Synthesis of nearly enantiopure allylic amines by aza-Claisen rearrangement of Z-configured allylic trifluoroacetimidates catalyzed by highly active ferrocenylbispalladacycles

Article abstract of DOI:10.1002/chem.200701642

The development of the first highly active enantioselective catalyst for the aza-Claisen rearrangement of Z-configured allylic trifluoroacetimidates generating valuable almost enantiopure protected allylic amines is described. Usually Z-configured allylic

Full text of DOI:10.1002/chem.200701642

DOI: 10.1002/chem.200701642
Synthesis of Nearly Enantiopure Allylic Amines by Aza-Claisen
Rearrangement of Z-Configured Allylic Trifluoroacetimidates
Catalyzed by Highly Active Ferrocenylbispalladacycles
Sascha Jautze,[a] Paul Seiler,[a, b] and RenØ Peters*[a]
Abstract: The development of the first
highly active enantioselective catalyst
for the aza-Claisen rearrangement of
Z-configured allylic trifluoroacetimi-
dates generating valuable almost enan-
tiopure protected allylic amines is de-
planar chiral bispalladacycle complex,
is rapidly prepared from ferrocene in
four simple steps.Key step of this pro-
tocol is an unprecedented highly dia-
stereoselective biscyclopalladation pro-
viding dimeric macrocyclic complexes
of fascinating structure.In the present
study as little as 0.1 mol% of catalyst
precursor were sufficient for most of
the alkyl substituted substrates to give
in general almost quantitative yields.
NMR investigations revealed a mono-
meric structure for the active catalyst
species.The bispalladacycle can also be
used for the formation of almost enan-
tiomerically pure allylic amines (eeꢀ
96%) substituted with important func-
tional groups such as ester, ketone,
ether, silyl ether, acetal or protected
amino moieties providing high-added-
value allylic amine building blocks in
excellent yield (ꢀ94%).The prepara-
tive advantages should render this
scribed.Usually
Z-configured allylic
imidates react significantly slower than
their E-configured counterparts, but in
the present study the opposite effect
was observed. Z-Configured olefins
have the principal practical advantage
that a geometrically pure C=C double
bond can be readily obtained, for ex-
ample, by semihydrogenations of al-
methodology highly appealing as
a
Keywords: allylic amines
· aza-
practical and valuable tool for the for-
mation of allylic amines in target ori-
ented synthesis.
Claisen rearrangement · cyclopalla-
dation · ferrocene · imidazoline
kynes.Our catalyst, a
C2-symmetric
Introduction
mercury trifluoroacetate.[5] The trihaloacetamide protecting
groups can be readily removed from the N atom of the ini-
tial rearrangement products 3 and therefore the overall
transformation leads to primary allylic amines 4.Those are
valuable building blocks due to the presence of two highly
versatile functional groups: the amino and the olefin moiety.
Due to its conceptual attractiveness this approach[6] has
been frequently applied in organic synthesis: more than 180
publications report the use of the thermal or non-enantiose-
The asymmetric aza-Claisen rearrangement[1] of allylic tri-
chloro-[2] and trifluoroacetimidates[3] is a useful synthetic
tool for the formation of allylic trihaloacetamides 3 starting
from readily available allylic alcohols 1 (Scheme 1).In 1974,
37 years after the first description of the thermal aza-Claisen
rearrangement of allylic benzimidates by Mumm and
Mçller,[4] Overman reported the first application of
trichloroacetimidate substrates and demonstrated that the
A
rearrangement can be catalyzed by soft Lewis acids such as
[a] S.Jautze, P.Seiler, Prof.Dr.R.Peters
ETH Zürich, Department of Chemistry and Applied Biosciences
Laboratory of Organic Chemistry
Wolfgang-Pauli-Strasse 10, Hçnggerberg HCI E 111
8093 Zürich (Switzerland)
Fax : (+41)44-633-1226
E-mail: peters@org.chem.ethz.ch
[b] P.Seiler
X-Ray crystal structure analysis.
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Scheme 1.Aza-Claisen (Overman) rearrangement of trihaloacetimidates.
1430
ꢀ 2008 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1430 – 1444

FULL PAPER
lective Lewis acid catalyzed aza-Claisen rearrangement to
prepare allylic amines.[2b]
imidazoline palladacycle (FIP),[8] is the result of a highly
modular catalyst design and the optimization of the catalyst
structure in which both the steric and the electronic proper-
ties of the PdII center could be adjusted by five different
modules: the counteranion X, the bottom Cp spectator
ligand (here CpF =C5Ph5), the oxidation state of the Fe
atom (ferrocene vs.ferrocenium), the imidazoline backbone
and the residue on the amino N atom (here a tosyl group).
The exceptionally high catalytic activity is mainly attribut-
ed to four factors: the possibility to apply almost solvent-
free reaction conditions, the robustness of the active catalyst
under the reaction conditions, the formation of a ferroceni-
um system as catalytically active species by oxidation of the
ferrocene core with a silver salt and the electron withdraw-
ing nature of the five phenyl substituents on the CpF specta-
tor ligand.The last two factors enhance the Lewis acidity of
the Pd center in 7.[9,10] However, complex 7 showed a much
lower catalytic activity for the rearrangement of Z-config-
ured substrates 2 which is a practical disadvantage, since
high enantioselectivities are only achieved with isomerically
pure substrates and Z-configured olefins are in principal
more readily accessible in isomerically pure form, for exam-
ple, by semihydrogenation of a triple bond.For this reason
our goal was to develop a highly enantioselective catalyst
with an extraordinary catalytic activity for Z-configured al-
lylic trifluoroacetimidates.In addition, to be attractive as
synthetic tool, the catalyst should be readily accessible to
allow also large-scale applications.
In 1997 a chiral PdII complex was reported—again by
Overman et al.—acting as the first enantioselective catalyst
for the rearrangement of allylic imidates.[7] The subsequent
development of catalysts with increasing enantioselectivity
concentrated on the rearrangement of allylic benzimidates
yielding benzoylamides as products, which are of limited
synthetic value though, since the hydrolytic cleavage of the
benzoyl amide group proceeds typically with very low yield.
In 2003, the Overman group disclosed the first highly enan-
tioselective catalyst for the rearrangement of the less reac-
tive trihaloacetimidates providing the allylic trihaloaceta-
mides 3 in synthetically useful yields.[2a,3a] However, for tri-
fluoroacetimidates high catalyst loadings were usually re-
quired (10 mol% of PdII) owing to the electron withdrawing
nature of the trihalomethyl group, which diminishes the nu-
cleophilicity of the imidate N atom since it destabilizes the
positive charge in the cationic reaction intermediate 5,
whereas in the case of benzimidates the generated positive
charge is in a benzylic position thus stabilizing the transition
state leading to intermediate 6.
Results and Discussion
Even with these high catalyst loadings long reaction times
were generally required for high conversion.In the case of
trifluoroacetimidate substrates, which appear to be syntheti-
cally slightly more attractive than the corresponding tri-
chloro derivatives due to milder deprotection conditions for
the resulting amide, the scope was limited to substrates
bearing a-unbranched alkyl substituents R’ at the 3-position
of the allylic imidate.These limitations are the reason why
the catalytic asymmetric aza-Claisen rearrangement of tri-
fluoroacetimidates—despite its conceptual potential—has
found almost no application in target oriented synthesis, in
particular on a technical scale.To develop a practical cata-
lyst system of synthetic impact, the issues of catalytic activi-
ty and scope were recently addressed by us with the devel-
opment of the first highly active chiral catalyst (catalyst
loadings are usually at 0.05–0.1 mol%) for the rearrange-
ment of E-configured trifluoroacetimidates providing allylic
trifluoroacetamides 3 with ex-
To further decrease the electron
density of the PdII-centers, we
have designed 1,1’-ferrocenyl-
A
N
tems 8 (FBIP) which could be
prepared by the first direct dia-
stereoselective biscyclopallada-
tion of an enantiomerically
pure sandwich complex as the
key step.[11]
The ligand preparation takes
advantage of the ensuing C2-symmetry reducing the number
of required steps starting from parent ferrocene.[12] The ini-
tial plan to adopt the conditions used for the synthesis of
ferrocenyl monoimidazolines, namely to make use of a pri-
mary amide functionality for the imidazoline formation by
first activating the amide group via alkylation and subse-
quent condensation of the in situ generated iminium ether
with a diamine, was given up, since ferrocenyl-1,1’-diamide
is almost insoluble in aprotic non-nucleophilic solvents such
as CH2Cl2, DCE or CHCl3 thus suppressing the alkylation
step.Bisthioamide 9 was therefore utilized, which is highly
soluble in chlorinated solvents and which was formed by di-
lithiation of ferrocene at room temperature in Et2O and
subsequent trapping of the biscarbanion with N,N-dimethyl-
cellent
enantioselectivities
while tolerating a broad sub-
strate scope: for the first time
a-branched alkyl groups as well
as aromatic substituents were
well tolerated as substituent R’
in the 3-position.[3e] The catalyst
7, a planar chiral ferrocenium
Chem. Eur. J. 2008, 14, 1430 – 1444
ꢀ 2008 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
www.chemeurj.org
1431

R.Peters et al.
aminothiocarbamoyl chloride using a modified literature
procedure (Scheme 2).[13]
Figure 1.ORTEP representation of ligand 11a (R2 =p-Tol).The 30%
probability ellipsoids have the following color code: C (black); N (blue);
O (red); S (yellow); Fe (aquamarine).H atoms are omitted for clarity.
Scheme 2.Synthesis of the C2-symmetric ligands 11.
A
The reactivity of the thioamide functionality was not yet
sufficient for a direct condensation with primary 1,2-di-
minimize unfavorable steric interactions with the metallo-
cene core (Figure 2).
amines even under Lewis acid activation.However, activa-
tion of the thioamide moieties by S-alkylation using Meer-
weinꢁs salt followed by treatment of the solution of the re-
sulting iminium thioether with (R,R)-1,2-diphenylethane-
1,2-diamine (10a) or (S,S)-cyclohexane-1,2-diamine (10b)
N
provided the corresponding bisimidazolines, which could not
be obtained in analytically pure form by column chromatog-
raphy on silica gel or alumina owing to an increased basicity
as compared to ferrocenyl monoimidazolines.For these rea-
sons the crude material was directly subjected to sulfonyl-
Figure 2.Chirality transfer to the sulfonylated N-atoms as major ligand
design principle resulting in a preferred equilibrium conformation of 11
to allow for a highly diastereoselective biscyclopalladation.
A
11a–f (Table 1).
Table 1.Preparation of bisimidazoline ligands 11a–f.
Due to this preferred conformation, the attacking PdII
electrophile is directed by the coordinating imino type N-
atom to only one of the two possible ortho-positions.For
that reason the direct biscyclopalladation proceeds with ex-
cellent diastereoselectivity using various aromatic or non-ar-
Entry
Product
R1
R2
Yield[a] [%]
1
2
3
4
5
6
11a
11b
11c
11d
11e
11 f[b]
Ph
Ph
Ph
Ph
Ph
p-Tol
C6F5
mesityl
p-Ph-C6H4
CF3
57
54
60
41
62
62
omatic sulfonyl residues (Table 2, entries 1–6).CH Cl2 was
2
(CH2)4
p-Tol
employed as a cosolvent to increase the solubility of the
ligand systems in the biscyclopalladation step, while MeOH
is required to dissolve the PdII source.To our knowledge
these are the first and only examples for direct diastereose-
lective biscyclopalladations of enantiomerically pure sand-
wich complexes reported in literature.[16] Previous syntheses
of non-racemic ferrocenyl bispalladacycles relied on double
ortho-lithiation, iodination and subsequent oxidative addi-
tion of Pd0 to the corresponding bisiodoferrocenes.[17]
At the outset of this project we were expecting that the
bispalladacycles would mainly form polymeric linear chains
in which the monomers would be connected via chloride
bridges.However, the 1H NMR spectra for 12a–f show only
a single set of sharp signals indicating the formation of a C2-
[a] Isolated yield over 2 steps starting from 9.[b] ( S,S)-Cyclohexane-1,2-
diamine (10b) was used for the imidazoline formation.The absolute con-
figuration of 11 f is thus opposite to the one depicted in Scheme 2.
The key structural design element to achieve high diaste-
reoselectivity for the subsequent biscyclopalladation is the
strong pyramidalization of the two sulfonylated N atoms
which is confirmed by X-ray crystal structure analysis
(Figure 1):[14,15] steric repulsion between the residue R1 at
the imidazoline 5-position and the sulfonyl group effects a
transfer of chirality to the N atom.
The consequence is a preferred conformation in which the
sulfonyl residue and therefore the substituent at the imid-
1432
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1430 – 1444

Enantiopure Allylic Amines
FULL PAPER
Table 2.Preparation of bispalladacycles 12.
Entry
Product
R1
R2
Yield[a] [%]
dr[b]
1
2
3
4
5
6
12a
12b
12c
12d
12e
12 f[c]
Ph
Ph
Ph
Ph
Ph
p-Tol
C6F5
mesityl
p-Ph-C6H4
CF3
56
34
40
40
35
>100:1
>100:1
>100:1
>100:1
>100:1
>100:1
(CH2)4
p-Tol
5–25
[a] Isolated yield after purification by silica gel filtration.[b] dr referring
to a single Cp unit of the product after silica gel filtration as determined
by 1H NMR.[33] [c] The absolute configuration of 11 f and 12 f is opposite
to the one depicted in Table 2.
symmetric species.An X-ray crystal structure analysis of
12a revealed that the bispalladacycles are C2-symmetric
dimers with an (Sp,S’p,S*p,S*’p)-configuration[18] (Figure 3).[15]
The formation of a single dimeric isomer is remarkable in
the sense, that in principle seven different diastereomeric
dimers with regard to the planar chirality could have been
formed.[19] Due to the theoretical stereochemical complexity,
the dr values provided in Table 2 refer to a single Cp unit
per complex.The complexes 12 can also be regarded as
macrocyclic systems comprising two ferrocenyl units and
four chloride bridged palladium ions (see bottom view in
Figure 3).The dimeric nature of 12 forces two chloride
bridged palladium square planes to be arranged in an
almost co-planar fashion which is otherwise unusual.More-
over, halide bridged ferrocenyl monoimidazoline palladacy-
cle complexes and other conventional halide bridged palla-
dacycles usually form geometric isomers about the PdII cen-
ters.In contrast, in the present case the planar chirality dic-
tates for geometric reasons the coordination sphere about
the PdII square planes in a sense that the all-(Sp)-configured
complexes 12 exist exclusively as all-trans-configured iso-
mers, that is, the imidazoline N-atoms, which are positioned
in the same (PdCl)2 plane, are always arranged trans to each
other.
The moderate but reproducible yields for the biscyclopal-
ladation step (best yield: 56% for R1 =Ph, R2 =p-Tol) is the
consequence of the formation of a side product, which is
most likely the expected oligomeric/polymeric material.This
assumption is supported by extremely complex 1H NMR
spectra.Attempts to transform this side-product into the Cl-
bridged dimer were not successful.Moreover, attempts to
form monomeric acac complexes or PPh3 adducts to simplify
the stereochemical analysis and to allow the unambiguous
identification of the side product remained fruitless.In fact
we cannot rule out the possibility that diastereomers pos-
Figure 3.Ortep-representations of the dimeric bispalladacycle 12a.The
30% probability ellipsoids have the following color code: C (black); N
(blue); O (red); S (yellow); Fe (aquamarine); Pd (bronze); Cl (green).H
atoms are omitted for clarity.At the top: view along the ferrocene axis;
below: view perpendicular to the ferrocene axis.
sessing one (Rp)- and one (Sp)-configured Cp unit per ferro-
cene moiety are consumed by polymerization since no
stable dimeric geometry might be found for such a system.
The isolation of the pure dimeric complexes 12 is neverthe-
less readily achieved by a simple filtration over silica gel
using dichloromethane as eluent, since the side products
have significantly different Rf values and require a more
polar eluent such as pure EtOAc.While in the case of imi-
dazoline systems 11a–e derived from (R,R)-1,2-diphenyl-
ethane-1,2-diamine (10a), no oxidative decomposition was
observed during the biscyclopalladation, with ligand 11 f de-
rived from (R,R)-cyclohexane-1,2-diamine (10b), the prod-
uct was formed in low and irreproducible yield (ranging
from 5–25%) presumably as a result of a significantly lower
stability of both the ligand and the generated Pd complex
(Table 2, entry 6).
The diastereomerically pure bispalladacycle 12a was then
examined in the aza-Claisen rearrangement of the Z-config-
ured trifluoroacetimidate model substrate 13a bearing an
Chem. Eur. J. 2008, 14, 1430 – 1444
ꢀ 2008 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
www.chemeurj.org
1433

R.Peters et al.
Table 3.Screening of conditions for the rearrangement of model substrate 13a.
CHCl3 as solvent for the rear-
rangement, a marked influence
of the activation time was
found (entries 6–9) which is
presumably best explained by
the fact that the chloride bridg-
ed dimeric bispalladacycles 12
are kinetically much more inert
than standard halide bridged
monopalladacycle dimers owing
to the rigid cage-like structure
containing two almost parallel
(PdCl)2 planes.The best turn-
over numbers were accom-
plished with a catalyst activa-
tion time of two days at room
temperature (entry 8) or of 3 h
at 408C (entry 9).With the op-
timized catalyst activation pro-
cedure the catalyst amount
could be reduced to 0.5 mol%
(entry 10) while still obtaining
reasonable conversions at room
temperature.In contrast the re-
action became very slow with
only 0.1 mol% precatalyst acti-
Entry
x mol% 12a
AgX (mol%)
T [8C]
Solvent
Activation
time [h]
t [h]
Yield[a] [%]
ee[b] [%]
1[d]
5
5
5
5
5
5
5
5
5
0.5
0.1
0.1
0.1
AgTFA (10)
AgTFA (20)
AgTFA (30)
AgTFA (30)
AgOTs (30)
AgOTs (30)
AgOTs (30)
AgOTs (30)
AgOTs (30)
AgOTs (3)
20
20
20
20
20
20
20
20
20
20
20
40
55
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
3
3
3
3
3
91
44
20
89
18
24
24
24
24
24
24
72
72
87
80
94
87
95
18
76
83
77
72
14
80
98
91
89
89
91
94
94
96
95
95
97
96
94
95
2[d]
3[d]
4[c,d]
5[d]
6[e]
3
7[e]
22
46
3
45
45
46
46
8[e]
9[e,g]
10[f]
11[f]
12[f]
13[f]
AgOTs (0.6)
AgOTs (0.6)
AgOTs (0.6)
[a] The reactions were performed on 8 mg scale and the yields were determined by 1H NMR after filtration
over silica gel.The remaining material is starting material ( <2% decomposition). 13a contained 2% of the E
isomer.[b] ee determined by HPLC on a chiral stationary phase (Daicel OD-H) after hydrolysis of 14a to the
corresponding secondary amine.[c] Reaction in the presence of 20 mol% proton sponge.[d] 200 mL of solvent
were used.[e] 50 mL of solvent were used.[f] 10 mL of solvent were used.[g] The catalyst was activated at
408C.
nPr substituent at the imidate 3-position.[20] The chloride
bridged dimer displayed no catalytic activity at all, presuma-
bly since in the dimeric rigid cage-like structure an olefin
moiety of the substrate is not able to substitute one of the
strongly binding bridging chloride anions.Consequently, the
chloride counteranion was exchanged by the more labile tri-
fluoroacetate ligand using silver trifluoroacetate (AgTFA)
as a chloride trap.The rearrangement of 13a proceeded still
very sluggishly in CH2Cl2 using 2 equiv of AgTFA/dimer
12a even in the presence of 5 mol% of the bispalladacycle
yet provided (S)-14a in high yield and with 91% ee
(Table 3, entry 1).An increased amount of the silver salt (4
or 6 equiv) improved the reaction rate significantly while
the high enantioselectivity was maintained (entries 2–3).
Proton sponge (1,8-bis(dimethylamino)naphthalene), which
has been used as an additive with several catalysts to sup-
press elimination reactions and which often has a positive
impact on the enantioselectivity of the aza-Claisen rear-
rangement reactions, had in the present case no beneficial
effect with regard to the enantioselectivity, but considerably
slowed down the reaction (entry 4).Compared to AgTFA,
AgOTs slightly increased the catalyst activity and since
product (S)-14a was also formed with enhanced enantiose-
lectivity (94% ee, entry 5), AgOTs was selected as the pre-
ferred catalyst activating agent.To further improve both cat-
alytic activity and enantioselectivity, the influence of various
solvents was investigated.For the solvent screening and all
following experiments, the activation of the precatalyst was
still performed in CH2Cl2.Coordinating solvents such as
acetonitrile, acetone or dioxane resulted in significantly
slower conversion and reduced enantioselectivities.Using
vated by 0.6 mol% AgOTs (entry 11). However, the excel-
lent thermal robustness and the high solubility of the cata-
lyst allowed us to perform the rearrangement at elevated
temperatures and at higher concentrations (entries 12–13):
at 558C, the product was formed with 95% ee in almost
quantitative yield after three days, although the model sub-
strate 13a was not isomerically pure (Z/E 98:2).
After optimization of the model reaction conditions, the
diastereomerically pure bispalladacycles 12a–f which differ
in the residue R2 connected to the sulfonyl moiety and in
the imidazoline backbone (Table 4) were investigated under
identical conditions (0.1 mol% 12, 0.6 mol% AgOTs, cata-
lyst activation time: 46 h at room temperature, CHCl3,
558C).The reactions were stopped after 24 h to directly
compare the catalytic activity.While 12a and 12b gave simi-
lar activities and enantioselectivities demonstrating that the
electronic influence of an aromatic residue R2 is almost neg-
ligible (entries 1–2), the bulkier mesityl- and biphenylsulfon-
yl groups reduced the catalyst activity considerably while
still maintaining high enantioselectivities (entries 3–4).With
the non-aromatic and strongly electron withdrawing SO2CF3
residue, the catalytic activity was surprisingly reduced to a
large degree, while the ee was still excellent (entry 5).The
influence of the imidazoline backbone was demonstrated by
catalyst precursor 12 f derived from cyclohexane-1,2-di-
A
sulted in a considerably lower enantioselectivity (entry 6)
which might be a direct consequence of the catalyst stability.
To achieve a high reproducibility of the catalytic proce-
dure, stem solutions of the hygroscopic silver salt in acetoni-
trile were prepared in a glove box under nitrogen atmos-
1434
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1430 – 1444

Enantiopure Allylic Amines
FULL PAPER
Table 4.Catalyst screening with model substrate 13a.
Entry
12
R1
R2
Reaction
time [h]
Yield[a]
[%]
ee[b]
[%]
1
2
3
4
5
6
a
b
c
d
e
f
Ph
Ph
Ph
Ph
Ph
p-Tol
C6F5
mesityl
p-Ph-C6H4
CF3
24
24
24
24
24
24
78
77
37
21
54
17
96
92
90
92
95
(CH2)4
p-Tol
83[c]
[a] The reactions were performed on 8 mg scale (use of 10 mL of CHCl3)
and the yields were determined by 1H NMR after filtration over silica
gel.The remaining material is starting material ( <2% decomposition).
[b] ee determined by HPLC on a chiral stationary phase (Daicel OD-H)
after hydrolysis of 14a to the corresponding secondary amine.[c] The
R
enantiomer was formed in excess.
Figure 5.NMR investigations of the activated catalyst (in CDCl 3 at
218C) generated from 12a in CH2Cl2 and a stem solution of AgOTs in
MeCN.Spectrum 1 shows the freshly activated catalyst species (after
15 min at high vacuum), which consists largely of the monomeric sym-
metric complex 15.Free MeCN is detected at 19. 9 ppm.Spectra 2–5 were
recorded after repetitive dissolution in CDCl3 and subsequent evapora-
tion of the solvent.Spectrum 6 was obtained after dissolution in a mix-
ture of CDCl3 and toluene and subsequent removal of the solvent (re-
peated twice).Treatment of the resulting material with acetonitrile re-
generated complex 15 as revealed by spectrum 7.
phere.After the precatalyst activation in CH Cl2, the het-
2
A
N
CaH2 (1:1) to remove the AgCl precipitate and unreacted
AgOTs and to remove traces of acid which might cause de-
composition of the acid labile substrates.To acquire more
information about the nature and structure of the catalyti-
cally active species formed from the best precatalyst 12a,
1H NMR investigations were performed showing the pres-
ence of a structurally well defined C2-symmetric major spe-
cies which is formed in large excess and which is in equilibri-
um with a complex mixture.[21] Although the solvent of the
AgOTs stem solution is removed in vacuum prior to the cat-
alyst activation, one acetonitrile molecule is coordinated to
each Pd center in the major species 15 of the activated cata-
lyst (Figures 4 and 5, spectrum 1).
trile as coordinating ligand presumably resulting in the for-
mation of dimeric and oligomeric complexes.This process is
reversible and the coordinated acetonitrile can rapidly ex-
change with free acetonitrile as revealed by the use of free
CD3CN (10 equiv).After 2 min, ca.60% of the coordinated
CH3CN were already replaced at room temperature in
CDCl3 (catalyst concentration: 10 mmolmLꢁ1).
Although we have no direct proof, since attempts to crys-
tallize the activated catalyst for an X-ray crystal structure
1
analysis failed and H NMR NOE analysis was without ex-
planatory power in that case, it is very likely that acetoni-
trile coordinates in the major species trans to the imidazo-
line N-atom as a result of the trans effect[22] in analogy to
the structures of the large majority of palladacycle com-
plexes bearing an additional monodentate N-ligand such as
pyridine or N-methylimidazole.[23] The second coordination
site, which should therefore be cis to the imidazoline unit, is
occupied by tosylate.[24] Even with an excess of free acetoni-
trile, the tosylate is not replaced by the neutral ligand mean-
ing that a cationic Lewis acid is not formed.
Figure 4.Proposed structure of the activated catalyst as revealed by
1H NMR analysis.
If the activated catalyst is subjected to vacuum for extend-
ed periods of time, the amount of the complex side prod-
uct(s) is significantly increased.Repetitive dissolution of the
activated catalyst in CDCl3 and subsequent removal of the
solvent has the same effect (see Figure 5, spectra 2–5).This
effect is dramatically enhanced by evaporation from a tolu-
ene/CDCl3 mixture (spectrum 6).However, addition of ace-
tonitrile regenerates 15 (spectrum 7).The minor species of
the activated catalyst is thus formed by removal of acetoni-
Identical spectra for the activated catalyst were obtained
whether by treatment of 12a with 4, 5 or 6 equiv of AgOTs.
No paramagnetic species was observed even with an excess
of the silver salt indicating that there is almost no oxidation
of the ferrocenyl core providing a ferrocenium salt.This
stands in contrast to previous reports which have shown that
ferrocenyl oxazoline and ferrocenyl imidazoline monopalla-
dacycles are oxidized by silver salts to the corresponding fer-
rocenium species which are the catalytically active spe-
Chem. Eur. J. 2008, 14, 1430 – 1444
ꢀ 2008 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
www.chemeurj.org
1435

R.Peters et al.
cies.[3c,e,25] Even so, 6 equiv of AgOTs per palladacycle dimer
were generally utilized to ensure a complete chloride ex-
change.
For the first time even Z-configured trifluoroacetimidates
bearing an a-branched alkyl moiety were well tolerated as
substrates as exemplified for 13e (R’=iPr, entry 13) provid-
ing 14e in a yield of 64% and with 93% ee.1 mol% of pre-
catalyst and a temperature of 558C were employed in this
case to obtain preparatively useful conversions.The lower
yield as compared to the use of substrates bearing a-un-
branched alkyl groups is a direct consequence of the lower
reaction rate.[26]
Similar arguments as those used for our FIP catalyst 7
bearing the CpF spectator ligand can be used to rationalize
the high enantioselectivity and the absolute configuration of
the rearrangement product.The enantioselectivity determin-
ing step is expected to be the enantioface selective coordina-
tion of the olefin moiety to one PdII center (Figure 5).[27] As-
suming that the olefin will coordinate trans to one imidazo-
line N atom due to the trans effect thus replacing the neutral
acetonitrile ligand,[23] the imidate N atom will attack the
olefin at the face remote to the Pd atom (Figure 6).In the
preferred orientation of the olefin part parallel to the ferro-
cene axis, the sterically undemanding allylic C(1) methylene
moiety should point towards the bulky metallocene core to
minimize unfavorable steric interactions.Coordination of
the enantiotopic olefin face should be less favorable again
owing to steric arguments.[28]
Employing the best catalyst precursor 12a, the scope of
Z-configured imidates 13 was studied (Table 5).The rate of
the rearrangement depends primarily on the steric bulk of
the residue R’.With a-unbranched substituents (R’=Me,
nPr, (CH2)2Ph, iBu), (S)-14 is formed in excellent yield and
enantioselectivity with 1.0 mol% precatalyst at room tem-
perature (entries 1, 4, 6, 9).By decreasing the precatalyst
amount to 0.1 to 0.2 mol% and increasing the reaction tem-
perature to 558C, yields and enantioselectivities were only
marginally affected due to the high robustness of the cata-
lyst against decomposition (entries 2, 5, 7, 10).The rear-
rangements were performed using 25 mg of substrates 13,
yet, working on larger scale (2.6 to 11.6 mmol) provided
similar results (entries 3, 8, 11).
Table 5.Screening of substrates 13 bearing different groups R’.
Entry 13 R’
x mol% 12a T [8C] Yield[a] [%] ee[b] [%]
1[c]
a
a
a
b
b
c
c
c
d
d
d
e
nPr
nPr
nPr
Me
Me
1.0
0.1
0.1
1.0
0.1
1.0
0.2
0.2
1.0
0.1
0.1
1.0
20
55
55
20
55
20
55
55
20
55
55
55
96
94
97
94
97
99
90
92
87
86
93
64
98
97
97
96
94
96
95
97
98
98
99
93
Figure 6.Explanation of the enantioselectivity and absolute configuration
by enantioface selective olefin coordination.
2[c]
3[c,d]
4[e]
5[f]
6[c]
ACHTREUNG
However, using 3-monosubstituted substrates with E-con-
figuration under the above-described reaction conditions op-
timized for the Z-configured systems gave significantly
lower and synthetically unattractive enantioselectivities
(Table 6, ee=62–78%) while the catalytic activities are
slightly reduced with the exception of substrate (E)-13e
bearing the a-branched iPr group (entry 4).The major enan-
tiomers formed from (E)-13 have, as expected and in ac-
cordance with our model, the opposite absolute configura-
tion as those derived from (Z)-13.We can only speculate
about the origin of the reduced selectivity.One of many
possible explanations would, for example, be that the E-con-
figured substrates partially undergo a two-point-binding to
PdII thus replacing both MeCN and the tosylate counterion
by the olefin moiety and the imidate N atom.This would
mean that the nucleophilic attack of the imidate on the
olefin could proceed via the expected anti-aminopalladation,
but in competition with a syn-aminopalladation thus lower-
ing the enantioselectivity.The two-point-binding mode
would also explain the reduced reactivity of the E sub-
7[c]
ACHTREUNG
8[g]
ACHTREUNG
9[c]
iBu
iBu
iBu
iPr
10[c]
11[h]
12[c]
[a] Isolated yield.The reactions were unless otherwise noted performed
on 25 mg scale.[b] ee determined by HPLC on a chiral stationary phase
(Daicel OD-H) after hydrolysis of 14 to the corresponding secondary
amine.[c] Reaction time 3 d.[d] 35. g of substrate 13a (11.6 mmol) were
used. [e] Reaction time 1.5 d. [f] Reaction time
(2.59 mmol) of substrate 13c were used.[h] 12.1 g of substrate
1 d. [g] 0.94 g
13d
(3.83 mmol) were used.
Particularly noteworthy in this series are the excellent ee
values obtained with the most difficult 3-alkyl substituted
substrate in terms of enantioselectivity, imidate 13b bearing
the small methyl substituent (94–96% ee, entries 4–5).The
highest literature value so far was 86% for this specific ex-
ample.[3a]
1436
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1430 – 1444

Enantiopure Allylic Amines
FULL PAPER
Table 6.Investigation of E-configured substrates.
Entry
(E)-10
R’
mol% 12a
Yield[a] [%]
ee[b] [%]
1
2
3
4
a
b
d
e
nPr
Me
iBu
iPr
0.2
0.1
0.5
1.0
84
91
92
87
70
62
78
63
Scheme 3.Preparation of functionalized substrates 19a–c.
[a] The reactions were performed on 25 mg scale for 72 h.The yields
were determined after column chromatography.The remaining material
is mostly starting material.[b] ee determined by HPLC on a chiral sta-
tionary phase (Daicel OD-H) after hydrolysis of 14a to the correspond-
ing secondary amine.
etherification.Boc-protected benzyl amine 31 was synthe-
sized by monoalkylation of benzyl amine, THP cleavage and
carbamate formation.
All of the imidate substrates prepared through intermedi-
ate 23 could be rearranged by the action of a catalytic
strates, since the catalyst might be partially blocked by the
chelating substrate.
amount of [Cl2Pd(NCCH3)2] (10 mol%) providing racemic
A
The observed reaction rate tendency is quite unusual,
since in general E-configured 3-monosubstituted allylic imi-
dates are known to rearrange significantly faster than their
Z-configured counterparts.This is attributed to the fact that
the 3-substituent adopts in the former case a pseudoequato-
rial position in the six-membered cyclic transition state lead-
ing to a cationic intermediate like 5, whereas in the latter
case the substituent is expected to be in the less favorable
pseudoaxial position.
The results for Z-configured substrates presented in
Table 5 are restricted to unfunctionalized trifluoroacetimi-
date substrates.However, a prerequisite for the use of the
easily accessible FBIP in target-oriented synthesis would be
next to high catalyst activities and enantioselectivities its tol-
erance to synthetically important functional groups.[29] The
commercially available Z-configured but-2-en-1,4-diol
served as starting material to synthesize functionalized sub-
strates.Imidates 19a–c containing a silyl ether, benzyl ether
or THP acetal moiety were prepared in two steps by mo-
noprotection[30] followed by imidate formation (Scheme 3).
Attempts to synthesize the racemic rearrangement prod-
ucts 20a–c as reference samples using achiral PdII catalysts
amides 32a–d (Scheme 5).
For the enantioselective rearrangement of functionalized
substrates, FBIP precatalyst 12a was again activated by
AgOTs.Experiments on 8 mg scale at 55 8C using 0.05, 0.1,
0.2, or 0.5 mol% were performed to find the lowest catalyst
loading which still provides complete conversion within
72 h.The best conditions were subsequently applied on
250 mg scale (Table 7).For most of the investigated sub-
strates the required catalyst amount was slightly higher than
for the non-functionalized substrates.This outcome is attrib-
uted to the competitive reversible binding of the heteroatom
substituents to the catalyst.For the TBS protected alcohol
19a as little as 0.05 mol% resulted in a yield of 94–96% on
250 mg and 4 g scale, respectively (entries 5 and 6), since the
TBS protected oxygen atom is not an efficient Lewis base.
Ester, ketone, ether, silyl ether, acetal and Boc protected
amino groups are all well tolerated and the optimized reac-
tion conditions provided the protected, functionalized allylic
amines in nearly quantitative isolated yield for each exam-
ple with the catalyst loading generally not exceeding 0.2–
0.5 mol%. All of these high-added-value building blocks
were produced with excellent enantioselectivity (ee=96–
98%).The value of the presented methodology is further
enhanced by the operational simplicity.The solution of the
activated catalyst is added to neat substrate, the reaction
vessel is subsequently sealed and heated for three days to
558C.After cooling to room temperature, a mixture of cy-
clohexane/ethyl acetate is added causing a precipitation of
the Pd complex.The product is subsequently isolated in an-
alytically pure form by a simple filtration.
such as [Cl2Pd(NCCH3)2] were unsuccessful and resulted
A
either in no conversion or in decomposition.The racemic
samples had to be prepared via thermally induced [3,3]-sig-
matropic rearrangements in the presence of catalytic
amounts of proton sponge to trap traces of acid in order to
avoid substrate decomposition (Scheme 3).
Additional imidate substrates were prepared via alkyl-
A
sequent Krapcho decarboxylation providing THP acetal
23,[30a,31] which was used to prepare various functionalized
derivatives (Scheme 4).Acidic hydrolysis of the acetal
moiety and ensuing imidate formation gave access to 24
containing a methyl ester group.A methyl ketone derivative
was prepared by formation of morpholine amide 25 and
monoaddition of MeLi to the amide moiety,[32] while benzyl
ether 29 was obtained via reduction of 23 and Williamson
Conclusion
We have developed the first highly active enantioselective
catalyst for the aza-Claisen rearrangement of Z-configured
trifluoroacetimidates requiring as little as 0.1 mol% of cata-
lyst precursor for most of the alkyl substituted substrates.
Chem. Eur. J. 2008, 14, 1430 – 1444
ꢀ 2008 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
www.chemeurj.org
1437

R.Peters et al.
Scheme 5.Synthesis of racemic functionalized allylic amine derivatives
32a–d.
Table 7.Catalytic asymmetric synthesis of functionalized allylic amine
building blocks.
Entry Product R’
x mol% Yield[a] [%] ee[b] [%]
1
2
3
4
32a
32b
32c
32d
20a
20a
20b
20c
C
0.5
98
97
100
99
94
96
94
99
98
97
97
98
97
97
98
96
0.2
5
0.05
0.05
0.2
6[c]
7[d]
8
0.2
[a] The reactions were unless otherwise noted performed on 250 mg
scale.Isolated yield after filtration over silica gel.[b] Determined by
HPLC on a chiral stationary phase (Daicel OD-H).[c] The reaction was
performed on 4.0 g scale. [d] The reaction was performed on 25 mg scale.
Proton sponge (0.2 mol%) was required to avoid decomposition of the
acetal moiety.
value building blocks in excellent yield (ꢀ94%; 0.05–
0.5 mol% of catalyst precursor). These results—the simple
and rapid ligand and catalyst preparation, the unprecedent-
ed catalytic activity for Z-configured allylic trifluoroacetimi-
date substrates, the excellent enantioselectivity, scalability
and functional group compatibility as well as the operational
simplicity—should render this methodology highly appealing
as a practical and valuable tool for target oriented synthesis.
Experimental Section
General methods and additional selected experimental procedures are
given in the Supporting Information.
Ferrocene-1,1’-bis-N,N-dimethylthioamide (9):[13] Ferrocene (13.36 g,
71.8 mmol) was dissolved in diethyl ether (300 mL) and nBuLi (99 mL,
158.4 mmol, 2.2 equiv, 1.6 molLꢁ1 in n-hexane) and TMEDA (26 mL,
180 mmol, 2.4 equiv) were added successively. After 19.5 h the reaction
mixture was cooled down to ꢁ788C and a solution of N,N-dimethylthio-
carbamate chloride (19.5 g, 160 mmol, 2.20 equiv) in Et2O (170 mL) was
added.After 5 min the cooling bath was removed and stirring was contin-
ued for additional 120 min.The reaction mixture was filtered, the filter
cake was extracted with CHCl3 and the solvent was removed.The re-
maining solid residue was washed with Et2O.The product was obtained
as a dark red crystalline solid (13.9 g, 38.8 mmol, 54%). M.p. 163–1648C;
1H NMR (300 MHz, CDCl3, 218C): d=4.78 (t, J=1.8 Hz, 4H, o-C5H4R),
Scheme 4.Preparation of functionalized substrates 24, 27, 29 and 31.
Key step of the catalyst formation is an unprecedented
highly diastereoselective biscyclopalladation providing di-
meric macrocyclic complexes of fascinating structure.This
novel catalyst type also allowed for the formation of almost
enantiomerically pure allylic amines (eeꢀ96%) substituted
with important functional groups (esters, ketones, ethers,
silyl ethers, acetals, protected amines) providing high-added-
1438
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1430 – 1444

Enantiopure Allylic Amines
FULL PAPER
4.43 (t, J=1.8 Hz, 4H, m-C5H4R), 3.49 (s, 6H, NCH3), 3.32 ppm (s, 6H,
NCH3); 13C NMR (75 MHz, CDCl3, 218C): d=198.3, 88.9, 76.2, 72.4,
45.2, 44.7 ppm; IR (film): n˜ =3074, 2925, 1510, 1505, 1434, 1386, 1333,
4H, p-H of ph), 7.21 (m, 20H, arom. H), 7.02 (t, J=7.5 Hz, 8H, m-H),
6.95 (d, J=8.4 Hz, 8H, arom. H), 6.60 (d, J=7.5 Hz, 8H, arom. H), 5.06
(dd, J1 =2.4, J2 =0.9 Hz, 4H, o-C5H3RR’), 4.94 (d, J=2.4 Hz, 4H, CHPh),
4.80 (d, J=2.4 Hz, 4H, CHPh), 4.56 (dd, J1 =2.4, J2 =0.9 Hz, 4H, o-
C5H3RR’), 4.41 (t, J=2.4 Hz, 4H, m-C5H3RR’), 2.36 ppm (s, 12H, CH3);
13C NMR (75 MHz, CDCl3, 218C): d=169.9, 144.9, 140.7, 139.9, 134.3,
129.8, 129.5, 128.7, 128.5, 127.3, 127.0, 126.8, 125.4, 96.6, 77.1, 75.9, 75.2,
71.6, 69.6 (2C), 21.5 ppm; IR (film): n˜ =3029, 1596, 1555, 1454, 1360,
1273, 1133, 1057, 990, 913 cmꢁ1
; HRMS (MALDI): m/z: calcd for
C16H21N2S2Fe: 361.0490 [M+H]+; found: 361.0482; elemental analysis
calcd (%) for C16H20N2S2Fe (360.32): C 53.33, H 5.59, N 7.77; found: C
53.44, H 5.58, N 7.81.
(4’’R,5’’R)-1,1’-Bis-(2’’-4’’,5’’-dihydro-4’’,5’’-diphenyl-1’’-H-imidazolyl) fer-
1168, 1095, 1028, 969 cmꢁ1
108H88N8O8S4Cl4Fe2Pd4: 2432.9232 [M]+; found: 2432.9192; elemental
;
HRMS (MALDI): m/z: calcd for
rocene (33a):
A
solution of Meerweinꢁs salt (579 mg, 3.05 mmol,
(536 mg,
C
2.05 equiv) in CH2Cl2 (7 mL) was added to thioamide
6
analysis calcd (%) for C108H88N8O8S4Cl4Fe2Pd4 (2433.34): C 53.31, H 3.64,
N 4.60.; found: C 53.59, H 3.42, N 4.52.
1.48 mmol) in CH2Cl2 (10 mL).After 35. h ( R,R)-1,2-diphenylethane-1,2-
diamine (10a) (663 mg, 2.10 equiv, 3.12 mmol) dissolved in CH2Cl2
(7 mL) was added rapidly.After 17 h at room temperature, ca.75% of
the solvent were removed and diisopropyl ether (40 mL) was added caus-
ing precipitation.The supernatant was removed by decantation and the
solid residue was taken up in CH2Cl2 and washed with 2n NaOH.The or-
ganic phase was dried with Na2CO3 and the solvent was removed after
filtration.The crude product (988 mg) was directly used for the sulfonyl-
General procedure (GP 3) for the catalyst activation (Table 5): A solu-
tion of AgOTs (37.4 mg, 0.134 mmol, 6.0 equiv) in MeCN (0.155 mL) was
evaporated to dryness and a solution of 12a (54.4 mg, 0.0224 mmol) in
CH2Cl2 (2 mL) was added to the silver salt.The reaction mixture was
subsequently diluted under nitrogen atmosphere with CH2Cl2 (14 mL)
and stirred in the absence of light.After 46 h the mixture was filtered
over CaH2/Celite (1:1; vol: 1 cm3).The filter cake was extracted with
CH2Cl2 until the organic solution was colorless.The solvent was removed
with a steady flow of N2 and finally by using high vacuum (15 min).A
stem solution of activated catalyst 15 was prepared by dissolving the
solid in dry CHCl3 (1.05 mL; accounts for 20.0 mmolLꢁ1 of dimer 12a in
its activated form). 1H NMR (300 MHz, CDCl3, 218C): d=7.63–7.12 (m,
48H, arom. H), 7.01 (t, J=7.8 Hz, 8H, arom. H), 6.92 (d, J=7.6 Hz, 8H,
arom. H), 6.54 (d, J=7.2 Hz, 8H, arom. H), 5.78 (d, J=2.1 Hz, 4H,
C5H3RR), 5.29 (d, J=2.1 Hz, 4H, C5H3RR), 5.07 (d, J=3.3 Hz, 4H,
CHPh), 5.02 (d, J=3.3 Hz, 4H, CHPh), 2.64 (s, 12H, CH3CN!Pd), 2.41
A
20H, arom. H), 6.31 (br, 2H, NH), 4.78–4.63 (br, 4H, CHPh) 4.74 (br,
2H, o-C5H4R), 4.68 (br, 2H, o-C5H4R), 4.58 (br, 2H, m-C5H4R),
4.43 ppm (br, 2H, m-C5H4R); 13C NMR (75 MHz, CDCl3, 218C): d=
164.6, 143.2, 128.5, 127.2, 126.7, 74.5, 73.0, 71.1, 70.6, 69.1 ppm; HRMS
(MALDI): m/z: calcd for C40H35N4Fe: 627.2206 [M+H]+; found:
627.2195.
General procedure (GP 1) for the sulfonylation of bisimidazolines 33:
A
N
DMAP and the sulfonylating agent were successively added at 08C.The
reaction mixture was slowly warmed to room temperature.After 3–18 h
additional CH2Cl2 was added and the mixture was washed with saturated
aqueous Na2CO3 solution.The aqueous phase was extracted once with
CH2Cl2 and the combined organic phases were dried over NaHCO3.The
crude products were purified by flash column chromatography.
(s, 12H, N-Ts
N
N
free CH3CN); MS (MALDI): m/z: calcd for C61H51FeN4O7Pd2S3: 1316.96
[M]+; found: 1316.96.
General procedure (GP 4) for the rearrangement of Z-configured sub-
strates 13 using 12a as catalyst precursor (Table 5): To substrates 13
(25 mg) a solution of the activated catalyst (see GP 3) in CHCl3 (entry 1:
39.0 mL; entry 4: 43.0 mL; entry 6: 32.3 mL; entry 9: 37.3 mL; entry 12:
39.0 mL; for entries 2, 5, 7 and 10 the catalyst stem solution was further
diluted to a concentration of 4.47 mmolLꢁ1; amount used of the diluted
stem solution: entry 2: 18.5 mL; entry 5: 20.5 mL; entry 7: 30.8 mL;
entry 10: 17.8 mL) was added and the reactions were stirred in the ab-
sence of light under nitrogen atmosphere for the indicated time at the in-
dicated temperature in a sealed and shielded flask.The remaining solvent
was removed in vacuo at room temperature, the residue was suspended
in cyclohexane/EtOAc 9:1 and subsequently purified by filtration over
silica gel (for 14a–d) or column chromatography (14e; cyclohexane/
EtOAc 9:1).
(4’’R,5’’R)-1,1’-Bis-(2’’-4’’,5’’-dihydro-4’’,5’’-diphenyl-1’’-tosyl-imidazolyl)
ferrocene (11a): According to GP 1, crude 33a (395 mg, < 0.630 mmol)
was treated with NEt3 (351 mL, 2.52 mmol, > 4.0 equiv), DMAP (14 mg,
0.11 mmol, > 0.18 equiv) and TsCl (480 mg, 2.52 mmol, > 4.0 equiv) for
3 h in CH2Cl2 (10 mL).After flash chromatography (cyclohexane/EtOAc
4:1, 3% NEt3) 11a was obtained as a red crystalline solid (334 mg,
0.357 mmol, 57% starting from 9).Mp..225
8C (decomp); [a]2D5 =(c=
0.110 gdLꢁ1, CHCl3)=ꢁ578.1; 1H NMR (300 MHz, CDCl3, 218C): d=
7.53–7.37 (m, 8H, arom. H), 7.36 (t, J=7.2 Hz, 2H, p-H of Ph), 7.24–7.16
(m, 6H, arom. H), 7.11 (t, J=7.2 Hz, 4H, m-H of Ph), 7.03 (d, J=7.8 Hz,
4H, arom. H), 6.70 (d, J=6.9 Hz, 4H, arom. H), 5.29 (t, J=1.5 Hz, 2H,
o-C5H4R), 5.08 (d, J=3.9 Hz, 2H, CHPh), 5.06 (t, J=1.2 Hz, 2H, o-
C5H4R), 5.02 (d, J=3.6 Hz, 2H, CHPh), 4.51 (m, 2H, m-C5H4R), 4.43
(m, 2H, m-C5H4R), 2.37 ppm (s, 6H, CH3); 13C NMR (75 MHz, CDCl3,
218C): d=158.9, 144.1, 142.2, 141.7, 134.6, 129.4, 129.2, 128.4, 128.0,
127.6, 127.0, 125.8, 77.3, 74.5, 74.1, 73.6, 73.5, 73.2, 72.6, 21.6 ppm; IR
(S)-2,2,2-Trifluoro-N-(4-methoxyphenyl)-N-(1-propylallyl)acetamide [(S)-
14a]: GP 4: Catalyst amount: 1.0 mol%; reaction time: 3 d; reaction tem-
perature: 208C, yield: 96%.Catalyst amount: 01. mol%; reaction time:
3 d; reaction temperature: 558C, yield: 94%.The analytical data were in
accordance with the literature.[3a,e]
(film): n˜ =2924, 1620, 1494, 1454, 1366, 1282, 1172, 1090, 1022 cmꢁ1
;
HRMS (MALDI): m/z: calcd for C54H47N4O4S2Fe: 935.2383 [M+H]+;
found: 935.2366; elemental analysis calcd (%) for C54H46N4O4S2Fe
(934.94): C 69.37, H 4.96, N 5.99; found: C 69.38, H 4.79, N 5.99.
Large-scale rearrangement of 13a (Table 5, entry 3):
A solution of
AgOTs (22.4 mg, 0.0801 mmol, 6.0 equiv) in MeCN (0.655 mL) was
evaporated to dryness and a solution of 12a (32.5 mg, 0.0134 mmol) in
CH2Cl2 (9.5 mL) was added in the absence of light under nitrogen atmos-
General procedure (GP 2) for the biscyclopalladation of bisimidazolines
11: A solution of bisimidazolines 11 in a 1:1 mixture of CH2Cl2/methanol
was cooled to 08C.Na 2PdCl4 and NaOAc were added and the reaction
was allowed to slowly warm to room temperature.After stirring for the
indicated time at this temperature, the solvent was completely removed
in vacuo at 408C and the crude product was subjected to a filtration over
silica gel with CH2Cl2 as eluent.
phere to the silver salt.After 46 h the mixture was filtered over CaH
/
2
Celite (1:1; vol: 1.5 cm3).The filter cake was extracted with CH 2Cl2 until
the organic solution was colorless.The solvent was removed with a
steady flow of N2 and finally by using high vacuum.A stem solution of
the activated catalyst was prepared by dissolving the solid in dry CHCl3
(1.00 mL; accounts for 13.3 mmolLꢁ1 of the dimer 12a in its activated
Bis(di-m-chloro-m-{[bis-h5-(4’R,5’R)-(Sp)-2-(2’-4’,5’-dihydro-4’,5’-diphenyl-
1’-tosyl-imidazolyl)cyclopentadienyl)] iron(II) 1-C,3’-N} dipalladium (II))
(12a): According to GP 2, 11a (224 mg, 0.240 mmol) was treated with
Na2PdCl4 (166 mg, 0.563 mmol, 2.35 equiv) and NaOAc (55 mg,
0.67 mmol, 2.8 equiv) in 18 mL of solvent for 18 h. 12a was isolated as a
dark red crystalline substance (162 mg, 0.133 mmol, 56%). M.p.
form).806. mL of the solution were added to neat
13a (3.500 g,
11.62 mmol) and the solution was stirred in a sealed and shielded flask
for 3 d at 558C in the absence of light under nitrogen atmosphere.The
residue was suspended in cyclohexane/EtOAc 9:1 and subsequently puri-
fied by filtration over silica gel (cyclohexane/EtOAc 9:1) to obtain 14a
(3.404 g, 11.29 mmol, 97%, ee=97%).
1
>2308C; [a]2D5 =(c=0.0215 gdLꢁ1, CHCl3)=ꢁ305.4; H NMR (300 MHz,
(S)-2,2,2-Trifluoro-N-(4-methoxyphenyl)-N-(1-methylallyl)acetamide
[(S)-14b]: GP 4: Catalyst amount: 1.0 mol%; reaction time: 40 h; reac-
CDCl3, 218C): d=7.63 (d, J=7.5 Hz, 8H, arom. H), 7.35 (t, J=7.5 Hz,
Chem. Eur. J. 2008, 14, 1430 – 1444
ꢀ 2008 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
www.chemeurj.org
1439

R.Peters et al.
tion temperature: 208C, yield: 94%.Catalyst amount: 01. mol%; reac-
tion time: 1 d; reaction temperature: 558C, yield: 97%.The analytical
data were in accordance with the literature.[3a,e]
(S)-N-(4-Methoxyphenyl)-3-amino-5-methyl-1-hexene from allylic amide
[(S)-14d]: Application of the hydrolysis procedure developed by Over-
man et al.[3a] gave the deacylated product in 98/98/99% ee (1.0 (Table 5,
entry 9)/0.1 (entry 10)/0.1 mol% (entry 11) precatalyst) for the formation
(S)-2,2,2-Trifluoro-N-(4-methoxyphenyl)-N-(1-phenethylallyl)acetamide
[(S)-14c): GP 4: Catalyst amount: 1.0 mol%; reaction time: 3 d; reaction
temperature: 208C, yield: 99%.Catalyst amount: 02. mol%; reaction
time: 3 d; reaction temperature: 558C, yield: 90%.The analytical data
were in accordance with the literature.[3a,e]
of (S)-14d) (Chiralcel OD-H, 99.8:0.2 n-hexane/iPrOH, 0.8 mLminꢁ1
with S configuration.The other analytical data were in accordance with
the literature.[3a,e]
)
(S)-N-(4-Methoxyphenyl)-3-amino-4-methyl-1-pentene from allylic amide
[(S)-14e]: Application of the hydrolysis procedure developed by Over-
man et al.[3a] gave the deacylated product in 93% ee (1.0 mol% precata-
lyst for the formation of 14e, Table 5, entry 12) (Chiralcel OD-H,
99.8:0.2 n-hexane/iPrOH, 0.8 mLminꢁ1) with (S)-configuration. 1H NMR
(300 MHz, CDCl3, 218C): d=6.77–6.73 (m, 2H, arom-H), 6.59–6.55 (m,
2H, arom-H), 5.71 (ddd, J=17.1, 10.2, 6.6 Hz, 1H, CH2=CH), 5.14 (m,
2H, CH2=CH), 3.73 (s, 3H, OCH3), 3.56 (dd, J=6,3, 5.1 Hz, NCH), 1.86
Large-scale rearrangement of 13c (Table 5, entry 8):
A solution of
AgOTs (22.9 mg, 0.08020 mmol, 6.0 equiv) in MeCN (0.260 mL) was
evaporated to dryness and a solution of 12a (33.8 mg, 0.0139 mmol) in
CH2Cl2 (9.6 mL) was added to the silver salt in the absence of light
under nitrogen atmosphere.After 46 h the mixture was filtered over
CaH2/Celite (1:1; vol: 1.5 cm3).The filter cake was extracted with CH Cl2
2
until the organic solution was colorless.The solvent was removed with a
steady flow of N2 and finally by using high vacuum.A stem solution of
the activated catalyst was prepared by dissolving the solid in dry CHCl3
(1.00 mL; accounts for 13.3 mmolLꢁ1 of the dimer 12a in its activated
(m, CH
N
N
CH(CH3)2).The other analytical data were accordance with the literature
T
data for the R isomer.[3e]
General procedure (Gp 5) for the aza-Claisen rearrangement of Z-con-
form).03.78 mL of the solution were added to neat
13c (0.940 g,
figured functionalized substrates (Table 7):
A solution of AgOTs
2.59 mmol) and the solution was stirred in a sealed and shielded flask for
3 d at 558C in the absence of light under nitrogen atmosphere.The resi-
due was suspended in cyclohexane/EtOAc 9:1 and subsequently purified
by filtration over silica gel (cyclohexane/EtOAc 9:1) to obtain 14c
(0.862 g, 2.37 mmol, 92%, ee=97%).
(27.5 mg, 0.0986 mmol, 6.0 equiv/12a) in MeCN (0.214 mL) was evapo-
rated to dryness and a solution of FBIP-Cl (12a, 40.0 mg, 0.0164 mmol)
in CH2Cl2 (12.0 mL) was added to the silver salt. The reaction mixture
was stirred in the absence of light.After 46 h the mixture was filtrated
over CaH2/Celite (1:1, vol: 1 cm3).The filter cake was extracted with
CH2Cl2 until the filtrate was colorless.The solvent was removed with a
steady flow of N2 and finally by high vacuum.A stem solution of the acti-
vated catalyst was prepared by dissolving the solid in dry CHCl3 (1.0 mL
for entries 1, 2, 3, 4, 8 ꢂ 16.4 mmolLꢁ1 of dimer 12a in its activated
form; 3.0 mL for entries 5, 7 ꢂ 5.4 mmolLꢁ1 of dimer 12a in its activated
form).To substrates 24, 27, 29, 31 or 19a (250 mg, entries 1–5), 19c
(210 mg, entry 8) and 19b (25.0 mg, entry 7) a solution of the activated
catalyst in CHCl3 [entry 1: 225 mL (0.5 mol%); entry 2: 236 mL
(0.5 mol%); entry 3: 76 mL (0.2 mol%); entry 4: 61 mL (0.2 mol%);
entry 5: 57.0 mL (0.05 mol%); entry 7: 24.6 mL (0.5 mol%); entry 8:
69 mL (0.5 mol%)] was added. In case of 19b (entry 7) a solution of
proton sponge (0.03 mg, 0.2 mol%) in CHCl3 (15.1 mL) was previously
added.The reactions were stirred for 72 h at 55 8C in a sealed flask in the
absence of light and under nitrogen atmosphere.The residue was subse-
quently suspended in cyclohexane/EtOAc 9:1 and filtrated over a patch
of silica gel (cyclohexane/EtOAc 9:1).Since a uniform reaction pathway
can be assumed,[3e] the absolute configuration has been assigned based on
the configurations determined for unfunctionalized rearrangement prod-
ucts prepared by the action of the (Sp)-configured FBIP catalyst.
(S)-2,2,2-Trifluoro-N-(4-methoxyphenyl)-N-(1-isobutylallyl)acetamide
[(S)-14d]: GP 4: Catalyst amount: 1.0 mol%; reaction time: 3 d; reaction
temperature: 208C, yield: 87%.Catalyst amount: 01. mol%; reaction
time: 3 d; reaction temperature: 558C, yield: 86%.The analytical data
were in accordance with the literature.[3a,e]
Large-scale rearrangement of 13d (Table 5, entry 11): A solution of
AgOTs (22.9 mg, 0.08020 mmol, 6.0 equiv) in MeCN (0.260 mL) was
evaporated to dryness and a solution of 12a (33.8 mg, 0.0139 mmol) in
CH2Cl2 (9.6 mL) was added to the silver salt in the absence of light
under nitrogen atmosphere.After 46 h the mixture was filtered over
CaH2/Celite (1:1, vol: 1.5 cm3).The filter cake was extracted with CH Cl2
2
until the organic solution was colorless.The solvent was removed with a
steady flow of N2 and finally by using high vacuum.A stem solution of
the activated catalyst was prepared by dissolving the solid in dry CHCl3
(1.00 mL; accounts for 13.3 mmolLꢁ1 of the dimer 12a in its activated
form).02.79 mL of the solution were added to neat
13d (1.207 g,
3.828 mmol) and the solution was stirred in a sealed and shielded flask
for 3 d.The residue was suspended in cyclohexane/EtOAc 9:1 and subse-
quently purified by filtration over silica gel (cyclohexane/EtOAc 9:1) to
obtain 14d (1.117 g, 3.542 mmol, 93%, ee 99%).
rac-N-(1-(tert-Butyldimethylsilyloxy)but-3-en-2-yl)-2,2,2-trifluoro-N-(4-
methoxyphenyl)acetamide (20a): A solution of 19a (142 mg, 0.352 mmol)
and proton sponge (14 mg, 0.070 mmol, 0.20 equiv) in mesitylene
(10 mL) was heated to 1608C for 43 h.After solvent removal the crude
product was submitted to column chromatography (pentane/EtOAc
200:1) to provide racemic 20a as a colorless oil (129 mg, 0.320 mmol,
91%). 1H NMR (300 MHz, CDCl3, 218C): d=7.32–7.29 (m, 1H, arom.
H), 7.11–7.07 (m, 1H, arom. H), 6.88–6.85 (m, 2H, arom. H), 5.69–5.57
(m, 1H, CH2=CH), 5.31–5.22 (m, 2H, CH2=CH), 5.08 (q, J=7.2 Hz, 1H,
CH-N), 3.82 (s, 3H, OCH3), 3.64 (d, J=7.2 Hz, 2H, CH2-O), 0.90 (s, 9H,
tBu), 0.07 ppm (s, 6H, Si-CH3); 13C NMR (75 MHz, CDCl3, 218C): d=
159.7, 156.8 (q, J=34.8 Hz, C-CF3), 132.0, 131.6, 131.3, 127.9, 120.2, 116.3
(q, J=286.5 Hz, CF3), 113.6, 62.9, 61.6, 25.9, 18.2, ꢁ5.3 ppm; 19F NMR
(282 MHz, CDCl3, 218C): d=ꢁ67.2 ppm; IR (film): n˜ =2957, 2932, 2860,
1699, 1610, 1513, 1466, 1301, 1254, 1207, 1154, 1108, 1035 cmꢁ1; HRMS
(EI): m/z: calcd for C15H19F3NO3Si: 346.1081 [MꢁC4H9]+; found:
346.1080; elemental analysis calcd (%) for C19H28F3NO3Si (403.51): C
56.56, H 6.99, N 3.47. Found: C 56.79, H 6.91, N 3.46.
(S)-2,2,2-Trifluoro-N-(4-methoxyphenyl)-N-(1-isopropylallyl)acetamide
[(S)-14e]: GP 4: Catalyst amount: 1.0 mol%; reaction time: 3 d; reaction
temperature: 558C, yield: 65%.The analytical data were in accordance
with the literature.[3e]
(S)-N-(4-Methoxyphenyl)-3-amino-1-hexene from allylic amide [(S)-14a]:
Application of the hydrolysis procedure developed by Overman et al.[3a]
gave the deacylated product in 98/97% ee (1.0 (Table 5, entry 1)/
0.1 mol% (entries 2, 3) precatalyst) for the formation of (S)-14a) (Chiral-
cel OD-H, 99.8:0.2 n-hexane/iPrOH, 0.8 mLminꢁ1) with S configuration.
The other analytical data were in accordance with the literature.[3a]
(S)-N-(4-Methoxyphenyl)-3-amino-1-butane from allylic amide [(S)-14b]:
Application of the hydrolysis procedure developed by Overman et al.[3a]
gave the deacylated product in 96/94% ee (1.0 (Table 5, entry 4)/
0.1 mol% (entry 5) precatalyst) for the formation of (S)-14b) (Chiralcel
OD-H, 99.8:0.2 n-hexane/iPrOH, 0.8 mLminꢁ1) with S configuration.The
other analytical data were in accordance with the literature.[3a]
(S)-N-(4-Methoxyphenyl)-3-amino-5-phenyl-1-pentene from allylic amide
[(S)-14c]: Application of the hydrolysis procedure developed by Over-
man et al.[3a] gave the deacylated product in 96/95/97% ee (1.0 (Table 5,
entry 6)/0.2 (entry 7)/0.2 mol% (entry 8) precatalyst) for the formation of
(S)-14c) (Chiralcel OD-H, 98.2:1.8 n-hexane/iPrOH, 0.8 mLminꢁ1) with
S configuration.The other analytical data were in accordance with the lit-
erature.[3a]
(R)-N-(1-(tert-Butyldimethylsilyloxy)but-3-en-2-yl)-2,2,2-trifluoro-N-(4-
methoxyphenyl)acetamide (20a): GP 5 (Table 7, entry 5): colorless oil,
yield: 94%, ee: 97%. The ee value was determined by chiral column
HPLC after hydrolysis of the amide and TBS group (see below) to give
2-(4-methoxyphenylamino)but-3-en-1-ol: Chiralcel OD-H, n-hexane/
iPrOH 90:10, 0.8 mLminꢁ1, detection at 210 nm.[ a]2D5 =(c=0.902 gLꢁ1
,
1440
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1430 – 1444

Enantiopure Allylic Amines
FULL PAPER
1
CHCl3)=ꢁ4.8; H NMR (300 MHz, CDCl3, 218C): d=7.32–7.29 (m, 1H,
Column chromatography (cyclohexane/EtOAc 4:1, 3% NEt3 ! cyclo-
arom. H), 7.11–7.07 (m, 1H, arom. H), 6.88–6.85 (m, 2H, arom. H), 5.69–
5.57 (m, 1H, CH2=CH), 5.31–5.22 (m, 2H, CH2=CH), 5.08 (q, J=7.2 Hz,
1H, CH-N), 3.82 (s, 3H, Ar-OCH3), 3.64 (d, J=7.2 Hz, 2H, CH2-O), 0.90
(s, 9H, tBu), 0.07 ppm (s, 6H, Si-CH3).The other analytical data were in
accordance with the racemate (see above).
hexane/EtOAc 2:1, 3% NEt3) of the residue gave 2-(4-methoxyphenyl-
G
1
(c=0.835 gdLꢁ1, CHCl3)=ꢁ25.0; H NMR (300 MHz, CDCl3, 218C): d=
6.78 (dt, J=9.0, 2.4 Hz, 2H, arom. H), 6.65 (dt, J=9.0, 2.4 Hz, 2H, arom.
H), 5.79 (ddd, J=17.1, 10.5, J=5.4 Hz, 1H, CH2=CH), 5.30 (dt, J=17.4,
1.5 Hz, 1H, CH2=CH), 5.24 (dt, J=10.5, 1.5 Hz, 1H, CH2=CH), 3.98–
3.92 (m, 1H, CH-N), 3.81–3.75 (m, 1H, OH or NH), 3.75 (s, 3H, Ar-
OCH3), 3.64–3.57 (m, 2H, CH2-O), 1.94 ppm (b, 1H, OH or NH);
13C NMR (75 MHz, CDCl3, 218C): d=152.6, 141.0, 136.5, 117.3, 115.6,
114.7, 64.8, 58.8, 55.6 ppm; IR (film): n˜ =3381, 2935, 2833, 1618, 1513,
1465, 1239, 1180, 1037, 993 cmꢁ1; HRMS (EI): m/z: calcd for C11H15NO2:
193.1098 [M]+; found: 193.1098; elemental analysis calcd (%) for
C11H15NO2 (193.24): C 68.37, H 7.82, N 7.25. Found: C 68.13, H 7.61, N
7.07.
Large-scale rearrangement of 19a (Table 7, entry 6):
A solution of
AgOTs (24.9 mg, 0.0893 mmol, 6.0 equiv) in MeCN (0.478 mL) was
evaporated to dryness and a solution of 12a (36.5 mg, 0.0149 mmol) in
CH2Cl2 (10.5 mL) was added in the absence of light under nitrogen at-
mosphere to the silver salt.After 46 h the mixture was filtered over
CaH2/Celite (1:1; vol: 1.5 cm3).The filter cake was extracted with CH Cl2
2
until the organic solution was colorless.The solvent was removed with a
steady flow of N2 and finally by using high vacuum.A stem solution of
the activated catalyst was prepared by dissolving the solid in dry CHCl3
(1.00 mL; accounts for 14.9 mmolLꢁ1 of the dimer 12a in its activated
rac-N-(1-(Benzyloxy)but-3-en-2-yl)-2,2,2-trifluoro-N-(4-methoxyphen
A
form).03.34 mL of the solution were added to neat
19a (4.008 g,
AHCTREUNG
9.945 mmol) and the solution was stirred in a sealed and shielded flask
for 3 d at 558C in the absence of light under nitrogen atmosphere.The
residue was suspended in cyclohexane/EtOAc 9:1 and subsequently puri-
fied by filtration over silica gel (cyclohexane/EtOAc 9:1) to obtain 20a
(3.844 g, 9.540 mmol, 96%, ee=97%).
sponge in mesitylene (13 mL) was heated to 1608C for 18 h.After solvent
removal the crude product was submitted to column chromatography
(pentane/EtOAc 9:1) to provide racemic 20c as a colorless oil (104 mg,
0.274 mmol, 70%). 1H NMR (300 MHz, CDCl3, 218C): d=7.39–7.28 (m,
2H, arom. H), 7.22–7.19 (m, 1H, arom. H), 7.10–7.07 (m, 1H, arom. H),
6.88–6.81 (m, 2H, arom. H), 5.71–5.60 (m, 1H, CH2=CH), 5.39–5.25 (m,
2H, CH2=CH, CH-N), 4.53 (q, J=15.0 Hz, 2H, CH2Ph), 3.83 (s, 3H, Ar-
OCH3), 3.50 ppm (d, J=7.2 Hz, 2H, BnO-CH2); 13C NMR (75 MHz,
CDCl3, 218C): d=159.8, 157.2 (q, J=34.6 Hz, C-CF3), 137.6, 131.9, 131.6,
131.4, 128.4, 127.7, 120.3, 116.4 (q, J=286.5 Hz, CF3), 113.7, 113.6, 72.9,
68.2, 59.8, 55.5 ppm; 19F NMR (282 MHz, CDCl3, 218C): d=ꢁ67.2 ppm;
IR (film): n˜ =3066, 3032, 2939, 2864, 1694, 1608, 1513, 1455, 1420, 1300,
1252, 1192, 1152, 1108, 1031, 995 cmꢁ1; HRMS (EI): m/z: calcd for
C20H20F3NO3: 379.1390 [M]+; found: 379.1393; elemental analysis calcd
(%) for C20H20F3NO3 (379.37): C 63.32, H 5.31, N 3.69. Found: C 63.60,
H 5.36, N 3.77.
rac-2,2,2-Trifluoro-N-(4-methoxyphenyl)-N-(1-(tetrahydro-2H-pyran-2-
yloxy)but-3-en-2-yl)acetamide (20b):
0.592 mmol) and proton sponge (25 mg, 0.12 mmol, 0.20 equiv) in
mesitylene (15 mL) was heated to 1608C for 22 h.After solvent removal
A solution of 19b (221 mg,
ACHTREUNG
the crude product was submitted to column chromatography (cyclohex-
ane/EtOAc 9:1) to provide racemic 20b as a colorless oil (174 mg,
0.466 mmol, 79%, 1:1 diastereomeric mixture). Analytical data are pro-
1
vided for the diastereomeric mixture: H NMR (300 MHz, CDCl3, 218C):
d=7.46–7.31 (m, 1H, arom. H), 7.22–7.05 (m, 3H, arom. H), 6.92–6.84
(m, 4H, arom. H), 5.80–5.58 (m, 2H, CH2=CH), 5.33–5.14 (m, 6H, CH2=
CH, CH-N), 4.68 (t, J=3.3 Hz, 1H, O-CH-O), 4.53 (t, J=3.3 Hz, 1H, O-
CH-O), 3.87–3.37 (m, 14H, Ar-OCH3, N-CH-CH2, OCH2-CH2), 1.90–
1.51 ppm (m, 12H, CH-CH2-CH2-CH2); 13C NMR (75 MHz, CDCl3,
218C): d=159.8, 157.1 (q, J=34.6 Hz, C-CF3), 157.0 (q, J=34.6 Hz, C-
CF3), 132.0, 131.9, 131.7, 131.5, 131.4, 131.0, 128.3, 127.8, 120.3, 120.0,
116.4 (q, J=286.5 Hz, CF3), 116.3 (q, J=286.5 Hz, CF3), 113.7, 98.6, 98.2,
65.4, 65.0, 62.4, 62.0, 61.2, 59.5, 30.6, 25.5, 19.5, 19.2 ppm; 19F NMR
(282 MHz, CDCl3, 218C): d=ꢁ67.7, ꢁ67.8 ppm; IR (film): n˜ =2945,
2873, 1694, 1609, 1514, 1467, 1444, 1418, 1354, 1300, 1252, 1203, 1152,
1076, 1035, 975 cmꢁ1; HRMS (EI): m/z: calcd for C18H22F3NO: 373.1496
[M]+; found: 373.1497; elemental analysis calcd (%) for C18H22F3NO4
(373.37): C 57.90, H 5.94, N 3.75. Found: C 57.63, H 6.03, N 3.85.
(R)-N-(1-(Benzyloxy)but-3-en-2-yl)-2,2,2-trifluoro-N-(4-methoxyphen
acetamide (20c): GP 5 (Table 7, entry 8): colorless oil, yield: 99%, ee:
96%.The ee value was determined by chiral column HPLC: Chiralcel
OD-H, n-hexane/iPrOH 99:1, 0.8 mLminꢁ1, detection at 210 nm.[ a]2D5
A
AHCTREUNG
=
(c=1.89 gdLꢁ1, CHCl3)=ꢁ32.3; 1H NMR (300 MHz, CDCl3, 218C): d=
7.39–7.28 (m, 5H, arom. H), 7.22–7.19 (m, 1H, arom. H), 7.10–7.07 (m,
1H, arom. H), 6.88–6.81 (m, 2H, arom. H), 5.71–5.60 (m, 1H, CH2=CH),
5.39–5.25 (m, 2H, CH2=CH, CH-N), 4.57 (d, J=12.0, 1H, CHHPh), 4.48
(d, J=12.0 Hz, 1H, CHHPh), 3.83 (s, 3H, Ar-OCH3), 3.50 ppm (d, J=
7.2 Hz, 2H, BnO-CH2).The other analytical data were in accordance
with the racemate (see above).
2,2,2-Trifluoro-N-(4-methoxyphenyl)-N-((2R)-1-(tetrahydro-2H-pyran-2-
yloxy)but-3-en-2-yl)acetamide (20b): GP 5 (Table 7, entry 7): colorless
oil, yield: 94%, ee: 98%, dr 1:1.The ee value was determined by chiral
column HPLC after hydrolysis of the amide and THP group (see below)
to give 2-(4-methoxyphenylamino)but-3-en-1-ol: Chiralcel OD-H, n-
hexane/iPrOH 90:10, 0.8 mLminꢁ1, detection at 210 nm.[ a]2D5: not deter-
mined since mixture of diastereomers. 1H NMR (300 MHz, CDCl3,
218C): d=7.46–7.31 (m, 1H, arom. H), 7.22–7.05 (m, 3H, arom. H),
6.92–6.84 (m, 4H, arom. H), 5.80–5.58 (m, 2H, CH2=CH), 5.33–5.14 (m,
6H, CH2=CH, CH-N), 4.68 (t, J=3.3 Hz, 1H, O-CH-O), 4.53 (t, J=
3.3 Hz, 1H, O-CH-O), 3.87–3.37 (m, 14H, Ar-OCH3, N-CH-CH2, OCH2-
CH2), 1.90–1.51 ppm (m, 12H, CH-CH2-CH2-CH2).The other analytical
data were in accordance with the racemate (see above).
rac-Methyl
4-(2,2,2-trifluoro-N-(4-methoxyphenyl)acetamido)hex-5-
2(CH3CN)2
enoate (32a): A solution of 24 (109 mg, 0.303 mmol) and PdCl
(8 mg, 0.03 mmol, 0.1 equiv) in CH2Cl2 (2 mL) was stirred for 75 min.
The solvent was removed and the residue was purified by column chro-
ACHTREUNG
matography (cyclohexane/EtOAc 9:1) giving 32a as
a colorless oil
(53 mg, 0.15 mmol, 49%). 1H NMR (300 MHz, CDCl3, 218C): d=7.12–
7.03 (m, 2H, arom. H), 6.90–6.86 (m, 2H, arom. H), 5.62–5.50 (m, 1H,
CH2=CH), 5.31–5.22 (m, 2H, CH2=CH), 4.98 (t, J=6.9 Hz, 1H, CH-N),
3.83 (s, 3H, Ar-OCH3), 3.67 (s, 3H, COOCH3), 2.36 (t, J=7.5 Hz, 2H,
CH2-COOMe), 2.05–1.93 (m, 1H, NCH-CH2-CH2), 1.90–1.77 ppm (m,
1H, NCH-CH2-CH2); 13C NMR (75 MHz, CDCl3, 218C): d=172.9, 159.9,
156.8 (q, J=34.8 Hz, C-CF3), 134.3, 131.7 and 130.9, 127.7, 120.2, 116.3
(q, J=288.1 Hz, CF3), 114.0 and 113.8 (hindered rotation of the amide
bond), 60.5, 55.5, 51.8, 30.9, 26.9 ppm; 19F NMR (282 MHz, CDCl3,
218C): d=ꢁ67.2 ppm; IR (film): n˜ =2955, 2843, 1739, 1694, 1514, 1439,
1421, 1358, 1299, 1254, 1205, 1152, 1034, 997, 940 cmꢁ1; HRMS (EI): m/z:
calcd for C16H18F3NO4: 345.1182 [M]+; found: 345.1184; elemental analy-
sis calcd (%) for C16H18F3NO4 (345.31): C 55.65, H 5.25, N 4.06; found: C
55.56, H 5.62, N 4.06.
(R)-2-(4-Methoxyphenylamino)but-3-en-1-ol
Method A: TBAF (1m in THF, 0.12 mmol, 1.1 equiv, 122 mL) was added
to a solution of 20a (45.1 mg, 0.112 mmol) in THF (2 mL). After stirring
for 45 min the reaction mixture was quenched by the addition of water,
extracted with CH2Cl2 and dried over Na2SO4.Column chromatography
(cyclohexane/EtOAc 4:1, 3% NEt3
NEt3) gave 2-(4-methoxyphenylamino)but-3-en-1-ol as a slightly brown
oil (12.1 mg, 0.0418 mmol, 37%).
!
cyclohexane/EtOAc 2:1, 3%
(S)-Methyl
4-(2,2,2-trifluoro-N-(4-methoxyphenyl)acetamido)hex-5-
enoate (32a): GP 5 (Table 7, entry 1): colorless oil, yield: 98%, ee: 98%.
The ee value was determined by chiral column HPLC: Chiralcel OD-H,
n-hexane/iPrOH 99.5:0.5, 0.8 mLminꢁ1, detection at 210 nm.[ a]D25 =(c=
Method B: A solution of 20b (51 mg, 0.140 mmol) and TsOH*H2O
(5 mg, 0.03 mmol, 0.2 equiv) in MeOH (5 mL) was heated to 558C for
2.5 h and the solvent was subsequently removed at reduced pressure.
0.885 gdLꢁ1, CHCl3)=ꢁ1.7; H NMR (300 MHz, CDCl3, 218C): d=7.12–
1
Chem. Eur. J. 2008, 14, 1430 – 1444
ꢀ 2008 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
www.chemeurj.org
1441

R.Peters et al.
7.03 (m, 2H, arom. H), 6.90–6.86 (m, 2H, arom. H), 5.62–5.50 (m, 1H,
CH2=CH), 5.31–5.22 (m, 2H, CH2=CH), 4.98 (t, J=6.9 Hz, 1H, CH-N),
3.83 (s, 3H, Ar-OCH3), 3.67 (s, 3H, COOCH3), 2.36 (t, J=7.5 Hz, 2H,
CH2-COOMe), 2.05–1.93 (m, 1H, NCH-CH2-CH2), 1.90–1.77 ppm (m,
1H, NCH-CH2-CH2).The other analytical data were in accordance with
the racemate (see above).
submitted to column chromatography (cyclohexane/EtOAc 9:1) giving
32d as
a colorless, highly viscous oil (160 mg, 0.316 mmol, 85%).
1H NMR (300 MHz, CDCl3, 218C): d=7.33–7.20 (m, 5H, Ph), 7.01 (d,
J=8.7 Hz, 2H, arom. H), 6.86 (d, J=8.7 Hz, 2H, arom. H), 5.58–5.46 (m,
1H, CH2=CH), 5.25–5.17 (m, 2H, CH=CH2), 5.00–4.93 (m, 1H, N-CH),
4.39 (b, 2H, Ph-CH2), 3.83 (s, 3H, O-CH3), 3.15 (b, 2H, N-CH2-CH2),
1.52 (b, 2H, CH2-CH2-CH2), 1.46 ppm (b, 9H, C-CH3); 13C NMR
(75 MHz, CDCl3, 218C): d=159.8, 156.8 (q, J=34.8, C-CF3), 155.9 and
155.5 (hindered rotation of the amide bond), 138.4, 135.0, 131.7, 130.9,
128.4, 127.6, 127.1, 119.5, 116.4 (q, J=288.0 Hz, CF3), 113.8 and 113.7
(hindered rotation of the amide bond), 79.8, 60.6, 55.5, 50.6 and 50.1 (hin-
dered rotation of the amide bond), 46.2, 28.8, 28.5, 24.9. 19F NMR
(282 MHz, CDCl3, 218C): d=ꢁ67.1 ppm; IR (film): n˜ =2976, 2936, 1694,
rac-2,2,2-Trifluoro-N-(4-methoxyphenyl)-N-(6-oxohept-1-en-3-yl)aceta-
mide (32b): A solution of 27 (196 mg, 0.595 mmol) and PdCl2(CH3CN)2
A
(15 mg, 0.060 mmol, 0.10 equiv) in CH2Cl2 (2 mL) was stirred for 30 min.
The solvent was removed at 408C and the residue was submitted to
column chromatography (pentane/EOAc 9:1 ! 4:1) giving 32b as a col-
1
orless oil (124 mg, 0.376 mmol, 63%). H NMR (300 MHz, CDCl3, 218C):
d=7.10–7.00 (m, 2H, arom. H), 6.88–6.84 (m, 2H, arom. H), 5.50 (ddd,
J=15.0, 9.9, 7.8 Hz, 1H, CH2=CH), 5.27–5.17 (m, 2H, CH2=CH), 4.94 (t,
J=7.8 Hz, 1H, NCH) 3.80 (s, 3H, OCH3), 2.46 (t, J=7.2 Hz, 2H, O=C-
CH2), 2.12 (s, 3H, CH3), 1.94–1.71 ppm (m, 2H, CH2-CH2-CH);
13C NMR (75 MHz, CDCl3, 218C): d=207.0, 159.8, 156.9 (q, J=34.8, C-
CF3), 134.6, 131.8, 130.9, 127.5, 120.0, 116.3 (q, J=288.1 Hz, CF3), 60.3,
55.5, 40.1, 30.2, 25.5 ppm; 19F NMR (282 MHz, CDCl3, 218C): d=
ꢁ67.1 ppm; IR (film): n˜ =1694, 1513, 1419, 1361, 1300, 1254, 1204, 1152,
1034, 999, 937 cmꢁ1; HRMS (MALDI): m/z: calcd for C16H18F3NO3Na:
529.2285 [M+Na]+; found: 529.2284; elemental analysis calcd (%) for
C16H18F3NO3 (329.31): C 58.36, H 5.51, N 4.25; found: C 58.45, H 5.70, N
4.26.
1513, 1464, 1417, 1367, 1298, 1252, 1154, 1034, 935 cmꢁ1
; HRMS
(MALDI): m/z: calcd for C27H33F3N2O4Na: 529.2285 [M+Na]+; found:
529.2284; elemental analysis calcd (%) for C18H27NO3 (506.56): C 64.01,
H 6.71, N 5.57. Found: C 64.02, H 6.75, N 5.52.
(S)-tert-Butyl benzyl(4-(2,2,2-trifluoro-N-(4-methoxyphenyl)acetamido)-
hex-5-enyl)carbamate (32d): GP 5 (Table 7, entry 4): colorless oil, yield:
99%, ee: 98%. The ee value was determined by chiral column HPLC:
Chiralcel OD-H, n-hexane/iPrOH 99:1, 0.8 mLminꢁ1
, detection at
210 nm.[ a]2D5 =(c=1.07 gdLꢁ1
,
CHCl3)=ꢁ8.7; 1H NMR (300 MHz,
CDCl3, 218C): d=7.33–7.20 (m, 5H, Ph), 7.01 (d, J=8.7 Hz, 2H, arom.
H), 6.86 (d, J=8.7 Hz, 2H, arom. H), 5.58–5.46 (m, 1H, CH2=CH), 5.25–
5.17 (m, 2H, CH=CH2), 5.00–4.93 (m, 1H, N-CH), 4.39 (b, 2H, Ph-CH2),
3.83 (s, 3H, O-CH3), 3.15 (b, 2H, N-CH2-CH2), 1.52 (b, 2H, CH2-CH2-
CH2), 1.46 ppm (b, 9H, C-CH3).The other analytical data were in ac-
cordance with the racemate (see above).
(S)-2,2,2-Trifluoro-N-(4-methoxyphenyl)-N-(6-oxohept-1-en-3-yl)aceta-
mide (32b): GP 5 (Table 7, entry 2): colorless oil, yield: 97%, ee: 97%.
The ee value was determined by chiral column HPLC: Chiralcel OD-H,
n-hexane/EtOH 98.2:1.8, 0.8 mLminꢁ1, detection at 210 nm.[ a]=(c=
1.00 gdLꢁ1, CHCl3)=ꢁ2.4; 1H NMR (300 MHz, CDCl3, 218C): d=7.10–
7.00 (m, 2H, arom. H), 6.88–6.84 (m, 2H, arom. H), 5.50 (ddd, J=15.0,
9.9, 7.8 Hz, 1H, CH2=CH), 5.27–5.17 (m, 2H, CH2=CH), 4.94 (t, J=
7.8 Hz, 1H, NCH) 3.80 (s, 3H, OCH3), 2.46 (t, J=7.2 Hz, 2H, O=C-
CH2), 2.12 (s, 3H, CH3), 1.94–1.71 ppm (m, 2H, CH2-CH2-CH).The
other analytical data were in accordance with the racemate (see above).
Acknowledgements
This work was financially supported by ETH research grant TH-01 07-1
and F.Hoffmann-La Roche.We thank Reuter Chemische Apparatebau
KG in Freiburg, Germany, for the generous donation of enantiomerically
pure 1,2-diaminocyclohexane and Raphal Rochat and Lorenz Urner for
skilful experimental contributions.
rac-N-(6-(Benzyloxy)hex-1-en-3-yl)-2,2,2-trifluoro-N-(4-methoxyphen
acetamide (32c): A solution of 29 (283 mg, 0.695 mmol) and PdCl2-
(CH3CN)2 (16 mg, 0.70 mmol, 0.10 equiv) in CH2Cl2 (3 mL) was stirred
yl)-
ACHTREUNG
ACHTREUNG
for 60 min.The solvent was removed at 40 8C and the residue was submit-
ted to column chromatography (pentane/EtOAc 9:1) giving 32c as a col-
1
orless oil (238 mg, 0.584 mmol, 84%). H NMR (300 MHz, CDCl3, 218C):
d=7.37–7.25 (m, 5H, Ph), 7.13–7.03 (m, 2H, arom. H), 6.88–6.81 (m,
2H, arom. H), 5.63–5.51 (m, 1H, CH2=CH), 5.29–5.19 (m, 2H, CH2=
CH), 5.04 (q, J=7.5 Hz, 1H, N-CH), 4.49 (s, 2H, Ph-CH2), 3.82 (s, 3H,
CH3), 3.49 (t, J=5.7 Hz, 2H, CH2-OBn), 1.77–1.53 ppm (m, 4H, CH-
CH2-CH2); 13C NMR (75 MHz, CDCl3, 218C): d=160.0, 156.8 (q, J=
34.7 Hz, C-CF3), 138.4, 135.1, 131.8, 131.1, 128.3, 127.5, 119.4, 116.4 (q,
J=288.8 Hz, CF3), 113.7, 113.6, 72.8, 69.5, 60.4, 55.3, 28.2, 26.4 ppm;
19F NMR (282 MHz, CDCl3, 218C): d=ꢁ67.1 ppm; IR (film): n˜ =3065,
3031, 2938, 2860, 1694, 1608, 1513, 1455, 1418, 1361, 1299, 1253, 1204,
[1] a) Review about enantioselective aza-Claisen rearrangements: T.K.
Hollis, L.E. Overman, J. Organomet. Chem. 1999, 576, 290–299;
b) review about the rearrangement of allylic trihaloacetimidates:
L.E.Overman, N.E.Carpenter, Org. React. 2005, 66, 1–107.
[2] a) C.E. Anderson, L.E. Overman,
J. Am. Chem. Soc. 2003, 125,
12412–12413; b) S.F.Kirsch, L.E.Overman, M.P.Watson,
Chem. 2004, 69, 8101–8104; c) H.Nomura, C.J.Richards,
Eur. J. 2007, 13, 10216–10224.
J. Org.
Chem.
[3] a) LE. .Overman, CE. .Owen, M.M.Pavan, CJ..Richards,
Org.
1187, 1151, 1109, 1034, 996, 936 cmꢁ1
; HRMS (EI): m/z: calcd for
Lett. 2003, 5, 1809–1812; b) R.S. Prasad, C.E. Anderson, C.J. Ri-
chards, L.E. Overman, Organometallics 2005, 24, 77–81; c) C. E.
C22H24F3NO3: 407.1703 [M]+; found: 407.1703; elemental analysis calcd
(%) for C22H24F3NO3 (407.43): C 64.85, H 5.94, N 3.44; found: C 64.85, H
5.91, N 3.46.
Anderson, Y.Donde, C.J.Douglas, L.E.Overman,
J. Org. Chem.
2005, 70, 648–652; d) R. Peters, Z-.q. Xin, D.F. Fischer, W.B.
Schweizer, Organometallics 2006, 25, 2917–2920; e) M.E. Weiss,
DF..Fischer, Z-q. .Xin, S.Jautze, WB. .Schweizer, R.Peters,
(S)-N-(6-(Benzyloxy)hex-1-en-3-yl)-2,2,2-trifluoro-N-(4-methoxyphen
A
ACHTREUNG
97%.The ee value was determined by chiral column HPLC: Chiralcel
Angew. Chem. 2006, 118, 5823–5827; Angew. Chem. Int. Ed. 2006,
OD-H, n-hexane/iPrOH 99:1, 0.8 mLminꢁ1, detection at 210 nm.[ a]2D5
=
45, 5694–5698; f) D.F.Fischer, Z-.q.Xin, R.Peters,
Int. Ed. 2007, 46, 7704–7707.
Angew. Chem.
(c=0.910 gdLꢁ1, CHCl3)=ꢁ27.2; H NMR (300 MHz, CDCl3, 218C): d=
7.37–7.25 (m, 5H, Ph), 7.13–7.03 (m, 2H, arom. H), 6.88–6.81 (m, 2H,
arom. H), 5.63–5.51 (m, 1H, CH2=CH), 5.29–5.19 (m, 2H, CH2=CH),
5.04 (q, J=7.5 Hz, 1H, N-CH), 4.49 (s, 2H, Ph-CH2), 3.82 (s, 3H, CH3),
3.49 (t, J=5.7 Hz, 2H, CH2-OBn), 1.77–1.53 ppm (m, 4H, CH-CH2-CH2).
The other analytical data were in accordance with the racemate (see
above).
1
[4] O.Mumm, F.Mçller, Ber. Dtsch. Chem. Ges. 1937, 70, 2214–2227.
[5] a) L.E. Overman, J. Am. Chem. Soc. 1974, 96, 597–599; b) L.E.
Overman, J. Am. Chem. Soc. 1976, 98, 2901–2910; c) L.E. Over-
man, Angew. Chem. 1984, 96, 565–573; Angew. Chem. Int. Ed. Engl.
1984, 23, 579–586.
[6] An alternative approach is the regioselective allylic amination; se-
rac-tert-Butyl benzyl(4-(2,2,2-trifluoro-N-(4-methoxyphenyl)acetamido)-
hex-5-enyl)carbamate (32d): A solution of 31 (188 mg, 0.371 mmol) and
lected examples: a) T.Ohmura, JF. .Hartwig,
J. Am. Chem. Soc.
2002, 124, 15164–15165; b) C.A. Kiener, C. Shu, C. Incarvito, J.F.
Hartwig, J. Am. Chem. Soc. 2003, 125, 14272–14273; c) C.Shu, A.
Leitner, J.F.Hartwig, Angew. Chem. 2004, 116, 4901–4904; Angew.
PdCl2
A
stirred for 60 min.The solvent was removed at 40 8C and the residue was
1442
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1430 – 1444

Enantiopure Allylic Amines
FULL PAPER
Chem. Int. Ed. 2004, 43, 4797–4800; d) D.Polet, A.Alexakis, K.
Tissot-Croset, C.Corminboeuf, K.Ditrich, Chem. Eur. J. 2006, 12,
ni, G.Polidori, R.Spagna, J. Appl. Crystallogr. 1999, 32, 115–119)
and refined by full-matrix least-squares analysis (SHELXL-97;
G.M.Sheldrick, SHELXL-97, Program for the Refinement of Crys-
tal Structures, University of Gçttingen (Germany), 1997), using an
isotropic extinction correction.All non-H atoms were refined aniso-
tropically, H atoms isotropically, whereby H-positions are based on
stereochemical considerations.X-ray crystal structure of 11a.Crys-
tal data at 220(2) K for C54H46FeN4O4S2, Mw =934.92; orthorhombic
3596–3609; e) R.Weihofen, O.Tverskoy, G.Helmchen,
Angew.
Chem. 2006, 118, 5673–5676; Angew. Chem. Int. Ed. 2006, 45, 5546–
5549; f) R.Takeuchi, S.Kezuka, Synthesis 2006, 3349–3366; g) O.V.
Singh, H.Han, J. Am. Chem. Soc. 2007, 129, 774–775; h) C.Defieb-
er, M.A.Ariger, P.Moriel, E.M.Carreira,
Angew. Chem. 2007, 119,
3200–3204; Angew. Chem. Int. Ed. 2007, 46, 3139–3143; i) Y.Yama-
shita, A.Gopalarathnam, J.F.Hartwig, J. Am. Chem. Soc. 2007, 129,
7508–7509; j) S.Spiess, C.Berthold, R.Weihofen, G.Helmchen,
Org. Biomol. Chem. 2007, 5, 2357–2360; k) G.Helmchen, A.
space group P212121 (no.19),
11.1108(2), b=11.3676(2), c=35.4662(7) Š, V=4479.49(14) Š3.
Linear crystal dimensions ca. 0.21”0.19”0.17 mm, m=0.483 mmꢁ1
1calcd =1.386 gcmꢁ3
, Z=4, a=
.
Final R(F)=0.042, wR(F2)=0.097 for 635 parameters and 6264 re-
flections with I > 2s(I) and 2.49 < q < 26.018 (corresponding R
values based on all 7194 reflections are 0.054 and 0.105, respective-
ly).CCDC 617968 contains the supplementary crystallographic 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.X-ray crystal structure of 12a.Crystal data at
220(2) K for C108H88Cl4Fe2N8O8Pd4S4·7CHCl3, Mw =3268.91; ortho-
rhombic space group P21 (no.4), 1calcd =1.392 gcmꢁ3, Z=2, a=
15.5124(3), b=30.3423(9), c=15.8007(3) Š, b=118.956(1), V=
6507.4(3) Š3. Linear crystal dimensions ca. 0.14”0.12”0.10 mm, m=
1.392 mmꢁ1.Final R(F)=0.049, wR(F2)=0.115 for 1499 parameters,
1 restraint, and 21058 reflections with I > 2s(I) and 2.95 < q <
26.028 (corresponding R values based on all 23932 reflections are
0.059 and 0.122, respectively). CCDC 617969 contains the supple-
mentary crystallographic data for this paper.These data can be ob-
tained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif
Dahnz, P.Dübon, M.Schelwies, R.Weihofen,
2007, 675–691.
[7] M.Calter, T.K.Hollis, L.E.Overman, J.Ziller, G.G.Zipp,
Chem. Commun.
J. Org.
Chem. 1997, 62, 1449–1456.
[8] Recent reviews about palladacycles: a) W.A. Herrmann, V.P.W.
Bohm, C.P.Reisinger, J. Organomet. Chem. 1999, 576, 23–41; b) M.
Albrecht, G.van Koten, Angew. Chem. 2001, 113, 3866–3898;
Angew. Chem. Int. Ed. 2001, 40, 3750–3781; c) J.Dupont, M.Pfeff-
er, J.Spencer, Eur. J. Inorg. Chem. 2001, 1917–1927; d) R.B. Bed-
ford, Chem. Commun. 2003, 1787–1796; e) M.E.van der Boom, D.
Milstein Chem. Rev. 2003, 103, 1759–1792; f) J.T. Singleton, Tetra-
hedron 2003, 59, 1837–1857; g) F.Bellina, A.Carpita, R.Rossi, Syn-
thesis 2004, 2419–2440; h) R.B.Bedford, C.S.J.Cazin, D.Holder,
Coord. Chem. Rev. 2004, 248, 2283–2321; i) I.Omae, Coord. Chem.
Rev. 2004, 248, 995–1023; j) I.P.Beletskaya, A.V.Cheprakov, J. Or-
ganomet. Chem. 2004, 689, 4055–4082; k) V.V.Dunina, O.N.Goru-
nova, Russ. Chem. Rev. 2004, 73, 309–350; l) J.Dupont, C.S.Con-
sorti, J.Spencer, Chem. Rev. 2005, 105, 2527–2572.
[9] For other applications of imidazolines in asymmetric catalysis see
ref.[3d–f] and for example, a) F.Menges, M.Neuburger, A.Pfaltz,
Org. Lett. 2002, 4, 4713–4716; b) C.A. Busacca, D. Grossbach,
[16] Even few direct diastereoselective monocyclopalladations of chiral,
enantiopure ferrocene derivatives are known: see ref.[3d,e] and
a) L.G. Kuz ’mina, Y.T. Struchkov, L.L. Troitskaya, V.I. Sokolov,
O.A. Reutov, Isv. Akad. Nauk SSSR, Ser. Khim. 1979, 1528–1534;
R.C.So, E.M.OꢁBrien, E.M.Spinelli,
c) S.Bhor, G.Anilkumar, M.K.Tse, M.Klawonn, C.Dobler, B.Bit-
terlich, A.Grotevendt, M.Beller, Org. Lett. 2005, 7, 3393–3396;
Org. Lett. 2003, 5, 595–598;
b) C.López, R.Bosque, X.Solans, M.Font-Bardia,
Tetrahedron:
Asymmetry 1996, 7, 2527–2530; c) G.Zhao, F.Xue, Z-.Y.Zhang,
T.C.W.Mak, Organometallics 1997, 16, 4023–4026; d) Z.Gang, Q-.
d) K.Ma, J.You, Chem. Eur. J. 2006, 12, 1863–1871; e) T.Arai, T.
Mizukami, A.Yanagisawa, Org. Lett. 2007, 9, 1145–1147; f) S.En-
thaler, B.Hagemann, S.Bhor, G.Anilkumar, M.K.Tse, B.Bitter-
lich, K.Junge, G.Erre, M.Beller, Adv. Synth. Catal. 2007, 349, 853–
860.
G.Wang, TC. W. .Mak,
1561; e) G.Zhao, Q-.G.Wang, T.C.W.Mak,
Tetrahedron: Asymmetry 1998, 9, 1557–
Organometallics 1998,
17, 3437–3443; f) J.-B. Xia, S.-L. You, Organometallics 2007, 26,
4869–4871.
[10] For the first preparation of chiral ferrocenyl imidazolines see: a) R.
Peters, D.F.Fischer, Org. Lett. 2005, 7, 4137–4140.For further prep-
arations and/or applications see references [3d–f] and b) J.Ma, X.
Cui, L.Gao, Y.Wu, Inorg. Chem. Commun. 2007, 10, 762–766; c) J.
Ma, X.-L. Cui, B. Zhang, M.-P. Song, Y.-J. Wu, Tetrahedron 2007, 63,
5529–5538.
[11] Preliminary results were reported as a communication: S.Jautze, P.
Seiler, R.Peters, Angew. Chem. 2007, 119, 1282–1286; Angew.
Chem. Int. Ed. 2007, 46, 1260–1264.
[17] J. Kang, K.H. Yew, T.H. Kim, D.H. Choi,
Tetrahedron Lett. 2002,
43, 9509–9512.
[18] The stereodescriptors with regard to the planar chirality are used ac-
cording to K.Schlçgl, Top. Stereochem. 1967, 1, 38–39.We employ
different superscripts to distinguish the four Cp ligands.( S’p) indi-
cates the configuration of a bottom Cp while an * indicates a second
ferrocene unit.
[19] The seven diastereomers would have the following configurations:
(Sp,S’p,S*p,S*’p), (Rp,S’p,S*p,S*’p), (Rp,R’p,S*p,S*’p), (Rp,S’p,R*p,S*’p),
(Rp,S’p,S*p,R*’p), (Rp,R’p,R*p,S*’p) and (Rp,R’p,R*p,R*’p).
[12] Recent reviews about applications of ferrocene in asymmetric catal-
ysis: a) R.Gómez Arrayµs, J.Adrio, J.C.Carretero,
Angew. Chem.
[20] Using the same enantiomerically pure catalyst, E- and Z-configured
substrates generally give the opposite major enantiomers.For that
reason it is important to have isomerically pure substrates to get
maximum ee values. 13a used for the catalyst screening experiments
contained 2% of the E-configured isomer.
[21] Attempts to activate the side product formed during the biscyclopal-
ladation resulted in oxidative decomposition of the complex mixture
and the formation of silver metal.
2006, 118, 7836–7878; Angew. Chem. Int. Ed. 2006, 45, 7674–7715;
b) R.C.J.Atkinson, V.C.Gibson, N.J.Long,
33, 313–328.
[13] K.Hamamura, M.Kita, M.Nonoyama, J.Fujita,
Chem. Soc. Rev. 2004,
J. Organomet.
Chem. 1993, 463, 169–177.The reported original yield was 29% and
the product isolation required two column chromatographic separa-
tions and one recrystallization.By our modifications no column
chromatography is necessary and the product is formed in 54%
yield.
[22] F.R.Hartley, Chem. Soc. Rev. 1973, 2, 163–179.
[23] Selected examples: for N-methylimidazole see for example, ref.
[16b] and a) C.López, R.Bosque, X.Solans, M.Font-Bardꢂa, D.
Tramuns, G.Fern, J.Silver, J. Chem. Soc. Dalton Trans. 1994, 3039–
3046; for pyridine derivatives see for example, b) B.Calmuschi, U.
Englert, Acta Crystallogr. Sect. C 2002, 58, 545–548; c) B.Calmu-
schi, M.Alesi, U.Englert, Dalton Trans. 2004, 12, 1852–1857; d) B.
Calmuschi, U.Englert, CrystEngComm 2005, 7, 171–176.
[14] The pyramidality at a given atom can be expressed by the difference
between 3608 and the sum of the three-bond angles at that atom.
For the present analysis, corresponding values at N(13) and N(43)
are 21 and 168, respectively.For further information of the X-ray
analysis see ref.[15].
[15] X-Ray crystal structure analyses.Bruker-Nonius Kappa-CCD dif-
fractometer, MoKa radiation, l=0.7107 Š, The structures were
solved by direct methods (SIR-97; A.Altomare, M.Burla, M.Ca-
malli, G.Cascarano, C.Giacovazzo, A.Guagliardi, A.G.G.Moliter-
[24] Mass spectroscopic investigations (MALDI) showed as major spe-
cies a monocationic monomeric bispalladacycle still possessing one
Chem. Eur. J. 2008, 14, 1430 – 1444
ꢀ 2008 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
www.chemeurj.org
1443

R.Peters et al.
tosylate unit, but no remaining acetonitrile ligands (determined
mass: 1316.96).
[30] a) P.Deslongchamps, S.Lamothe, H.S.Lin,
Can. J. Chem. 1987, 65,
1298–1307; b) A.Thurner, F.Faigl, L.Tçke, A.Mordini, M.Valac-
chi, G.Reginato, G.Czira, Tetrahedron 2001, 57, 8173–8180; c) P.
[25] TP..Remarchuk, PhD. .Dissertation, University of California,
Irvine CA (USA), 2003, pp.175–235 (Overman group).
[26] Like previous catalysts, 12a is not a suitable catalyst for the Z-con-
figured substrate bearing a Ph substituent R’, since the rearrange-
ment is not only exceedingly slow, but also accompanied by decom-
position.
Ding, M.J.Miller, Y.Chen, P.Helquist, A.J.Oliver, O.Wiest,
Lett. 2004, 6, 1805–1808.
[31] -JD. .Moriggi, LJ..Brown, JL. .Castro, RD.C. .Brown,
Biomol. Chem. 2004, 2, 835–844.
Org.
Org.
[32] For the use of morpholine amides as cost-efficient substitute for the
commonly used Weinreb amides see e.g.: a) R. Peters, P. Waldmeier,
A.Joncour, Org. Process Res. Dev. 2005, 9, 508–512; b) R.Peters,
[27] For the kinetic and computational analysis of the asymmetric rear-
rangement of related 3-monosubstituted allylic trichloroacetimidates
C.Diolez, A.Rolland, E.Manginot, M.Veyrat,
Heterocycles 2007,
see: M.P.Watson, L.E.Overman, R.G.Bergman,
J. Am. Chem.
72, 255–273.
Soc. 2007, 129, 5031–5044.
[33] Before short pad silica gel filtration the 1H NMR spectra were too
complex for an accurate dr determination.
[28] For a confirmation of this mechanistic hypothesis see ref.[3f].
[29] A study of the tolerance of various functional groups in the rear-
rangement of trichloroacetimidates has been reported: C.E.Ander-
son, L.E.Overman, J. Am. Chem. Soc. 2003, 125, 12412–12413.
Received: October 16, 2007
Published online: December 20, 2007
1444
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1430 – 1444