysis
FULL PAPER
Preparation ofthe bislithium salt of( R)-(+)-2,2’-bis(cyclopentylamino)-
1,1’-binaphthyl: In an argon-filled glove box, (R)-(+)-2,2’-bis(cyclopentyl-
amino)-1,1’-binaphthyl (50.0 mg, 0.12 mmol) was solubilised in hexanes
(2 mL) in a Schlenk flask equipped with a magnetic stirring bar. nBuLi
(1.6m in hexanes, 0.15 mL, 0.24 mmol) was introduced by a micro syringe
and the reaction mixture was stirred for 10 min to give a yellow suspen-
sion. The solvent was evaporated in vacuo to afford the corresponding
for pyrrolidines resulting from aminoalkenes with internal
double bonds.[19]
Research is presently ongoing for the preparation of more
active catalysts for widening the scope of substrates and syn-
thesising structurally different nitrogen heterocycles by the
atom-efficient hydroamination transformation.
lithium amide salt as a yellow powder (40.0 mg, 0.09 mmol, 79% yield).
3
1H NMR (360 MHz, [D8]THF, 258C, TMS): d=7.47 (d, J
A
2H; 2CH Ar), 7.35 (d, 3J
(H,H)=9.0 Hz, 2H; 2CH Ar), 6.65 (dd, 3J
6.7 Hz, 2H; 2CH Ar), 6.56 (dd, 3J(H,H)=6.7 Hz, 3J
2CH Ar), 6.41(d, 3J
(H,H)=8.5 Hz, 2H; 2CH Ar), 3.84–3.90 (m, 2H;
ACHTREUNG
A
G
ACHTREUNG
T
ACHTREUNG
Experimental Section
AHCTREUNG
2CH), 2.09–2.14 (m, 2H; 2CH2), 1.99–2.03 (m, 2H; 2CH2), 1.61–1.63 (m,
8H; 4CH2), 1.22–1.30 ppm (m, 4H; 2CH2); 13C NMR (90 MHz,
[D8]THF, 258C, TMS): d=156.89 (2C Ar), 137.36 (2C Ar), 128.07 (2CH
Ar), 126.73 (2CH Ar), 124.47 (2CH Ar), 124.01 (2C Ar), 123.83 (2CH
Ar), 116.32 (2CH Ar), 115.81 (2CH Ar), 112.13 (2C Ar), 59.41 (2CH),
35.93 (2CH2), 34.29 (2CH2), 24.10 ppm (4CH2); IR (KBr, Nujol): n˜ =
General considerations: All manipulations were carried out under an
argon atmosphere by using standard Schlenk or glove box techniques.
THF and diethyl ether were distilled from sodium benzophenone ketyl
and degassed immediately prior to use. Hexane and toluene were dis-
tilled from CaH2 and degassed immediately prior to use. Deuterated ben-
zene was dried with sodium benzophenone ketyl and was transferred
under vacuum. Anhydrous YbCl3 and YCl3 and (R)-(+)-1,1’-binaphthyl-
1611, 1592, 1497, 1421, 1337, 1287, 1242, 812, 748 cmÀ1
Preparation ofcomplex [Li (thf)4][Yb{(R)-C20H12N2(C10H18)}2] (Li-3b): In
.
A
ACHTREUNG
2,2’-diamine were purchased. Compounds Y[N
(tms)2]3,[26] (R)-(+)-2,2’-
A
an argon-filled glove box, YbCl3 (50.3 mg, 0.18 mmol) was added to a so-
lution of the bislithium salt of (R)-(+)-2,2’-bis(cyclopentylamino)-1,1’-bi-
naphthyl (155.6 mg, 0.36 mmol) in THF (2 mL). The reaction mixture
was stirred at room temperature for 45 min and THF was evaporated in
vacuo. The resulting solid was extracted with toluene (5 mL), the solution
was centrifuged and the filtrate was concentrated in vacuo. The product
was obtained as a green powder (117.5 mg, 0.09 mmol, 52% yield). The
powder was dissolved in THF (1mL) and a slow condensation of hexanes
in the THF solution resulted in green crystals, suitable for X-ray analysis.
IR (KBr, Nujol): n˜ =1609, 1591, 1539, 1495, 1420, 1333, 1247, 1209, 1043,
885, 809, 743 cmÀ1; elemental analysis calcd (%) for C76H92LiN4O4Yb: C
69.92, H 7.10; found: C 69.74, H 7.10.
Bis(cyclopentylamino)-1,1’-binaphthyl,[27] and Ph2CHK[28] were prepared
according to reported procedures. The bis(N-neopentyl)binaphthylamine
ligand and the complex Li-3d [Li
N
N
prepared according to the procedure we previously described.[20] The sub-
strates 6a–h for the hydroamination/cyclisation reactions were prepared
as previously reported.[9] Substrate 8 was synthesised according to the
procedure described by Marks.[29] All other commercially available chem-
icals were used after the appropriate purification. Bruker AM250,
Bruker AV300 and AV360 NMR spectrometers, (operating at 250, 300
and 360 MHz, respectively) were used for recording the NMR spectra.
1
Chemical shifts for H and 13C spectra were referenced internally accord-
ing to the residual solvent resonances and reported relative to tetrame-
thylsilane. Infrared spectra were recorded on a Perkin–Elmer FT-IR
spectrometer as Nujol mulls and are reported in cmÀ1. Optical rotations
are reported as follows: [a]RDT (c in g per 100 mL in solvent). Elemental
analyses were performed by the Microanalytical Laboratory of the Insti-
tute of Organometallic Chemistry of RAS. Enantiomeric excesses of the
products have been determined by GC (GC Fisons 800, column DB1
30 m”0.32 mm”0.5 mm) or HPLC (Thermo Separation Product Spectra
Series tsp 100P100/UV100 or Perkin–Elmer Pump Series 200/DAD200)
analyses after derivatisation, and compared to racemic products prepared
Preparation of[Li
G
N
filled glove box, YCl3 (58.1mg, 0.30 mmol) was added to a solution of
the bislithium salt of (R)-(+)-2,2’-bis(cyclopentylamino)-1,1’-binaphthyl
(259.2 mg, 0.60 mmol) in THF (5 mL). The reaction mixture was stirred
at room temperature for 45 min and THF was evaporated in vacuo. The
resulting solid was extracted with toluene (5 mL), the solution was centri-
fuged and the filtrate was concentrated in vacuo. The product was ob-
tained as an orange powder (195.4 mg, 0.16 mmol, 53% yield). The
powder was dissolved in THF (1mL) and a slow condensation of hexanes
in the THF solution resulted in orange crystals, suitable for X-ray analy-
with Y[N(TMS)2]3. The methods are detailed for each compound below.
A
sis. 1H NMR (250 MHz, C6D6, 258C, TMS): d=7.91(d, 3J
(H,H)=8.8 Hz,
N
Preparation of( R)-(+)-2,2’-bis(cyclopentylamino)-1,1’-binaphthyl: An
aqueous solution of H2SO4 (20%, 8 mL) was added to a solution of cy-
clopentanone (4.1g, 49.24 mmol) in THF (15 mL). A solution of ( R)-bi-
naphthylamine (1.0 g, 3.52 mmol) in THF (30 mL) was then slowly added
by syringe. After one hour at room temperature, NaBH4 (1.9 g,
49.24 mmol) was added in portions at 08C. The reaction mixture was al-
lowed to warm up to room temperature overnight. It was then poured
into an aqueous solution of KOH (2%, 320 mL) and the aqueous phase
was extracted with diethyl ether (3”300 mL). The combined organic
layers were dried (MgSO4), filtered and concentrated in vacuo. The
crude product was purified by flash chromatography on silica gel (hep-
tanes/ethyl acetate=9/1) to give a white powder (1.5 g, 3.46 mmol, 98%
yield). [a]2D0 =+96.9 (c=0.84 in chloroform); 1H NMR (250 MHz, C6D6,
2H; 2CH Ar), 7.70–7.73 (m, 4H; 4CH Ar), 7.54–7.56 (m, 2H; 2CH Ar),
7.37–7.41 (m, 4H; 4CH Ar), 6.82–7.03 (m, 12H; 12CH Ar), 3.75–3.82
(m, 4H; 4CH), 3.33 (brs, 16H; 8aCH2-THF), 2.41–2.57 (m, 2H; CH2),
2.04–2.14 (m, 2H; CH2), 1.58–1.69 (m, 2H; CH2), 1.39–1.46 (m, 8H;
4CH2), 1.28 (brs, 16H; 8bCH2-THF), 1.12–1.21 (m, 8H; 4CH2), 0.87–
0.92 (m, 2H; CH2), 0.59–0.65 ppm (m, 8H, 4CH2); 13C NMR (62.5 MHz,
C6D6, 258C, TMS): d=150.87 (4C Ar), 135.60 (4C Ar), 131.74 (4CH
Ar), 131.27 (4CH Ar), 127.31 (4C Ar), 126.97 (4CH Ar), 124.65 (4CH
Ar), 120.44 (4CH Ar), 119.17 (4CH Ar), 116.51 (4C Ar), 68.03 (aCH2-
THF), 58.95 (2CH), 58.41(2CH), 35.65 (2CH 2), 35.09 (2CH2), 34.93
(2CH2), 34.33 (2CH2), 25.55 (bCH2-THF), 24.86 (2CH2), 24.40 (2CH2),
23.86 (2CH2), 23.60 ppm (2CH2); IR (KBr, Nujol): n˜ =1608, 1590, 1537,
1488, 1419, 1340, 1284, 1245, 1207, 1170, 1041, 953, 913, 883, 850, 808,
739 cmÀ1; elemental analysis calcd (%) for C76H92LiN4O4Y: C 74.73, H
7.59, Y 7.28; found: C 75.03, H 7.62, Y 7.57.
258C, TMS): d=7.84 (d, 3J
7.6 Hz, 2H; CH Ar), 7.28 (d, 3J
(H,H)=8.8 Hz, 2H; CH Ar), 7.02–7.12 (m, 4H; CH Ar), 3.81 (d, 3J-
(H,H)=8.2 Hz, 2H; NH), 3.65–3.75 (m, 2H; CH), 1.49–1.71 (m, 4H;
ACHTREUNG
AHCTREUNG
ACHTREUNG
Preparation of[K
A
E
ACHTREUNG
CH2), 1.15–1.25 (m, 8H; CH2), 0.92–1.12 ppm (m, 4H; CH2); 13C NMR
(62.5 MHz, C6D6, 258C, TMS): d=144.93 (2C Ar), 134.68 (2C Ar),
129.96 (2CH Ar), 128.62 (2CH Ar), 128.26 (2C Ar), 127.21 (2CH Ar),
124.46 (2CH Ar), 122.36 (2CH Ar), 115.26 (2CH Ar), 112.82 (2C Ar),
55.02 (2CH), 33.71(2CH 2), 33.63 (2CH2), 23.88 ppm (4CH2); IR (KBr):
n˜ =3400, 3049, 2954, 2867, 1616, 1595, 1510, 1492, 1422, 818, 807,
argon-filled glove box, (R)-(+)-2,2’-bis(cyclopentylamino)-1,1’-binaphthyl
(302.4 mg, 0.72 mmol) was solubilised in THF (5 mL) in a Schlenk flask
equipped with a magnetic stirring bar. Ph2CHK (296.6 mg, 1.44 mmol) in
THF (1mL) was added slowly and the reaction mixture was stirred for
10 min, forming the bispotassium salt. YbCl3 (99.7 mg, 0.36 mmol) was
then added to this solution. The reaction mixture was stirred at room
temperature for 45 min and THF was evaporated in vacuo. The resulting
solid was extracted with toluene (5 mL), the solution was centrifuged and
746 cmÀ1
443.2458.
; HRMS (ESI): calcd for C30H32N2Na: 443.2463; found:
Chem. Eur. J. 2008, 14, 2189 – 2200
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2197

J. Collin, E. Schulz et al.
the filtrate was concentrated in vacuo. The product was obtained as a
reddish brown powder (215.0 mg, 0.15 mmol, 43% yield). The powder
was dissolved in THF (1mL) and a slow condensation of hexanes in the
solution resulted in reddish brown crystals, not suitable for X-ray analy-
sis. IR (KBr, Nujol): n˜ =1616, 1595, 1507, 1491, 1421, 1349, 1327, 1248,
1210, 1185, 1050, 1019, 857, 808, 745 cmÀ1; elemental analysis calcd (%)
for C76H92KN4O4Yb: C 68.24, H 6.93, Yb 12.94; found: C 68.48, H 6.94,
Yb 13.00.
tered and concentrated. The residue was distilled (1008C, 0.4 mbar) to
afford a colourless liquid (6.04 g, 36.17 mmol, 53% yield). The product is
a mixture of the cis/trans isomers (10:90). 1H NMR (250 MHz, CDCl3,
258C, TMS) trans isomer: d=5.30–5.36 (m, 2H; 2CH), 2.39 (s, 2H;
CH2), 1.87 (d, 3J
A
N
CH3), 1.31–1.34 (m, 6H; 3CH2), 1.15–1.22 (m, 4H; 2CH2), 0.77–0.82 ppm
(brs, 2H; NH2); 13C NMR (62.5 MHz, C6D6, 258C, TMS): d=127.25
(CH), 127.15 (CH), 48.89 (CH2), 38.52 (CH2), 37.10 (C), 33.31 (2CH2),
26.50 (CH2), 21.58 (2CH2), 18.09 ppm (CH3); IR (CaF2): n˜ =3391, 3308,
3019, 2923, 2853, 1615, 1454, 1377, 968 cmÀ1; HRMS (ESI): calcd for
C11H22N: 168.1752; found: 168.1747.
Preparation ofcomplex [K
N
N
an argon-filled glove box, YCl3 (69.7 mg, 0.36 mmol) was added slowly to
a solution of the bispotassium salt of (R)-(+)-2,2’-bis(cyclopentylamino)-
1,1’-binaphthyl in THF, prepared as described above for K-3b. The reac-
tion mixture was stirred at room temperature for 45 min and THF was
evaporated in vacuo. The resulting solid was extracted with toluene
(5 mL), the solution was centrifuged and the filtrate was concentrated in
vacuo. The product was obtained as an orange powder (256.0 mg,
0.20 mmol, 57% yield). The powder was dissolved in THF (1mL) and a
slow condensation of hexanes into the solution resulted in orange crystals
Catalytic tests
General procedure for NMR-scale racemic hydroamination–cyclisation:
For the cyclisations of aminoalkenes performed from 258C to 808C, in
the glove box, the appropriate aminoalkene (0.20 mmol) was dissolved in
C6D6 (0.1mL) and dried on 4 Š molecular sieves for 2h at room temper-
ature. [Y{N(SiMe3)2}3] (11.2 mg, 0.02 mmol) was dissolved in C6D6
G
(0.7 mL) and introduced into a J. Young NMR tube equipped with a
teflon valve, and the aminoalkene solution was then introduced. The
NMR tube was heated out of the glove box at the appropriate tempera-
ture. The hydroamination reaction was monitored by 1H NMR spectros-
copy by observation of the decrease of the olefinic protons signals. After
the appropriate time, the reaction was quenched with CH2Cl2.
1
3
not suitable for X-ray analysis. H NMR (250 MHz, C6D6): d=7.84 (d, J-
(H,H)=8.8 Hz, 2H; 2CH Ar), 7.76 (d, 3J
(H,H)=7.0 Hz, 2H; 2CH Ar),
7.28 (d, 3J(H,H)=7.0 Hz, 2H; 2CH Ar), 7.23 (d, 3J
(H,H)=8.8 Hz, 2H;
A
ACHTREUNG
A
ACHTREUNG
2CH Ar), 7.02–7.12 (m, 16H; 16CH Ar), 3.70–3.82 (m, 4H; 4CH), 3.55–
3.60 (m, 16H; 8aCH2-THF), 1.47–1.70 (m, 8H; 4CH2), 1.39–1.44 (m,
16H; 8bCH2-THF), 1.30–1.36 (m, 4H; 2CH2), 1.17–1.23 (m, 8H; 4CH2),
0.89–1.13 ppm (m, 12H, 6CH2); 13C NMR (90 MHz, C6D6, 258C, TMS):
d=138.43 (4C Ar), 134.69 (4C Ar), 129.95 (2CH Ar), 129.30 (2CH Ar),
128.73 (2CH Ar), 128.61 (2CH Ar), 128.35 (4CH Ar), 127.23 (2CH Ar),
127.08 (4C Ar), 126.33 (2CH Ar), 124.46 (2CH Ar), 122.35 (2CH Ar),
117.52 (4C Ar), 115.25 (4CH Ar), 67.82 (aCH2-THF), 55.01(4CH),
42.19 (4CH2), 33.71(4CH 2), 33.61(4CH 2), 25.81( bCH2-THF),
23.87 ppm (4CH2); IR (KBr, Nujol): n˜ =1605, 1591, 1538, 1493, 1419,
1367, 1350, 1327, 1291, 1247, 1052, 812, 742 cmÀ1; elemental analysis
calcd (%) for C76H92KN4O4Y: C 72.82, H 7.40; found: C 72.26, H 7.00.
For the cyclisations of aminoalkenes performed at 1108C the appropriate
aminoalkene (0.20 mmol) was dissolved in toluene (0.1mL) in the glove
box and dried on 4 Š molecular sieves for 2h at room temperature. This
solution was added to [Y{N(SiMe3)2}3] (11.2 mg, 0.02 mmol) in solution in
H
toluene (0.7 mL) in a sealed tube which was sealed and heated at 1108C
for the appropriate time. The reaction was quenched with CH2Cl2.
Typical procedure for the in-situ preparation of the Li ate complexes Li-
3a, Li-3b, Li-3d and Li-5b: In an argon-filled glove box, the ligand
(0.12 mmol) was solubilised in hexanes (2 mL) in
a Schlenk flask
equipped with a magnetic stirring bar. nBuLi (1.6m in hexanes, 0.15 mL,
0.24 mmol) was introduced through a microsyringe and the reaction mix-
ture was stirred for 10 min (30 min for Li-3d) to give a yellowish suspen-
sion. The solvent was evaporated in vacuo. LnCl3 (0.06 mmol) was added
to the solution of the bislithium salt of the ligand in THF (2 mL). The re-
action mixture was stirred at room temperature for 45 min (3 h for Li-
3d) and THF was evaporated in vacuo. The crude mixture was directly
used for the catalytic tests.
Preparation of[K
N
A
argon-filled glove box, (R)-(+)-2,2’-bis(neopentylamino)-1,1’-binaphthyl
(200.0 mg, 0.47 mmol) was solubilised in THF (3 mL) in a Schlenk flask
equipped with a magnetic stirring bar. Ph2CHK (195.3 mg, 0.95 mmol) in
THF (1mL) was added slowly and the reaction mixture was stirred for
2h forming the bispotassium salt. YbCl3 (66.2 mg, 0.24 mmol) was then
slowly added to this solution, the reaction mixture was stirred at room
temperature overnight and THF was evaporated in vacuo. The resulting
solid was extracted with toluene (5 mL), the solution was centrifuged and
the filtrate was concentrated in vacuo. The product was obtained as a
reddish brown powder (206.2 mg, 0.15 mmol, 64% yield). The powder
was dissolved in THF (1mL) and a slow condensation of hexanes in the
solution resulted in reddish brown crystals, suitable for X-ray analysis. IR
(KBr, Nujol): n˜ =1607, 1590, 1538, 1496, 1421, 1364, 1336, 1317, 1283,
1244, 1210, 1094, 1055, 896, 810, 775, 744; elemental analysis calcd (%)
for C76H88KN4O4Yb: C 67.83, H 7.49, Yb 12.86; found: C 68.02, H 8.82,
Yb 13.16.
Typical procedure for the preparation of the K ate complexes K-3b, K-3d
and K-5b: In an argon-filled glove box, the ligand (0.12 mmol) was solu-
bilised in THF (2 mL) in a Schlenk flask equipped with a magnetic stir-
ring bar. Ph2CHK (49.0 mg, 0.24 mmol) in THF (0.5 mL) was added
slowly and the reaction mixture was stirred for 10 min (2 h for K-3d).
LnCl3 (0.06 mmol) was then slowly added to the solution of the bis potas-
sium salt of the ligand in THF. The reaction mixture was stirred at room
temperature for 45 min (overnight for K-3d) and THF was evaporated in
vacuo. The crude mixture was directly used for the catalytic tests.
General procedure for NMR-scale asymmetric hydroamination–cyclisation
of aminoalkenes: For the cyclisation of aminoalkenes performed from
258C to 808C the appropriate aminoalkene (0.20 mmol) was dissolved in
C6D6 (0.1mL) in the glove box and dried on 4 Š molecular sieves for 2 h
at room temperature. The lanthanide catalyst was dissolved in C6D6
(0.7 mL) and introduced into a J. Young NMR tube equipped with a
teflon valve and the aminoalkene solution was then introduced. The
NMR tube was heated out of the glove box at the appropriate tempera-
ture. The hydroamination reaction was monitored by 1H NMR spectros-
copy by observation of the decrease of the olefinic protons signals. After
the appropriate time, the reaction was quenched with CH2Cl2.
Synthesis of C-(1-but-2-enylcyclohexyl)methylamine (6 f): n-Buli 2.5m
(32.80 mL, 82.00 mmol) was added dropwise to a solution of diisopropyl-
amine (12.00 mL, 85.38 mmol) in THF (70 mL) at 08C and the reaction
mixture was stirred for 2 h.
A solution of cyclohexane carbonitrile
(8.2 mL, 68.32 mmol) in THF (15 mL) was added dropwise at 08C and
the reaction mixture was stirred for three additional hours. A solution of
but-2-enyl bromide (9.6 mL, 78.75 mmol) in THF (15 mL) was added
dropwise at 08C and the reaction mixture was allowed to warm up to
room temperature overnight. It was hydrolysed with water (100 mL) and
the aqueous phase extracted with diethyl ether (3”100 mL). The com-
bined organic layers were dried (MgSO4), filtered and concentrated in
vacuo. The crude product was not isolated but used in the following step
without purification. It was solubilised in diethyl ether (30 mL) and
added dropwise to a suspension of LiAlH4 (2.60 g, 68.32 mmol) in diethyl
ether (120 mL) at 08C. The reaction mixture was allowed to warm up to
room temperature overnight and then hydrolysed with water until the
formation of white hydroxide aluminium salts. The solid was separated
from the mixture by filtration. The organic layer was dried (MgSO4), fil-
For the cyclisation of aminoalkenes performed at 08C the appropriate
aminoalkene (0.20 mmol) was dissolved in toluene (0.1mL) in the glove
box and dried on 4 Š molecular sieves for 2 h at 08C. The lanthanide cat-
alyst was dissolved in toluene (0.5 mL) and cooled at 08C for 30 min.
The aminoalkene solution was added at 08C to the catalyst solution and
maintained at 08C during the appropriate time. The reaction mixture was
quenched with CH2Cl2.
2198
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2189 – 2200

Asymmetric Catalysis
FULL PAPER
For the cyclisation of aminoalkenes performed at 1108C the appropriate
aminoalkene (0.20 mmol) was dissolved in toluene (0.1mL) in the glove
box and dried on 4 Š molecular sieves for 2 h at room temperature. The
lanthanide catalyst was dissolved in toluene (0.7 mL) and introduced into
an ampoule which was sealed and heated at 1108C during the appropri-
ate time. The reaction was quenched with CH2Cl2.
parameters for all non-hydrogen atoms. Hydrogen atoms were located on
a difference Fourier map and introduced into the calculations as a riding
model with isotropic thermal parameters. All calculations were per-
formed by using the Crystal Structure crystallographic software package
WINGX.[33] The absolute configuration was determined by refining the
Flackꢁs[34] parameter using a large number of Friedelꢁs pairs.
3-Ethyl-2-azaspiro[4,5]decane (7 f): Colourless oil; b.p. 1008C, 0.4 mbar;
G
[a]2D0 =+3.6 (c=0.12 in chloroform) for 27% ee, 58% isolated yield;
1H NMR (400 MHz, C6D6, 258C, TMS): d=3.17–3.28 (m, 1H; CH), 3.07
(d, 3J
A
ACHTREUNG
ACHTREUNG
Acknowledgements
(100 MHz, C6D6, 258C, TMS): d=60.44 (CH), 42.54 (C), 37.62 (2CH2),
36.37 (CH2), 27.18 (CH2), 26.07 (CH2), 24.06 (CH2), 23.46 (2CH2),
11.78 ppm (CH3); IR (CaF2): n˜ =2926, 2854, 1623, 1450 cmÀ1; HRMS
(ESI): m/z calcd for C18H22N: 252.1747; found: 252.1749.
MENSR is acknowledged for financial support and CNRS for giving fi-
nancial support and a doctoral grant (together with the Conseil GØnØral
de lꢁEssonne) for I.A. We thank the UniversitØ Paris Sud XI for an invit-
ed professor chair for A.T. (2005). We are indebted to Dr. J.-C. Daran
for the RX analysis of compound K-3d. A.T. and D.L. thank the Russian
Foundation for Basic Research (Grant: 05-03-32390), the Russian Foun-
dation for Science Support, and the Federal Agency for Science and In-
novations (Contract 02.513.11.3029).
2-Ethyl-4,4-diphenylpyrrolidine (7g): Colourless oil; 1H NMR (250 MHz,
C6D6, 258C, TMS): d=7.06–7.28 (m, 10H; 10CH Ar), 3.56 (d, 3J
ACHTREUNG
10.7 Hz, 1H; CH2), 3.33 (d, 3J
ACHTREUNG
AHCTREUNG
3H; CH3); 13C NMR (62.5 MHz, CDCl3, 258C, TMS): d=148.70 (C Ar),
147.92 (C Ar), 128.53 (4CH Ar), 127.59 (4CH Ar), 126.12 (2CH Ar),
59.50 (CH), 58.26 (CH2), 56.93 (C), 45.51(CH 2), 30.62 (CH2), 11.79 ppm
(CH3); IR (CHCl3): n˜ =2918, 2877, 2710, 2553, 1601, 1493, 1456,
[1] a) S. Hong, T. J. Marks, Acc. Chem. Res. 2004, 37, 673–686; b) K. C.
Hultzsch, Adv. Synth. Catal. 2005, 347, 367–391; c) P. W. Roesky,
T. E. Müller, Angew. Chem. 2003, 115, 2812–2815; Angew. Chem.
Int. Ed. 2003, 42, 2708–2710; d) I. Aillaud, J. Collin, J. Hannedou-
che, E. Schulz, Dalton Trans. 2007, 5105–5118.
[2] a) M. R. GagnØ, L. Brard, V. P. Conticello, M. A. Giardello, C. L.
Stern, T. J. Marks, Organometallics 1992, 11, 2003–2005; b) M. A.
Giardello, V. P. Conticello, L. Brard, M. R. GagnØ, T. J. Marks, J.
Am. Chem. Soc. 1994, 116, 10241–10254.
1380 cmÀ1
167.1671.
; HRMS (ESI): m/z calcd for C11H21N: 167.1669; found:
Determination of enantiomeric excesses for the amine 7 f, 7g, 7h and 9
3-Ethyl-2-azaspiro[4,5]decane (7 f): Dimethylaminopyridine (6.0 mg,
0.04 mmol), triethylamine (27.5 mL, 0.20 mmol) and Mosher chloride
(37.0 mL, 0.20 mmol) were added to solution of pyrrolidine 7 f
ACHTREUNG
a
(0.20 mmol) in CH2Cl2 (5 mL). The reaction mixture was stirred at room
temperature for 2 h and then hydrolysed with water (5 mL). The aqueous
phase was extracted with CH2Cl2 (3”5 mL). After drying, the crude
product was dissolved in diethyl ether and injected on a GC capillary
column DB-1(80 8C (5 min) then 108CminÀ1 until 2008C (20 min),
[3] M. R. Douglass, M. Ogasawara, S. Hong, M. V. Metz, T. J. Marks,
Organometallics, 2002, 21, 283–292.
[4] P. N. OꢁShaughnessy, P. Scott, Tetrahedron: Asymmetry 2003, 14,
1979–1983.
[5] P. N. OꢁShaughnessy, P. D. Knight, C. Morton, K. M. Gillespie, P.
Scott, Chem. Commun. 2003, 1770–1771.
t
1(major) =28.86 min, t2(minor) =29.57 min).
2-Ethyl-4,4-diphenylpyrrolidine (7g): Dimethylaminopyridine (6.0 mg,
0.04 mmol), triethylamine (27.5 mL, 0.20 mmol) and naphthoyl chloride
[6] D. V. Gribkov, K. C. Hultzsch, F. Hampel, Chem. Eur. J. 2003, 9,
4796–4810.
[7] J. Collin, J. -C. Daran, E. Schulz, A. Trifonov, Chem. Commun.
2003, 3048–3049.
[8] S. Hong, S. Tian, M. Metz, T. J. Marks, J. Am. Chem. Soc. 2003, 125,
14768–14783.
[9] D. Riegert, J. Collin, A. Meddour, E. Schulz, A. Trifonov, J. Org.
Chem. 2006, 71, 2514–2517.
[10] J. Y. Kim, T. Livinghouse, Org. Lett. 2005, 7, 1737–1739.
[11] P. D. Knight, I. Munslow, P. N. OꢁShaughnessy, P. Scott, Chem.
Commun. 2004, 894–895.
(29.5 mL, 0.20 mmol) were added to
a solution of pyrrolidine 7g,
(49.2 mg, 0.20 mmol) in CH2Cl2 (5 mL). The reaction mixture was stirred
at room temperature for 2 h and then hydrolysed with water (5 mL). The
aqueous phase was extracted with CH2Cl2 (3”5 mL). The combined or-
ganic layers were washed with a saturated solution of NH4Cl, dried
(MgSO4), filtered and concentrated in vacuo. The crude product was pu-
rified by preparative silica plate (cyclohexane/AcOEt 75/25). The ee was
determined by HPLC analysis using a (S,S)-Whelk O1column (hexane/
EtOH 75/25; flow: 0.9 mLminÀ1, t1(minor) =8.10 min, t2(major) =15.01 min).
[12] H. Kim, Y. K. Kim, J. H. Shim, M. Kim, M. Han, T. Livinghouse,
P. H. Lee, Adv. Synth. Catal. 2006, 348, 2609–2618.
[13] N. Meyer, A. Zulys, P. W. Roesky, Organometallics 2006,25, 4179–
4182.
[14] D. V. Gribkov, K. C. Hultzsch, F. Hampel, J. Am. Chem. Soc. 2006,
128, 3748–3759.
[15] D. A. Watson, M. Chiu, R. G. Bergman, Organometallics 2006, 25,
4731–4733.
[16] M. C. Wood, D. C. Leitch, C. S. Yeung, J. A. Kozak, L. L. Schafer,
Angew. Chem. 2007, 119, 358–362; Angew. Chem. Int. Ed. 2007, 46,
354–358.
[17] A. L. Gott, A. J. Clarke, G. J. Clarkson, P. Scott, Organometallics
2007, 26, 1729–1737.
[18] P. H. Martínez, K. C. Hultzsch, F. Hampel, Chem. Commun. 2006,
2221–2223.
[19] a) J. -S. Ryu, T. J. Marks, F. E. McDonald, J. Org. Chem. 2004, 69,
1038–1052; b) J. -S. Ryu, T. J. Marks, F. E. McDonald, Org. Lett.
2001, 3, 3091–3094.
[20] J. Collin, J.-C. Daran, O. Jacquet, E. Schulz, A. Trifonov, Chem. Eur.
J. 2005, 11, 3455–3462.
2-Benzyl-4,4-diphenylpyrrolidine (7h): The same procedure for the prep-
aration of the amide was used as described for 7g. The ee was deter-
mined by HPLC analysis using a (S,S)-Whelk O1column (hexane/EtOH
75/25; flow: 0.9 mLminÀ1, t1 =9.90 min, t2 =20.83 min).
2-Methylindoline (9): The same procedure for the preparation of the
amide was used as described for 7g. The ee was determined by HPLC
analysis using a (S,S)-Whelk O1column (hexane/EtOH 75/25; flow:
0.7 mLminÀ1, t1 =24.67 min, t2 =26.70 min).
X-ray crystallography: Low-temperature diffraction data of Li-3b, Li-
5b[30] were collected on a Kappa X8 APPEX II Bruker diffractometer
with graphite-monochromated MoKa radiation (l=0.71073 Š), by per-
forming f and w scans. The temperature of the crystal was maintained at
the selected value (180 K) by means of a 700 series Cryostream cooling
device to within an accuracy of Æ1K. For complex K- 3d[30] the data
were collected on Stoe IPDS diffractometer operating with monochro-
matic MoKa radiation (l=0.71073). The data were corrected for Lorentz,
polarisation, and absorption effects. The structures were solved by direct
methods with SHELXS-97[31] and refined against F2 by full-matrix least-
squares techniques with SHELXL-97[32] with anisotropic displacement
Chem. Eur. J. 2008, 14, 2189 – 2200
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2199

J. Collin, E. Schulz et al.
[21] D. Riegert, J. Collin, J. -C. Daran, T. Fillebeen, E. Schulz, D.
Lyubov, G. Fukin, A. Trifonov, Eur. J. Inorg. Chem. 2007, 1159–
1168.
[22] R. D. Schannon, Acta Crystallogr. Sect. A 1976, 32, 751–767.
[23] X. Yu, T. J. Marks, Organometallics 2007, 26, 365–376.
[24] M. Shibasaki, N. Yoshikawa, Chem. Rev. 2002, 102, 2187–2210.
[25] a) G. A. Molander, E. D. Dowdy, J. Org. Chem. 1998, 63, 8983–
8988; b) G. A. Molander, E. D. Dowdy, J. Org. Chem. 1999, 64,
6515–6517.
[26] D. C. Bradley, J. S. Ghotra, F. A. Hart, J. Chem. Soc. Chem.
Commun. 1972, 349–350.
[27] S. Vyskocil, S. Jaracz, M. Smrcina, M. Sticha, V. Hanus, M. Polasek,
P. Kocovsky, J. Org. Chem. 1998, 63, 7727–7737.
[29] M. R. GagnØ, C. L. Stern, T. J. Marks, J. Am. Chem. Soc. 1992, 114,
275–294.
[30] CCDC 660504, 660505 and 661065 contain the supplementary crys-
tallographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif..
[31] G. M. Sheldrick, SHELXS-97, Program for Crystal Structure Solu-
tion, University of Gçttingen, Gçttingen, Germany, 1997.
[32] G. M. Sheldrick, SHELXL-97, Program for the refinement of crystal
structures from diffraction data, University of Gçttingen, Gçttingen,
Germany, 1997.
[33] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837–838.
[34] H. D. Flack, Acta Crystallogr. Sect. A 1983, 39, 876–881.
[28] E. B. Tjaden, G. L. Casty, J. M. Stryker, J. Am. Chem. Soc. 1993,
115, 9814–9815.
Received: July 16, 2007
Published online: December 14, 2007
2200
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2189 – 2200