Asymmetric synthesis of heterobimetallic planar chiral ferrocene pallada-/platinacycles and their application to enantioselective aza-claisen rearrangements

Article abstract of DOI:10.1021/om300600v

Ligand exchange reactions are usually slower for Pt(II) in comparison to Pd(II) centers. Mainly for that reason Pt(II) catalysts have often shown a reduced catalytic activity as compared to their Pd(II) counterparts. We are interested in the question if t

Full text of DOI:10.1021/om300600v

Article
pubs.acs.org/Organometallics
Asymmetric Synthesis of Heterobimetallic Planar Chiral Ferrocene
Pallada-/Platinacycles and Their Application to Enantioselective Aza-
Claisen Rearrangements
Marcel Weiss, Wolfgang Frey, and Rene Peters*
́
Institut fur Organische Chemie, Universitat Stuttgart, Pfaffenwaldring 55, D-70569 Stuttgart, Germany
̈
̈
S
* Supporting Information
ABSTRACT: Ligand exchange reactions are usually slower
for Pt(II) in comparison to Pd(II) centers. Mainly for that
reason Pt(II) catalysts have often shown a reduced catalytic
activity as compared to their Pd(II) counterparts. We are
interested in the question if this inherently slower ligand
exchange might also provide a chance for heterobimetallic
catalysts to accomplish an improved catalytic performance with
substrates in which the reactive center has a lower binding
constant than an additional Lewis basic moiety. For that purpose we have prepared the first diastereo- and enantiomerically pure
mixed pallada-/platinacycles based on ferrocene. These complexes have been prepared by sequential direct diastereoselective
cycloplatination and cyclopalladation. The investigation of the asymmetric aza-Claisen rearrangement of Z-configured
trifluoroacetimidates showed that a heterodinuclear Pt−Pd bis-metallacycle is an excellent catalyst for this reaction type, in
general allowing for very high enantioselectivities. Moreover, at a slightly elevated temperature (55 °C), the heterodinuclear
platina-/palladacycle could in certain cases outperform the corresponding bis-Pd complex, previously known to be the by far
most active highly enantioselective catalyst for the rearrangement of Z-configured trifluoroacetimidates. This effect, which might
be surprising at first sight due to the low efficiency of other Pt catalysts for aza-Claisen rearrangements, might be explained by an
enhanced lifetime of a productive monodentate olefin coordination of the substrate at the Pt center due to slower ligand
exchange processes.
INTRODUCTION
■
The cooperation of two metal centers, which is well-known for
a number of dinuclear metalloenzymes such as DNA
polymerases, phosphatases, or ureases,1 constitutes an intrigu-
ing design principle for artificial catalysts.2 The bimetallic
catalysts can be classified as homo- and heterobimetallic
complexes. The former are generally more readily available,
because both metals can often be introduced in the same step
by a double metalation, whereas the latter frequently require a
larger synthetic effort, because two different metalation steps
are necessary. Despite this inherent disadvantage, hetero-
bimetallic complexes obviously offer a large potential as dual
activation catalysts. This is due to an enhanced variability for an
optimization of the cooperation of both metals. In that way the
complementary properties of the different Lewis acid centers
can be employed most efficiently for the simultaneous
activation of both reacting substrates (normally an electrophile
and a nucleophile). In other words, the most advantageous
metals for the activation of each reactant might be selected.
Recently we have reported that planar chiral ferrocene
bisimidazoline bispalladacycles (FBIP, Figure 1)3 act as highly
enantioselective homobimetallic catalysts in different reaction
types such as enantioselective [3,3] rearrangements of allylic
imidates4,5 or Michael additions of various nucleophiles to
enones.6 For the 1,4-addition of α-cyanoacetates to enones,6a
detailed kinetic studies have provided strong evidence for the
Figure 1. Planar chiral bis-palladacycle precatalyst [FBIP-Cl]2, the
activated monomeric catalysts FBIP-X (X− is an anionic ligand), and
monoplatinacycle 1.
intramolecular cooperation of both metals by simultaneous
activation of both substrates.
In addition, we have recently reported the first diaster-
eoselective cycloplatination of an enantiomerically pure
ferrocene derivative using the same ferrocenebisimidazoline
ligand as for the synthesis of the FBIP systems.7 The resulting
Pt complex 1 gave only poor results in allylic imidate
rearrangements.8,9 On the other hand, it enabled the first
enantioselective intramolecular Friedel−Crafts alkylations of
Received: June 29, 2012
Published: August 22, 2012
© 2012 American Chemical Society
6365
dx.doi.org/10.1021/om300600v | Organometallics 2012, 31, 6365−6372

Organometallics
Article
Scheme 1. Synthesis of the Dimeric Planar Chiral Mixed Platina-/Palladacycle [FBIPP-Cl]2 and Formation of the Monomer 2 in
the Presence of [nBu4N]Cl
Figure 2. X-ray crystal structure analysis of dimeric mixed pallada-/platinacycle [FBIPP-Cl]2 (color code C (gray), N (blue), O (red), S (yellow), Fe
(orange), Pd and Pt (magenta), Cl (green); H atoms and included CHCl3 molecules (7 per unit cell) omitted for clarity): (left) view along the
ferrocene axes; (right) view nearly perpendicular to the ferrocene axes.
indoles with internal olefins that are not activated by π-acceptor
substituents.7 This application represents the first reported
highly enantioselective reaction catalyzed by a platinacycle. In
contrast, the related imidazoline mono-10 and bispallada-
cycles3,4 could not be used for the same purpose, demonstrat-
ing the different properties of pallada- and platinacycles for
substrate activation.
Both Pd(II) and Pt(II) are well-known to serve as
carbophilic Lewis acids capable of activating olefins as
electrophiles for a large number of atom- and step-economic
processes,11,12 but they generally offer a complementary
reactivity profile. Ligand exchange processes are usually much
slower for Pt(II) in comparison to Pd(II), often resulting in
sluggish catalytic processes. Another dissimilarity is given by the
reactivity of the σ-alkyl−metal intermediates formed by attack
of suitable nucleophiles at π-olefin complexes. Whereas alkyl−
Pd(II) complexes display a comparatively high tendency to
undergo a β-hydride elimination, their alkyl−Pt(II) counter-
parts prefer to undergo a protonolysis and a β-hydride
elimination pathway is usually significantly slower.12c,d,13 In
addition, the tendency to generate a nucleophile by C−H
activation is arguably more pronounced for Pt(II) than for
Pd(II).12c,d
Due to the different catalytic preferences of Pd(II) and Pt(II)
and the documented synthetic value of pallada- and platina-
cycles, we were interested in combining both metal centers in a
planar chiral heterobismetallacycle to explore their coopera-
tivity. Herein, we report the first synthesis of planar chiral
diastereo- and enantiomerically pure mixed pallada-/platina-
cycles on a ferrocene basis. These complexes have been
prepared from a ferrocene bisimidazoline ligand (FBI, Scheme
1) by sequential direct diastereoselective cycloplatination and
cyclopalladation and were investigated in the catalytic
asymmetric aza-Claisen rearrangement of trifluoroacetimidates
for an initial comparison with the related bispalladacycle FBIP.
6366
dx.doi.org/10.1021/om300600v | Organometallics 2012, 31, 6365−6372

Organometallics
Article
RESULTS AND DISCUSSION
■
The ligand FBI was prepared according to the literature in
three steps starting from ferrocene.3 Cycloplatination was
performed according to our previously published protocol using
Zeise’s salt as the Pt source (Scheme 1).7 This reaction
proceeds with excellent diastereoselectivity, most likely via a
chelate in which Pt is coordinated by two imidazoline-N
donors,3 providing a mixture of a highly strained monomer 1
and the chloride-bridged dimer 12.7,14
This mixture was then applied to a diastereoselective
cyclopalladation with (MeCN)2PdCl2 in the presence of
NaOAc, providing the mixed dimeric bismetallacycle [FBIPP-
1
Cl]2. H NMR spectra of this material are relatively complex
because of the formation of different regioisomers, since
chloride bridges are formed between either two identical or two
different metals. X-ray single-crystal structure analysis confirms
the dimeric nature of the complex15 and shows a structure very
similar to that of [FBIP-Cl]2 (Figure 2).4 The two different
regioisomers (ratio 2:1 as determined by 1H NMR) are
statistically distributed in the crystal, and it is thus not possible
to differentiate Pd and Pt in the structure. The distance
between a Pd and Pt center bound to the same ferrocene
fragment has been determined to be approximately 3.76−3.79
Å.
The crystal structure analysis of [FBIPP-Cl]2 confirmed the
expected all-Sp configuration also found in the corresponding
bispalladacycle.16 The dimeric molecules can be described as
macrocyclic systems comprising two ferrocenyl units and four
chloride-bridged metal ions. The dimeric structure of [FBIPP-
Cl]2 forces two chloride-bridged metal square planes to be
arranged in an almost coplanar fashion, which is otherwise
unusual for dimeric pallada- or platinacycles. Moreover, halide-
bridged pallada- or platinacycles usually form geometric
isomers about the PdII centers.17 In contrast, in the present
case the planar chirality and the 1,1′,2,2′-substitution pattern of
each ferrocene fragment dictate (as for [FBIP-Cl]2) the
coordination sphere about the PdII square planes for geometric
reasons: the all-Sp-configured complexes exist exclusively as all-
trans-configured isomers. That means both imino-type N
atoms, which are positioned in the same (MCl)2 plane, are
always arranged trans to each other.
Figure 3. X-ray crystal structure analysis of monomeric mixed platina-/
palladacycle 2 (color code C (gray), N (blue), O (red), S (yellow), Fe
(orange), Pd and Pt (both magenta), Cl (green)). The unit cell
contains two different conformers, and only one is depicted here. H
atoms and the tetrabutylammonium counterion are also omitted for
clarity.
A monomeric complex is also formed in quantitative yield by
chloride counterion exchange utilizing silver heptafluorobuty-
rate in acetonitrile (Scheme 2).
Scheme 2. Generation of the Monomeric Complex FBIPP-
O2CC3F7 by Chloride Counterion Exchange in the Presence
of MeCN
Treatment of [FBIPP-Cl]2 with [nBu4N]Cl in chloroform
generates monomer 2 with two anionic metal centers.
Unfortunately, the product could not be separated from
unreacted starting material on a preparative scale. However,
the identity of 2 was confirmed by single-crystal X-ray structure
analysis performed on the microcrystals obtained.18 Since the
structure could be refined to only R = 0.1334, an in-depth
discussion of the structural parameters in 2 showing two
independent conformers in the unit cell (Figure 3) is not
possible, but the analysis confirms the constitution and absolute
configuration. Moreover, it is possible to say that the distance
of both soft metal centers is largely increased in the monomeric
species (Pd−Pt distances ca. 6.0 and 6.7 Å for the two
conformers). This large variation of possible metal−metal
distances in ferrocene-derived bismetallacycles is considered to
be a valuable aspect with regard to cooperative bimetallic
catalysis, as each bimetallic reaction should prefer a different
optimal distance of the reacting centers in the corresponding
transition state. With the ferrocene backbone this distance is
readily self-adjustable due to the partial rotational freedom
around the Fe−Cp bond in the monomeric entities.
This transformation confirms that not only the cyclo-
platination but also the cyclopalladation proceeds with almost
complete diastereoselectivity, as only a single isomer is obtained
as judged by 1H NMR. The generated species FBIPP-O2CC3F7
is a monomer, because one MeCN coordinates to each noble
metal center.
To compare the catalytic properties with those of the
established bis-Pd precatalyst [FBIP-Cl]2, we have chosen the
aza-Claisen rearrangement5 of allylic imidates as the initial
application, as part of our program to study cooperative effects
6367
dx.doi.org/10.1021/om300600v | Organometallics 2012, 31, 6365−6372

Organometallics
Article
in asymmetric catalysis.4,6,19 The bis-Pd catalyst is the most
active enantioselective catalyst known so far for the rearrange-
ment of allylic trifluoroacetimidates with a Z configuration of
the olefin moiety.4 Since the aza-Claisen rearrangement
proceeds stereospecifically (E and Z substrates yield different
enantiomers in excess)4,5,20 and since (Z)-olefins are more
readily available in geometrically pure form than (E)-olefins
(e.g., by semihydrogenation of alkynes), rearrangement of the
Z-configured substrates provides a more practical access to
highly enantioenriched allylic amines. However, this rearrange-
ment is also considerably more difficult to achieve than the
analogous transformation of the E isomers. This is explained by
an unfavorable axial orientation of the (Z)-olefin substituent in
the assumed half-chair-like transition state and the ensuing
cyclic intermediate.21
The FBIP system has been demonstrated to be ca. 1−2
orders of magnitude more active than its mono-Pd counterparts
for Z-configured substrates.4 However, there is still a general
problem for Pd-catalyzed aza-Claisen rearrangements causing
low turnover frequencies even in the case of the more reactive
E-configured substrates. Typical values are about just 40−100
h−1 for the most active enantioselective catalysts.10d A major
issue of enantioselective aza-Claisen rearrangements is most
likely the circumstance that the substrate has two possible
coordination sitesthe olefin and the imidate N atomand
that coordination of the imidate N is strongly preferred. The
binding constant of the N atom of trichloroacetimidates has
been reported to be about 104 times higher than the binding
constant of an olefin.21 That means that in a snapshot almost all
of the catalyst would be blocked by an unproductive
coordination (Figure 4). A bimetallic catalyst might change
that situation in a way that coordination of the stronger N
ligand to one metal site might also facilitate the coordination of
the reactive olefin site by a chelating effect. Since both
coordination events are reversible,21 a subsequent attack of the
imidate N atom on the coordinated olefin is still possible
(Figure 4).
Figure 4. Comparison of the general coordination modes of a
bidentate substrate, in which the reactive center (in the case of allylic
imidates the olefin moiety) has a lower binding constant than another
Lewis basic functionality leading to a small degree of productive
coordination with a monometallic catalyst. In a bimetallic catalyst a
chelate effect might increase the binding constant of the reactive
center. However, the efficiency would just be increased in the allylic
imidate rearrangement, if the decomplexation of the imidate N atom is
sufficiently rapid to allow for a sufficiently fast outer-sphere attack at
the coordinated olefin.
With the heterodinuclear Pd−Pt catalyst, higher ligand
exchange rates are expected at the Pd center. However, the
relatively rapid coordination of the imidate N atom at Pd might
also lead to an accelerated ligand exchange at the Pt center by
formation of the bridged substrate complex 3 depicted in
Figure 4.22 Once coordinated to the Pt center, the olefin
exchange might be slower than in the case of the analogous bis-
Pd catalyst while the N ligand still rapidly exchanges. In total
this might lead to a larger lifetime of the complex with a
productive monodentate olefin coordination. To investigate
this idea, we studied the catalytic asymmetric rearrangement of
Z-configured allylic imidate substrates 4. Table 1 shows a
comparison of data obtained with the heterodinuclear Pd−Pt
complex [FBIPP-Cl]2 and the corresponding homodinuclear
Pd2 complex [FBIP-Cl]2. The experiments were conducted
under conditions previously optimized for the bis-Pd
precatalyst using silver tosylate for chloride ligand exchange.
The rearrangements were investigated at 20 and 55 °C. In
general, they are very clean in a way that side products are not
detectable. As a general trend, the bispalladium catalyst is
somewhat more active at 20 °C than the Pt−Pd catalyst
(compare entries 1/2, 7/8, and 14/15). In contrast, at 55 °C
which of the two catalyst systems performs in a superior
manner is dependent on the substrate: e.g., substrate 4b with R
= iBu requires a higher catalyst loading for the heterodinuclear
catalyst (compare entries 9−11) to obtain high product yields.
Similarly, a lower reactivity was also found for the rearrange-
ment of 4c with R = Me (entries 12 and 13). In that case, also
the enantioselectivity was lower with the heterobimetallic
catalyst (88 vs 94% ee). In contrast, for substrate 4a with R =
nPr, a significantly higher activity was noticed with the Pt−Pd
catalyst. Using 0.05 mol % of the precatalyst, the almost
enantiopure product was formed in nearly quantitative yield,
while the bispalladium catalyst gave a yield of just 68% under
otherwise identical conditions. In addition, a higher activity of
the Pt−Pd catalyst was also observed for substrate 4d with R =
(CH2)2Ph (entries 16 and 17). Both catalysts can also be
employed to generate N-substituted quaternary stereocenters
starting from allylic imidates such as 4e carrying residues RE
and RZ (entries 18 and 19) both different from H. The
successful enantioselective application of this challenging
substrate class has been reported with very few cata-
lysts.10b,d,20g,23 While the product was formed in both cases
in a yield of 94% using 2 mol % of the precatalysts, the
enantioselectivity was higher for the Pt−Pd catalyst (87% vs
80% ee).
For a further comparison of the catalytic activity of the
homo- and heterobimetallic complexes, the rearrangement of
substrates 4a,d was monitored starting after a reaction time of 2
h (Figure 5). For both substrates and with both catalysts, the
reactions proceeded in a clean manner (no side products could
be detected at any time). The Pt/Pd catalyst provided the
6368
dx.doi.org/10.1021/om300600v | Organometallics 2012, 31, 6365−6372

Organometallics
Article
Table 1. Comparison of the Catalytic Asymmetric Rearrangement of Allylic Imidates 4 Using the Bis-Pd Catalyst [FBIP-Cl]2 or
the Heterodinuclear Pd−Pt Complex [FBIPP-Cl]2
a
b
entry
M
4
RZ
RE
precat. loading: X (mol %)
T (°C)
t (h)
yield (%)
ee (%)
1
Pd
Pt
Pd
Pt
Pd
Pt
Pd
Pt
Pd
Pt
Pt
Pd
Pt
Pd
Pt
Pd
Pt
Pd
Pt
4a
4a
4a
4a
4a
4a
4b
4b
4b
4b
4b
4c
4c
4d
4d
4d
4d
4e
4e
nPr
nPr
nPr
nPr
nPr
nPr
iBu
iBu
iBu
iBu
iBu
Me
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Me
Me
0.5
0.5
0.1
0.1
0.05
0.05
1
20
20
55
55
55
55
20
20
55
55
55
55
55
20
20
55
55
55
55
72
72
72
72
72
72
72
72
72
72
72
24
48
72
72
72
72
72
72
97
94
94
99
68
99
87
87
86
62
>99
97
78
99
83
90
98
94
94
98
97
97
98
97
98
98
99
98
98
98
94
88
96
95
95
95
2
3
4
5
6
7
8
1.5
0.1
0.1
0.2
0.1
0.1
1.0
1.0
0.2
0.2
2.0
9
10
11
12
13
14
15
16
17
18
19
(CH2)2Ph
(CH2)2Ph
(CH2)2Ph
(CH2)2Ph
CH2OBn
CH2OBn
c
80
c
2.0
87
a
b
c
Yield of isolated product. Determined by HPLC after hydrolysis of the amide. Determined by HPLC without hydrolysis of the amide.
rearrangement product in both cases with a considerably higher
rate under otherwise identical conditions.
CONCLUSION
■
In conclusion, we have reported the first asymmetric synthesis
of planar chiral mixed pallada-/platinacycles based on
ferrocene. These complexes were prepared via sequential direct
cycloplatination and cyclopalladation. Both metalations pro-
ceeded with almost complete diastereoselectivity. In prelimi-
nary studies the Pt−Pd system was investigated in the catalytic
asymmetric aza-Claisen rearrangement of Z-configured trifluor-
oacetimidates for an initial comparison with the related
bispalladacycle FBIP. Although monoplatinacycles were
previously found to be not useful as catalysts for the aza-
Claisen rearrangement of trifluoroacetimidates, the hetero-
dinuclear Pt−Pd complex [FBIPP-Cl]2 proved to be an
excellent precatalyst for this reaction type, in general allowing
for very high enantioselectivities. In particular, at a slightly
elevated temperature, the heterodinuclear bismetallacycle could
for certain substrates outperform the bispalladacycle FBIP,
previously known as the by far most active enantioselective
catalyst for the title reaction using Z-configured allylic
trifluoroacetimidate substrates. This effect might be explained
by an enhanced lifetime of a productive monodentate olefin
coordination of the substrate at the Pt center due to slower
ligand exchange processes in comparison to Pd(II).
The active participation of the Pt center in the catalytic event
is likely as revealed by a comparison with the corresponding
ferrocene monoimidazoline monopalladacycles. The latter are
at least 1 order of magnitude less active for Z-configured
substrates than the bimetallic catalysts, as the monopallada-
cycles require a precatalyst loading of 5 mol % for substrate 4a
in order to obtain high product yields.10a,d The fact that
monoplatinacycle 1 is not a useful catalyst for aza-Claisen
rearrangements of trifluoroacetimidates further supports the
synergy of the Pd and Pt centers.
The activity difference of both catalysts at 20 °C favoring the
homobimetallic catalyst might reflect the traditional issues of
slower substrate coordination/product release at Pt. However,
these issues appear to be less relevant at 55 °C and the higher
activity in certain cases with the heterobimetallic catalyst
supports the initial hypothesis of an improved synergistic action
of both metal centers in the case of Pd and Pt, depending on
the temperature-dependent kinetic profile of each elementary
step. An increased lifetime of the substrate olefin coordination
might thus be beneficial for the catalyst activity at least under
certain conditions (vide supra).
6369
dx.doi.org/10.1021/om300600v | Organometallics 2012, 31, 6365−6372

Organometallics
Article
Figure 5. Comparison of the reactivity using the bis-Pd precatalyst [FBIP-Cl]2 (blue curves) and the Pt/Pd precatalyst [FBIPP-Cl]2 (red curves) for
1
the rearrangement of substrates 4a (top) and 4d (bottom) by monitoring via H NMR of samples taken.
operating at 500 or 300 MHz (1H), 125 or 75 MHz (13C), and 235
EXPERIMENTAL SECTION
General Considerations. All reactions were performed in oven-
■
MHz (19F). Deuterated solvents were used as purchased. IR spectra
were recorded by the IR service of the Universitat Stuttgart on a
̈
dried (150 °C) glassware under a positive pressure of dinitrogen using
standard Schlenk techniques. For catalysis all glassware used (also for
catalyst activation and preparation of stem solutions) was intensively
washed with demineralized water to remove all remaining chloride.
Methanol and benzene were stored in crown-capped bottles under
argon over 4 Å molecular sieves. Acetonitrile and dichloromethane
(DCM) were purified by distillation and subsequently by a solvent
purification system under a dinitrogen atmosphere. n-Hexane (HPLC
grade), 2-propanol (HPLC grade), and chloroform (>99%) were used
as purchased. NMR spectra were recorded at 21 °C on spectrometers
spectrometer with an ATR unit. Optical rotation was measured on a
polarimeter operating at the sodium D line with a 100 mm path cell
length. Melting points were measured in open glass capillaries and are
uncorrected. Mass spectra were obtained from the MS service of the
Universitat Stuttgart. Single-crystal X-ray analyses were performed by
̈
Dr. Wolfgang Frey (Universitat Stuttgart). Enantiomeric excesses
̈
(ee’s) were determined by high-performance liquid chromatography
(HPLC).
6370
dx.doi.org/10.1021/om300600v | Organometallics 2012, 31, 6365−6372

Organometallics
Article
General Procedure for the Enantioselective Aza-Claisen
Rearrangement. Silver p-toluenesulfonate (4.00 equiv) was dissolved
in MeCN (0.1 mL/mg), and the solvent was subsequently removed by
a stream of dinitrogen. A solution of [FBIPP-Cl]2 (1.00 equiv) in
DCM (0.1 mL/mg) was then added under a dinitrogen atmosphere.
The mixture was stirred overnight at room temperature and
subsequently filtered through Celite/CaH2. The filter cake was
extracted with DCM until the organic solution was colorless. The
solvent was removed by a steady stream of dinitrogen and finally by
high vacuum. A stem solution of the activated catalyst was prepared by
dissolving the solid in dry CHCl3 (20 mmol/L). The required amount
of this solution was added to the substrate (prepared according to
Overman et al.20a) under a nitrogen atmosphere. The reaction tube
was sealed, and the reaction mixture was stirred for the indicated time
at the indicated temperature. Afterwards, the reaction mixture was
suspended in petroleum ether/ethyl acetate (10/1) and subsequently
purified by filtration over silica gel.
H), 6.69 (d, J = 7.5, 2H, arom H), 6.45 (d, J = 2.7, 2H, arom H), 5.58
(m, 1H, Cp ), 5.55 (d, J = 2.7, 1H, Cp H), 5.51 (d, J = 2.7, 1H,
CHPh), 5.35 (d, J = 2.7, 1H, CHPh), 5.22 (d, J = 3.3, 1H, CHPh),
5.10 (d, J = 2.7, 2H, CHPh, Cp H), 4.94 (m, 1H, Cp H), 4.68 (m, 1H,
Cp H), 4.62 (dd, J = 2.7, J = 2.4, 1H, Cp H), 4.34 (m, 1H, Cp H), 2.45
(s, 3H, Ts-CH3), 2.33 (s, 3H, Ts-CH3). All further data are in
accordance with the literature.2
Bis{[μ-chloro-[(η5-(4′′R,5′′R)-(1-SP)-2-(2′′-4′′,5′′-dihydro-
4′′,5′′-diphenyl-1′′-tosyl-1′′H-imidazolyl)cyclopentadienyl-
κC1,κN3]palladium(II)][μ-chloro-[(η5-(4′′′R,5′′′R)-(1′-SP)-2′-(2′′′-
4′′′,5′′′-dihydro-4′′′,5′′′-diphenyl-1′′′-tosyl-1′′′H-imidazolyl)-
cyclopentadienyl-κC1′,κN3′]platinum(II)]iron(II)} ([FBIPP-Cl]2).
A mixture of platinacycles 1/12 (249.1 mg, 0.214 mmol, 1.00 equiv),
bis(acetonitrile)dichloropalladium (55.5 mg, 0.214 mmol, 1.00 equiv),
and sodium acetate (34.0 mg, 0.428 mmol, 2.00 equiv) was dissolved
in degassed 1,2-dichloroethane/tBuOH (1/1) (28.5 mL) and the
solution stirred for 7.5 h at 80 °C. After the mixture was cooled to
room temperature, the solvent was removed in vacuo and the reaction
mixture was filtered through silica gel (DCM). The solvent was
subsequently removed under reduced pressure. The product was
obtained as a red solid (161.1 mg, 0.0617 mmol, 58%). The product is
a mixture of two regioisomers (isomeric ratio 1/0.5). [α]D25 (c = 0.020
[FBIP-Cl]2 was prepared according to literature procedures.4
trans-N,N-μ-[(1S)-2-[(4R,5R)-1-(4-Tolylsulfonyl)-4,5-dihydro-
4,5-diphenyl-1H-imidazol-2-yl-κN3]-1′-[(4R,5R)-1-(4-tolylsul-
fonyl)-4,5-dihydro-4,5-diphenyl-1H-imidazol-2-yl-κN3]-1-ferro-
cene-κC1]chloroplatinum(II) (1) and Bis[(1S)-2-[(4R,5R)-1-(4-
tolylsulfonyl)-4,5-dihydro-4,5-diphenyl-1H-imidazol-2-yl-κN3]-
1′-[(4R,5R)-1-(4-tolylsulfonyl)-4,5-dihydro-4,5-diphenyl-1H-
imidazol-2-yl]-1-ferrocene-κC1]-μ-dichlorodiplatinum(II) (12).
(4′′R,5′′R)-1,1'-Bis(2′′-4′′,5′′-dihydro-4′′,5′′-diphenyl-1′′-
tosylimidazolyl)ferrocene4 (FBI; 200.0 mg, 0.214 mmol, 1.00 equiv)
was suspended in benzene (2.20 mL). To this mixture was added a
solution of K[(C2H4)PtCl3] (157.7 mg, 0.428 mmol, 2.00 equiv) and
sodium acetate (70.2 mg, 0.856 mmol, 4.00 equiv) in methanol (2.20
mL), and the resulting solution was stirred for 24 h at room
temperature. Afterwards the reaction mixture was filtered through
Al2O3 (ethyl acetate). Solvent removal under reduced pressure yielded
the product as a red solid (249.1 mg, 0.214 mmol, 100%). This
product is a mixture of the cis/trans dimeric and the monomeric
species 12 and 1 (4.2/1), respectively, mp > 250 °C dec. Data for
1
g/dL, CHCl3) = −225.0°. Mp: >200 °C dec. H NMR (500 MHz,
CD2Cl2, main isomer (*), minor isomer (#)): δ 7.44−7.36 (m, 12H,
arom H), 7.34−7.11 (m, 20H, arom H), 7.06−6.90 (m, 13H, arom H),
6.90−6.84 (m, 3H, arom H), 6.36−6.30 (m, 4H, arom H), 5.14* (d, J
= 2.2, 1H, Cp H), 6.34 (d, J = 7.7, 4H, arom H), 5.12# (d, J = 2.2, 1H,
Cp H), 5.06−5.03 (m, 2H, Cp H), 4.97* (d, J = 2.8, 1H, CHPh),
4.91# (d, J = 3.1, 1H, CHPh), 4.88# (d, J = 3.5, 1H, CHPh), 4.81* (d,
J = 3.8, 1H, CHPh), 4.66−4.63 (m, 2H, CHPh/Cp H), 4.60* (d, J =
2.8, 1H, CHPh), 4.56−4.53# (m, 2H, CHPh/Cp H), 4.49* (d, J = 3.9,
1H, CHPh/Cp H), 4.48−4.46* (m, 1H, Cp H), 4.46−4.44# (m, 1H,
Cp H), 4.38 (d, J = 2.4, 1H, Cp H), 4.37* (d, J = 2.4, 1H, Cp H),
4.29# (d, J = 2.2, 1H, Cp H), 4.27* (d, J = 2.2, 1H, Cp H), 2.41 (s, 3H,
Ts-CH3), 2.39 (s, 3H, Ts-CH3), 2.37 (s, 3H, Ts-CH3). 13C NMR (500
MHz, CD2Cl2): δ 172.3 (2C), 170.4 (2C), 145.8 (2C), 145.6 (2C),
140.9, 140.7 (2C), 140.5 (2C), 140.3, 134.1, 133.9 (2C), 133.8 (2C),
133.7, 130.3 (4C), 130.2 (4C), 129.6 (4C), 129.5 (4C), 128.9, 128.8,
128.7 (2C), 128.6, 128.5, 128.4 (2C), 127.6, 127.3, 127.2 (2C), 127.1,
127.0 (2C), 126.9 (3C), 126.6, 126.5 (2C), 125.7, 125.6 (2C), 125.5
(2C), 125.4, 126.2 (2C), 97.9 (2C), 97.2, 78.1 (2C), 77.7, 77.5, 77.3
(2C), 76.7 (2C), 76.4 (2C), 75.8, 75.5, 74.9 (2C), 74.7 (2C), 74.4
(2C), 73.3, 73.1 (2C), 73.0 (2C), 72.9 (2C), 71.2, 70.9, 70.0 (2C),
1
dimer 12 are as follows. H NMR (300 MHz, CDCl3, mixture of cis
and trans isomers): δ 7.80−6.53 (m, 82H, arom H), 6.38 (d, J = 7.4,
2H, arom H), 5.58 (d, J = 2.4, 1H, Cp H/CHPh), 5.53−5.49 (m, 1H,
Cp H/CHPh), 5.48 (d, J = 2.7, 1H, Cp H/CHPh), 5.44 (d, J = 2.7,
1H, Cp H/CHPh), 5.37−5.35 (m, 1H, Cp H/CHPh), 5.32 (d, J = 2.7,
1H, Cp H/CHPh), 5.27 (d, J = 2.7, 1H, Cp H/CHPh), 5.17−5.13 (m,
1H, Cp H/CHPh), 5.02 (d, J = 2.7, 2H, Cp H/CHPh), 4.98 (d, J =
3.8, 3H, CHPh), 4.96−4.94 (m, 2H, Cp H/CHPh), 4.92 (d, J = 3.8,
2H, CHPh), 4.87−4.85 (m, 1H, Cp H/CHPh), 4.78−4.73 (m, 1H, Cp
H/CHPh), 4.63−4.59 (m, 2H, Cp H/CHPh), 4.59−4.57 (m, 1H, Cp
H/CHPh), 4.56−4.51 (m, 2H, Cp H/CHPh), 4.48−4.40 (m, 3H, Cp
H/CHPh), 4.33−4.29 (m, 1H, Cp H/CHPh), 4.27 (d, J = 2.4, 2H, Cp
H/CHPh), 4.15−4.09 (m, 2H, Cp H/CHPh), 2.37 (s, 6H, 2 × CH3),
2.34 (s, 3H, CH3), 2.31 (s, 3H, CH3), 2.29 (s, 3H, CH3), 2.25 (s, 3H,
CH3). 13C NMR (75 MHz, CDCl3, mixture of cis and trans isomers): δ
183.8 (qC), 176.89 (qC), 171.0 (qC), 166.5 (qC), 163.3 (qC), 158.3
(qC), 146.0, 145.9, 145.2, 145.0, 144.5, 142.5, 141.1, 140.7, 140.6,
140.1, 139.7, 138.7, 134.5, 134.23, 134.20, 133.2, 130.4, 130.3, 130.1,
130.00, 129.97, 129.8, 129.61, 129.56, 129.53, 129.41, 129.39, 129.32,
129.30, 129.26, 129.1, 129.0, 128.9, 128.82, 128.77, 128.72, 128.67,
128.60, 128.57, 128.54, 128.52, 128.47, 128.3, 128.21, 128.17, 128.0,
127.9, 127.7, 127.64, 127.61, 127.59, 127.5, 127.4, 127.32, 127.27,
127.17, 127.1, 126.78, 126.76, 126.4, 125.94, 125.89, 125.87, 125.84,
125.77, 125.63, 125.57, 125.4, 125.1 (CAr), 85.3, 83.1, 82.7, 80.0, 78.1,
78.0, 77.3, 76.9, 76.8, 76.5, 76.1, 76.0, 75.7, 75.3, 74.7, 74.4, 74.3,
74.19, 74.16, 73.5, 73.3, 73.0, 72.9, 72.49, 72.45, 72.36, 72.2, 71.9,
70.84, 70.78, 67.7 (Cp C/HC(N)), 31.0, 21.77, 21.72, 21.68, 21.62,
̃
69.9, 69.0, 68.9 (2C), 68.5 (2C), 21.4 (4C). IR (in CDCl3): ν 3065,
2925, 2362, 2256, 1597, 1551, 1494, 1457, 1362, 1335, 1259, 1170,
1096, 1029, 971, 905, 812, 728, 670, 649, 607, 545 cm−1. MS (ESI):
m/z 2610.04 (100%, [M]+), 1305.00 (51%, [M + H]+). HRMS (ESI):
m/z calcd for [M]+ 2610.0433, found 2610.0427.
{Acetonitrile-[(η5-(4′′R,5′′R)-(1-SP)-2-(2′′-4′′,5′′-dihydro-
4′′,5′′-diphenyl-1′′-tosyl-1′′H-imidazolyl)cyclopentadienyl-
κC1,κN3]heptafluorobutyratopalladium(II)]}{acetonitrile-[(η5-
(4′′′R,5′′′R)-(1′-SP)-2′-(2′′′-4′′′,5′′′-dihydro-4′′′,5′′′-diphenyl-
1′′′ tosyl-1′′′H-imidazolyl)cyclopentadienyl-κC1',κN3']-
heptafluorobutyratoplatinum(II)]}iron(II) (FBIPP-O2CC3F7). A
solution of silver heptafluorobutyrate (4.9 mg, 0.0153 mmol, 4.00
equiv) in MeCN (0.1 mL) was added to [FBIPP-Cl]2 (10.0 mg, 3.83
μmol, 1.00 equiv) under an N2 atmosphere. The mixture was stirred
for 2 h at room temperature. The reaction mixture was then filtered
through Celite, and subsequently the solvent was removed by a stream
22
of N2 and finally by vacuum. [α]D (c = 0.050 g/dL, CHCl3) =
1
+540.0°. Mp: >200 °C dec. H NMR (500 MHz, CDCl3): δ 7.63−
6.98 (m, 24H, arom H), 6.61−6.56 (m, 2H, arom H), 6.56−6.50 (m,
2H, arom H), 5.83−5.77 (m, 1H, CHPh), 5.61−5.55 (m, 1H, CHPh),
5.19−5.14 (m, 1H, CHPh), 5.16−5.13 (m, 1H, Cp H), 5.11−5.07 (m,
1H, Cp H), 5.00−4.96 (m, 1H, Cp H), 4.92−4.88 (m, 1H, Cp H),
4.86−4.80 (m, 2H, Cp H), 4.67−4.62 (m, 1H, Cp H), 4.55−4.53 (m,
1H, CHPh), 2.49 (s, 3H, Ts-CH3), 2.46 (s, 3H, Ts-CH3), 2.01 (s, 3H,
NCCH3 + 2 × solvent NCCH3), 1.73 (s, 3H, NCCH3). 13C NMR
(500 MHz, CD2Cl2): δ 174.7, 172.0, 147.2, 146.9, 140.8, 140.3, 139.9,
139.7, 134.6, 134.4, 131.3, 131.15 (2C), 131.07 (2C), 130.2 (2C),
130.0 (4C), 129.7 (2C), 129.6 (2C), 129.3 (4C), 128.6, 128.4, 128.0
̃
21.59, 21.49 (Ts-CH3). IR (in CDCl3): ν 3033, 2254, 1971, 1596,
1531, 1495, 1453, 1367, 1170, 1090, 1028 cm−1. MS (ESI, only
monomeric species detectable): m/z 1129.19 (100%, [M + H − Cl]+),
935.23 (3%, [M + H − PtCl]+). HRMS (ESI): m/z calcd for [M + H
− Cl]+ 1129.1889, measd 1129.1886. Data for monomer 1 are as
1
follows. H NMR (300 MHz, CDCl3): δ 7.60−7.31 (m, 16H, arom
H), 7.17 (m, 3H, arom H), 7.01 (m, 5H, arom H), 6.89 (m, 2H, arom
6371
dx.doi.org/10.1021/om300600v | Organometallics 2012, 31, 6365−6372

Organometallics
Article
(2C), 127.8, 127.6 (2C), 127.5 (2C), 126.5, 126.3 (2C), 125.9, 125.5
(2C), 22.0 (2C), 3.2, 2.2. 19F NMR (235 MHz, CDCl3): −80.5 to
−81.0 (m, 3F, CF3), −80.6 to −80.2 (m, 3F, CF3), −115.4 to −115.9
(m, 2F, CF3CF2), −116.6 to −117.2 (m, 2F, CF3CF2), −126.6 to
−126.9 (m, 2F, CF2COO), −127.1 to −127.6 (m, 2F, CF2COO). IR
F.; Xin, Z.-q.; Peters, R. Angew. Chem., Int. Ed. 2007, 46, 7704. (c) Xin,
Z.-q.; Fischer, D. F.; Peters, R. Synlett 2008, 1495. (d) Fischer, D. F.;
Barakat, A.; Xin, Z.-q.; Weiss, M. E.; Peters, R. Chem. Eur. J. 2009, 15,
8722. (e) Peters, R.; Xin, Z.-q.; Maier, F. Chem. Asian J. 2010, 5, 1770.
(f) Eitel, S. H.; Bauer, M.; Schweinfurth, D.; Deibel, N.; Sarkar, B.;
̃
(in CDCl3): ν 3033, 2929, 2254, 1672, 1596, 1544, 1494, 1466, 1456,
Kelm, H.; Kruger, H.-J.; Frey, W.; Peters, R. J. Am. Chem. Soc. 2012,
̈
1374, 1333, 1305, 1261, 1211, 1189, 1171, 1118, 1083, 1053, 967, 931,
906, 814, 758, 719, 700, 670, 650, 600, 543 cm−1. MS (ESI): m/z
1447.0 (100%, [M − O2CC3F7 − 2MeCN]+) 1269.0 (59%, [M −
2O2CC3F7 − 2MeCN + Cl]+). HRMS (ESI): m/z calcd for [M −
O2CC3F7 − 2MeCN]+ 1446.0635, found 1446.0631.
134, 4683.
(11) (a) Henry, P. M. Handbook of Organopalladium Chemistry for
Organic Synthesis; Wiley-Interscience: New York, 2002; Vol. 2, p 2119.
(b) Fanning, K. N.; Jamieson, A. G.; Sutherland, A. Curr. Org. Chem.
2006, 10, 1007.
(12) (a) Dewar, M. J. S. Bull. Soc. Chim. Fr. 1951, 18, C71. (b) Chatt,
ASSOCIATED CONTENT
* Supporting Information
Further experimental procedures, characterization data, and
CIF files. This material is available free of charge via the
Internet at http://pubs.acs.org.
J.; Duncanson, L. A. J. Chem. Soc. 1953, 2939. (c) Furstner, A.; Davies,
P. W. Angew. Chem., Int. Ed. 2007, 46, 3410. (d) Chianese, A. R.; Lee,
S. J.; Gagne, M. R. Angew. Chem., Int. Ed. 2007, 46, 4042.
́
(13) (a) Widenhoefer, R. A. Pure Appl. Chem. 2004, 76, 671. (b) Liu,
C.; Widenhoefer, R. A. Tetrahedron Lett. 2005, 46, 285.
(14) The formation of a bisplatinacycle as a side product has never
been detected, even by using a large excess of Zeise salt or another
platinum source. A second platination is thus considered to be
significantly slower.
(15) Supplementary crystallographic data for this compound have
been deposited with the Cambridge Crystallographic Data Centre as
deposition 888130. This material is available free of charge via the
Internet at http://pubs.acs.org and http://www.ccdc.cam.ac.uk/
products/csd/request/.
̈
■
S
AUTHOR INFORMATION
Corresponding Author
*E-mail: rene.peters@oc.uni-stuttgart.de.
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
■
(16) The stereodescriptors with regard to the planar chirality are
used according to: Schlogl, K. Top. Stereochem. 1967, 1, 39.
̈
Notes
(17) (a) Djukic, J.-P.; Hijazi, A.; Flack, H. D.; Bernardinelli, G. Chem.
Soc. Rev. 2008, 37, 406. (b) Dupont, J.; Pfeffer, M. Palladacycles; Wiley-
VCH: Weinheim, Germany, 2008.
(18) Supplementary crystallographic data for this compound have
been deposited with the Cambridge Crystallographic Data Centre as
deposition 888129. This material is available free of charge via the
Internet at http://pubs.acs.org and http://www.ccdc.cam.ac.uk/
products/csd/request/.
(19) Selected applications: (a) Koch, F. M.; Peters, R. Chem. Eur. J.
2011, 17, 3679. (b) Kull, T.; Cabrera, J.; Peters, R. Chem. Eur. J. 2010,
16, 9132. (c) Tiseni, P. S.; Peters, R. Chem. Eur. J. 2010, 16, 2503.
(d) Zajac, M.; Peters, R. Chem. Eur. J. 2009, 15, 8204. (e) Kull, T.;
Peters, R. Angew. Chem., Int. Ed. 2008, 47, 5461. (f) Tiseni, P. S.;
Peters, R. Org. Lett. 2008, 10, 2019. (g) Koch, F. M.; Peters, R. Angew.
Chem., Int. Ed. 2007, 46, 2685.
(20) Rearrangement of trihaloacetimidates: (a) Overman, L. E.;
Owen, C. E.; Pavan, M. M.; Richards, C. J. Org. Lett. 2003, 5, 1809.
(b) Anderson, C. E.; Overman, L. E. J. Am. Chem. Soc. 2003, 125,
12412. (c) Kirsch, S. F.; Overman, L. E.; Watson, M. P. J. Org. Chem.
2004, 69, 8101. (d) Prasad, R. S.; Anderson, C. E.; Richards, C. J.;
Overman, L. E. Organometallics 2005, 24, 77. (e) Anderson, C. E.;
Donde, Y.; Douglas, C. J.; Overman, L. E. J. Org. Chem. 2005, 70, 648.
(f) Nomura, H.; Richards, C. J. Chem. Eur. J. 2007, 13, 10216.
(g) Jiang, G.; Halder, R.; Fang, Y.; List, B. Angew. Chem., Int. Ed. 2011,
50, 9752. (h) See ref 10.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Dr. Sascha Jautze for initial investigations to prepare
mixed pallada-/platinacycles and Dr. Assem Barakat for the
investigation of monoplatinacycle 1 for the rearrangement of
allylic trifluoroacetimidates. This work was financially sup-
ported by the Fonds der Chemischen Industrie (FCI, Ph.D.
fellowship to M.W.).
REFERENCES
(1) Wilcox, D. E. Chem. Rev. 1996, 96, 2435.
■
(2) Selected reviews about bimetallic catalysis: (a) van den Beuken,
E. K.; Feringa, B. L. Tetrahedron 1998, 54, 12985. (b) Shibasaki, M.;
Matsunaga, S. Chem. Soc. Rev. 2006, 35, 269. (c) Shibasaki, M.; Kanai,
M.; Matsunaga, S.; Kumagai, N. Acc. Chem. Res. 2009, 42, 1117.
(3) Jautze, S.; Diethelm, S.; Frey, W.; Peters, R. Organometallics 2009,
28, 2001.
(4) (a) Jautze, S.; Seiler, P.; Peters, R. Angew. Chem., Int. Ed. 2007,
46, 1260. (b) Jautze, S.; Seiler, P.; Peters, R. Chem. Eur. J. 2008, 14,
1430.
(5) (a) Nomura, H.; Richards, C. J. Chem. Asian J. 2010, 5, 1726.
(b) Peters, R.; Fischer, D. F.; Jautze, S. Top. Organomet. Chem. 2011,
33, 139.
(6) (a) Jautze, S.; Peters, R. Angew. Chem., Int. Ed. 2008, 47, 9284.
(b) Weber, M.; Jautze, S.; Frey, W.; Peters, R. J. Am. Chem. Soc. 2010,
132, 12222. (c) Weber, M.; Frey, W.; Peters, R. Adv. Synth. Catal.
2012, 354, 1443.
(7) Huang, H.; Peters, R. Angew. Chem., Int. Ed. 2009, 48, 604.
(8) Catalytic rearrangements of trichloroacetimidates have been
reported to proceed with 5−10 mol % of PtIICl2 or PtIVCl4 in CH2Cl2
or THF at room temperature within 18 h (yield, traces to 89%):
(a) Jaunzeme, I.; Jirgensons, A. Synlett 2005, 2984. (b) Jaunzeme, I.;
Jirgensons, A. Tetrahedron 2008, 64, 5794.
(21) Watson, M. P.; Overman, L. E.; Bergman, R. G. J. Am. Chem.
Soc. 2007, 129, 5031.
(22) A monomeric catalyst/substrate adduct such as 3 was observed
by ESI-MS (see the Supporting Information).
(23) The nonenantioselective version with PtCl2 or PtCl4 provided
only traces of rearrangement product if a N-substituted quaternary C
atom was formed using a trichloroacetimidate substrate with RE = RZ =
Me.9
(9) Almost no catalytic activity for the rearrangement of a
trichloroacetimidate was also found for a planar chiral oxazoline
platinacycle: Gunay, M. E.; Hughes, D. L.; Richards, C. J.
̈
Organometallics 2011, 30, 3901.
(10) (a) Weiss, M. E.; Fischer, D. F.; Xin, Z.-q.; Jautze, S.; Schweizer,
W. B.; Peters, R. Angew. Chem., Int. Ed. 2006, 45, 5694. (b) Fischer, D.
6372
dx.doi.org/10.1021/om300600v | Organometallics 2012, 31, 6365−6372