Synthesis and characterization of novel aminophosphine ligands based on ferrocenodecaline backbones

Article abstract of DOI:10.1007/s007060200069

Novel aminophosphine ligands for enantioselective transition metal catalysts based on different ferroceno cis- and trans-decaline backbones were synthesized and structurally characterized. Their palladium dichloride complexes were tested in the asymmetric

Full text of DOI:10.1007/s007060200069

Monatshefte fur Chemie 133, 991±1009 ꢀ2002)
Synthesis and Characterization
of Novel Aminophosphine Ligands Based
on Ferrocenodecaline Backbones
Max Weiûenbacher1, Thomas Sturm1, Hermann Kalchhauser1,
Christoph Kratky2, and Walter Weissensteiner1;Ã
1

Institut fur Organische Chemie, Universitat Wien, A-1090 Wien, Austria
Institut fur Chemie, Universitat Graz, A-8010 Graz, Austria
2

Summary. Novel aminophosphine ligands for enantioselective transition metal catalysts based on
different ferroceno cis- and trans-decaline backbones were synthesized and structurally characterized.
Their palladium dichloride complexes were tested in the asymmetric Grignard cross coupling reaction
of vinyl bromide and phenylethyl magnesium chloride, but only very low enantioselectivities were
obtained. Steric strain in the aminophosphine ligands causes a severe backbone deformation and in
addition leads to a slowed rotation of the respective dimethylamino group as was detected by variable
temperature NMR spectroscopy.
Keywords. Aminophosphines; Asymmetric catalysis; Ferrocene; Grignard cross coupling; Variable
temperature NMR; X-Ray diffraction.
Introduction
Some time ago we have reported the successful application of ferrocenyl amino-
phosphine PTFA ꢀScheme 1) in enantioselective Grignard cross coupling and other
catalytic reactions [1]. It is well documented that for this type of catalysts the steric
surroundings of the amino functionality are of prime importance [2]. For example,
use of PPFA in the nickel-catalyzed reaction of phenylethyl magnesium chloride
with vinyl bromide gave 3-phenylbutene in 68% e.e. as the ®nal product [3],
whereas application of its aminopropyl analogue ꢀMe-PPFA, Scheme 1) resulted
in 86% e.e. [4]. Hence, it was of interest to explore analogues of PTFA with
signi®cantly different steric requirements in close proximity to the dimethylamino
group.
We decided to extend the six-membered ring of PTFA to a decaline system
which, when substituted like PTFA, can exist in 4 different structural variations,
two with a cis- and two with a trans-decaline con®guration ꢀScheme 2, A±D). Such
systems are expected to be rather rigid and conformationally less ¯exible than
PTFA itself, providing different steric arrangements around the dimethylamino
Ã
Corresponding author. E-mail: walter.weissensteiner@univie.ac.at

992
M. Weiûenbacher et al.
Scheme 1
Scheme 2
functionality. We now describe synthesis routes to ferrocenyl aminophosphines 1
and 2 and their palladium dichloride complexes 1 Á PdCl2 and 2 Á PdCl2 with either a
cis ꢀ1) or a trans ꢀ2) ferrocenodecaline backbone ꢀScheme 2), their structural
properties, and some catalytic cross coupling experiments.
In the ®rst part, routes to different ferrocenodecaline systems are discussed. In
the second part, a diastereoselective synthesis of enantiopure 1 and 2, starting from
a commercially available enantiopure cyclohexenedicarboxylic acid derivative, is
described.
Results and Discussion
Synthesis of racemic ligands 1 and 2
Racemic 1 was prepared via a seven-step synthesis ꢀScheme 3) starting from ferro-
cene which was reacted with cis-cyclohexane-1,2-dicarboxylic acid anhydride and
AlCl3 in CS2 giving the ketoacid 3 in 62% yield. 3 was transformed in a Clemmensen
reduction to the methyl ester of 2-ferrocenylmethyl cyclohexanecarboxylic acid 4 in

Novel Ferrocene-Based Aminophosphine Ligands
993
Scheme 3
88% yield. This ester is also accessible via catalytic hydrogenation of 5 in acetic
acid ꢀPtO2, 66%). Saponi®cation of 4 led to a mixture of 2-ferrocenylmethyl-
cyclohexanecarboxylic acid 6 and a small amount of by-product 15 in an overall
yield of 71%. Cyclization with tri¯uoroacetic anhydride ꢀTFEA) afforded 71% of a
mixture of two ketones which could be separated by chromatography on silica
giving cis-ketone 7 as the major product ꢀ65%) and trans-ketone 16 as a by-product
ꢀ6%).
Interestingly, the cyclization of acid 6 gave trans-ketone 16 as the by-product
instead of the second cis-ketone of type B ꢀScheme 2) which, beside 7, was
expected to be formed in this ring closure reaction ꢀsee Scheme 3, 6 ! 7, dotted
arrow), indicating a partial epimerization of ester 4 in the saponi®cation step
leading to acid 15 as the by-product which functions as the precursor of ketone
16 ꢀsee below).
Reduction of 7 with either LiAlH4 in Et2O or B2H6 in THF resulted in exo-
alcohol 8 whose stereochemical integrity was proved by a crystal structure analysis
ꢀFig. 1) con®rming the cis-con®guration of the decaline system ꢀScheme 2, type A)
as well as the exo-position of the hydroxyl group [5]. For the structural character-
ization of ketone 16, see below.

994
M. Weiûenbacher et al.
Fig. 1. Molecular structures of 8, 9, and 1
Further reaction of alcohol 8 with ꢀCH3)2NH=AlCl3 in 1,2-dichloroethane gave
amine 9 with retention of con®guration as was con®rmed by X-ray diffraction ꢀFig. 1)
[6]. Finally, treatment of 9 with sec-BuLi in Et2O and quenching with ClPPh2 led
to aminophosphine 1 ꢀ60%) and to aminodiphosphine 1b ꢀ4%). It is important to
note that the lithiation step is especially sensitive to the concentration of amine 9
as well as to the choice of solvent and the reaction temperature. Acceptable yields
could only be achieved at high amine concentrations in diethyl ether. The molec-
ular structure of 1 was determined by X-ray diffraction and is shown in Fig. 1.
Racemic aminophosphine 2 with a trans-decaline backbone was accessible
from ketone 7 in 4 steps ꢀScheme 4). Epimerization of 7 with LDA in THF led
to ketone 10 in 85% yield. Subsequent reduction with LiAlH4 in Et2O resulted in a
mixture of alcohols 11a ꢀ57%) and 11b ꢀ28%) which could be separated by chro-
matography on silica. An X-ray diffraction study of 11a con®rmed the trans-con-
®guration of the decaline system as well as the exo-position of the hydroxyl group
ꢀFig. 2). Reaction of 11a with ꢀCH3)2NH=AlCl3 in 1,2-dichloroethane afforded
Scheme 4

Novel Ferrocene-Based Aminophosphine Ligands
995
Fig. 2. Molecular structures of 11a and 17
exo-amine 12 under retention of con®guration; ortho-lithiation of 12 with n-BuLi
in Et2O and quenching with ClPPh2 yielded aminophosphine 2 ꢀ75%).
Asynthesis route to derivatives with a trans-ferrocenodecaline backbone of
type D ꢀScheme 2) starts ± in analogy to the synthesis of 1 ± from ferrocene
and trans-cyclohexanedicarboxylic acid anhydride which, when reacted with AlCl3
in CH2Cl2, gave ketoacid 13 in 68% yield ꢀScheme 5).
Clemmensen reduction and esteri®cation of 13 led to the methyl ester 14 ꢀ58%)
which after saponi®cation ꢀ15, 70%) and cyclization with TFEA resulted in a
Scheme 5

996
M. Weiûenbacher et al.
mixture of ketones which could be separated on silica, the trans-ketone 16 ꢀ80%)
being the major and the cis-ketone 7 ꢀ4%) being the minor product. The formation
of 7 is again the result of a partial epimerization in the saponi®cation step of 14. A s
before, the cyclization of 15 to 16 is a highly diastereoselective reaction. Reduction
of 16 with LiAlH4 in Et2O gave exclusively endo-alcohol 17 whose structure was
elucidated by an X-ray diffraction study con®rming both the trans-decaline back-
bone of type D and the endo-position of the hydroxyl group ꢀFig. 2). In this
particular case, all attempts to transform either the keto group of 16 or the hydroxyl
group of 17 into an amino functionality failed.
In principle, like in the synthesis of trans-2 from cis-7 ꢀScheme 4), epimeriza-
tion of trans-ketone 16 could lead to the corresponding cis-ketone 18 with a deca-
line system of type B ꢀScheme 2). However, it was not possible to isolate such a
cis-ketone from the reaction mixture when 16 was treated with LDA in THF.
Synthesis of enantiopure ligands 1 and 2
In analogy to one of the preparation methods of PTFA [7] it was tried to separate
racemic ketones 7 and 10 by chromatography on chiral stationary phases such as
triacetyl or tribenzoyl cellulose. However, only on Chiralcel OD type stationary
phases reasonable separations could be achieved. Hence, a diastereoselective route
for the synthesis of ketone 7, the precursor of both aminophosphines 1 and 2, was
established.
Commercially available enantiopure cis-1,2-cyclohexenedicarboxylic acid
monomethylester was catalytically hydrogenated to the corresponding cyclohexane
derivative 19 in 92% yield ꢀScheme 6). Treatment with SOCl2 and reaction with
ferrocene gave 33% of enantiopure ketoester 20. Clemmensen reduction led to ester
4 ꢀ88%) which ®nally was transformed into the key intermediate 7 following the
preparation method for 1 as described above ꢀScheme 3). The enantiomeric purity
of 7 ꢀas well as those of 10 and 16) could be con®rmed by HPLC on Chiralcel OD.
It is interesting to note that the enantiopure ester 4 does not show any detectable
Scheme 6

Novel Ferrocene-Based Aminophosphine Ligands
997
Fig. 3. Molecular structures of 1 Á PdCl2 and 2 Á PdCl2
speci®c rotation, neither at 589 nor at 578 nm. Only in the CD spectrum transitions
of small intensity were observed. The synthesis of enantiopure ligand continues
exactly as described for the racemic ligand ꢀSchemes 2 and 3).
Synthesis of palladium dichloride complexes 1 Á PdCl2 and 2 Á PdCl2
Racemic as well as enantiomerically pure 1 Á PdCl2 and 2 Á PdCl2 are accessible by
reacting ligands 1 or 2 with ꢀCH3CN)2PdCl2 in benzene in 86 and 64% yield,
respectively. The molecular structures of both rac-1 Á PdCl2 and rac-2 Á PdCl2 were
determined ꢀFig. 3).
As mentioned above, the cyclization reactions of acid 6 to ketone 7 as well as of
acid 15 to ketone 16 are highly diastereoselective. As indicated in Schemes 3 and 5,
in each case two diastereomers are expected to be formed, but only 7 and 16 could be
isolated. Force ®eld calculations [8] including all 4 possible diastereomeric ketones
ꢀScheme 2, RˆOˆ) suggest that the cyclization reactions are likely to be kinetically
controlled since ketones of type A ꢀ7) and B ꢀ18) ꢀpossible reaction products from 6)
as well as ketones of type C ꢀ16) and D ꢀ10) ꢀpossible reaction products of 15) differ
in ground state energy only by 0.6 kJ=mol ꢀenergy differences to the minimum
energy structure of 16: 10: 0.6 kJ=mol, 7: 11.0 kJ=mol, 18: 11.6 kJ=mol). According
to these calculations, the trans-ketones 10 and 16 are signi®cantly more stable than
their cis-diastereomers 7 and 18 ꢀ10.4 and 11.6 kJ=mol, respectively), thus allowing
to rationalize the fact that LDA epimerizes cis-ketone 7 to the more stable trans-
ketone 10, but not trans-16 to the less stable cis-18.
Structural features of 1 Á PdCl2 and its precursors 8, 9, and 1
The molecular structures of 8, 9, 1 ꢀFig. 1), and 1 Á PdCl2 ꢀFig. 3) in the solid state
were determined by single crystal X-ray diffraction; the corresponding data are
given in Table 1. The general features of these 4 derivatives are rather similar: the
ferrocenodecaline backbone of type A ꢀScheme 7) adopts one of two interconver-
tible structures with the annelated six-membered ring system in a chair conforma-
tion and with the exo-substituents ꢀ±OH or ±NMe2) in a pseudo-equatorial position

998
M. Weiûenbacher et al.

Novel Ferrocene-Based Aminophosphine Ligands
999
Scheme 7
Fig. 4. Superposition of the molecular structures of 9 and 1 ꢀleft) and of the calculated structures of
12 and 2 ꢀright)
ꢀA-eq). The molecular structure of aminophosphine 1 is of particular interest since
both functional groups are well pre-organized for the formation of a chelate com-
plex, the amino and the phosphino lone pairs pointing almost towards each other. A
comparison of amine 9 and aminophosphine 1 shows that the phosphino group
obviously imposes severe steric strain on the system which is released by tilting the
ferrocene against the decaline unit ꢀFig. 4).
According to a simple analysis of the vicinal coupling constants JꢀHAHX) and
JꢀHBHX) ꢀScheme 7), the predominant conformation of 8, 9, 1, and 1 Á PdCl2 in
solution is identical to that found in the solid state ꢀfor coupling constant values,
see Experimental). The effect of strain in aminophosphine 1 can also be observed
in solution by variable temperature 1H NMR spectroscopy: upon lowering the tem-
perature, an exchange phenomenon involving the dimethylamino group is observed
ꢀfor details, see below).
Structural features of 2 Á PdCl2, its precursors 11a, 12, and 2,
and alcohol 17
Both ferroceno trans-decaline systems ꢀtypes C and D, Scheme 2), represented e.g.
by alcohols 11 and 17 ꢀFig. 2), can adopt only one single conformation with a
chair-like annelated six-membered ring. Force ®eld calculations on 11, 12, and 2
showed that, in agreement with the crystal structures of 11a and 2 Á PdCl2, the exo-
equatorial position of the hydroxyl or dimethylamino substituent is energetically

1000
M. Weiûenbacher et al.
either slightly ꢀ0.6 kJ=mol, 11a) or strongly ꢀ19.7 kJ=mol, 12; 24.7 kJ=mol, 2)
preferred over the endo-axial arrangement. Some structural properties are similar
to those of their cis-analogues 9 and 1. Again, steric strain in aminophosphine 2
leads to a strong tilt of the decaline system relative to the ferrocene unit ꢀFig. 4).
Like in 1, the functional groups are well pre-organized for chelate formation.
Dynamic processes in 1 and 2
Variable temperature 1H NMR measurements of 1 and 2 in toluene-d8 were carried
out. In both cases, the room temperature spectra show a rather broad singulet for
the dimethylamino group which decoalesces upon lowering the temperature ꢀTc:
302 Æ 2 K ꢀ1), 270 Æ 5 K ꢀ2)), ®nally leading to two singulets at the slow exchange
limit at 210 K ꢀÁꢆ: 12.5 Hz ꢀ1), 600 Hz ꢀ2)). The observed exchange phenomena
are interpreted as being caused by a slowed rotation of the dimethylamino unit
about the C±NMe2 bond. The activation barrier of this process was calculated as
65.8 Æ 2.1 kJ=mol for 1 and 49.9 Æ 2.1 kJ=mol for 2, respectively; the modi®ed
ˆ
Eyring equation ꢀÁG ˆ 4.57Tc  ꢀ9.97 ‡ log Tc=Áꢆ); Tc: coalescence tempera-
ture, Áꢆ: shift difference at the slow exchange limit) was used. Ahigher barrier
for cis-decaline 1 as compared to that of trans-oriented 2 ®ts the expectations.
Catalytic reactions
Complexes ꢀ‡)-ꢀ4S,4aS,8aS,Rm)-1 Á PdCl2 and ꢀÀ)-ꢀ4S,4aR,8aS,Rm)-2 Á PdCl2
were tested in the standard Grignard cross coupling reaction of vinyl bromide
and phenylethyl magnesium chloride ꢀScheme 1). In each case, high chemical
yields ꢀ1 Á PdCl2: 96%; 2 Á PdCl2: 84% based on vinyl bromide) were obtained, but
nearly racemic product 3-phenyl-but-1-ene was isolated ꢀ1 Á PdCl2: 4% e.e.;
2 Á PdCl2: 1% e.e.). This result strongly contrasts the 79% e.e. obtained with ꢀ4R,
Rm)-ꢀ4-N,N-dimethylamino-3-diphenylphosphino-ꢀꢇ5-4,5,6,7-tetrahydro-indenyl))-
ꢀꢇ5-cyclopenta-dienyl)-ironꢀII), ꢀR, Rm)-PTFA ꢀ[1a], Scheme 1). Possible structural
reasons could be either the rather strained ferrocenodecaline backbones of 1 Á PdCl2
and 2 Á PdCl2 ꢀsee above) or, more likely, the fact that in PTFA the dimethylamino
group is located in the endo-, in 1 Á PdCl2 and 2 Á PdCl2 in the exo-position. Although
obviously 1 Á PdCl2 and 2 Á PdCl2 are performing poorly in Grignard cross coupling
reactions, they seem to be interesting candidates for allylic alkylation and amination
reactions.
Experimental
General
NMR spectra were recorded on a Bruker AM-400 spectrometer in CDCl3 unless stated otherwise ꢀ1H:
400.0, 13C: 100.6, 31P: 161.9 MHz). Chemical shifts are given relative to internal TMS ꢀ1H and
13C NMR) or external 85% H3PO4 ꢀ31P NMR). The coupling constants given for 13C spectra refer
to 13C,31P-couplings. Melting points were determined on a Ko¯er melting point apparatus and are
uncorrected. Mass spectra were recorded on a Varian MAT-CH 7 spectrometer, CD spectra on a Jobin
Yvon CD 6 dichrograph. Optical rotations were measured on a Perkin Elmer 241 polarimeter. For
analytical HPLC, a Hewlett Packard HP1090 liquid chromatograph was used. The enantiomeric purity

Novel Ferrocene-Based Aminophosphine Ligands
1001
of ketones 7, 10, and 16 was determined by HPLC on Chiralcel OD ꢀDaicel); separation conditions:
¯ow 0.5 cm3=min, column temperature 35ꢀC, eluent hexane:isopropanol ˆ 95:5, c ˆ 2 mg=cm3,
injected 1 mm3. Elemental analyses were carried out at the Mikroanalytisches Laboratorium der
Universitat Wien; the results agreed favourably with the theoretical values. Reactions under inert
atmosphere ꢀAr) were carried out using standard Schlenk techniques. All solvents were dried by
standard procedures before use.
04S,4aS,8aS,Rm)-04-N,N-Dimethylamino-3-diphenylphosphino-0ꢇ5-5,6,7,8-tetrahydro-
4H,9H-benz[f]indenyl))-0ꢇ5-cyclopentadienyl)-iron0II) ꢀ1; C32H36FeNP)
To a degassed solution of 0.5 g ꢀ1.5mmol) of 9 in 2 cm3 of diethyl ether, 1.5 cm3 ꢀ2.4mmol) of n-BuLi
ꢀ1.6M in hexane) were added at 0ꢀC. After stirring for 18 h at room temperature, 0.52cm3 ꢀ2.9mmol)
of chlorodiphenylphosphine were added at À 10ꢀC, and the reaction mixture was stirred for additional
5 h at room temperature. After hydrolysis with 5 cm3 of saturated NaHCO3 solution the organic layer
was separated, diluted with 15cm3 of diethyl ether, and washed with H2O ꢀ2Â 10cm3). After drying
over MgSO4 and removal of the solvent, the crude product was chromatographed on silica ꢀpetrol
ether:triethylamine ˆ 97:3) to yield 0.46 g ꢀ0.9mmol, 60%) of 1.
1
M.p.: 161±165ꢀC; H NMR: ꢈ ˆ 1.04±1.17 ꢀm, 2H), 1.27±1.43 ꢀm, 3H), 1.49±1.70 ꢀm, 3H),
1.73±1.92 ꢀbs, 6H), 2.10 ꢀHB), 2.57 ꢀHA, A B,JAB ˆ 13.7 Hz, JAX ˆ 12.3Hz, JBX ˆ 2.9 Hz, 2H),
2.31 ꢀm, 1H), 2.47±2.54 ꢀm, 1H), 3.69 ꢀs, 5H), 3.82 ꢀm, 1H), 4.20 ꢀm, 1H), 4.57 ꢀd, J ˆ 7.4 Hz,
1H), 7.10±7.19 ꢀm, 3H), 7.21±7.29 ꢀm, 3H), 7.36±7.44 ꢀm, 2H), 7.58±7.66 ꢀm, 2H) ppm; 13C NMR:
ꢈ ˆ 20.96 ꢀCH2), 23.56 ꢀCH2), 25.01 ꢀCH2), 26.84 ꢀCH2), 32.04 ꢀCH2), 34.60 ꢀCH), 40.99 ꢀNꢀCH3)2),
41.88 ꢀCH), 63.31 ꢀCH), 67.38 ꢀCp), 70.48 ꢀd, J ˆ 5.3 Hz, Cp), 71.26 ꢀCp), 72.04 ꢀd, J ˆ 11.5Hz, Cp),
89.58 ꢀd, J ˆ 4.6 Hz, Cp), 92.85 ꢀd, J ˆ 21.4Hz, Cp), 127.63 ꢀd, J ˆ 7.6 Hz, 4Ph), 127.66 ꢀPh), 128.34
ꢀPh), 133.20 ꢀd, J ˆ 21.4 Hz, 2Ph), 135.17 ꢀd, J ˆ 22.9 Hz, 2Ph), 139.84 ꢀd, J ˆ 12.2Hz, Ph), 140.91 ꢀd
‡
J ˆ 11.4Hz, Ph) ppm; MS: m=z ꢀ%) ˆ 521 ꢀ47, M ), 478 ꢀ36), 292 ꢀ35), 242 ꢀ50), 199 ꢀ100), 185
ꢀ54), 149 ꢀ33), 121 ꢀ15), 108 ꢀ37), 91 ꢀ25); [ꢀ]20 ꢀnm)ˆ ‡ 49.3 ꢀ589), ‡ 52.4 ꢀ578)ꢀ ꢀc ˆ 0.87,
CHCl3).
04S,4aS,8aS,Rm)-004-N,N-Dimethylamino-3-diphenylphosphino-0ꢇ5-5,6,7,8-tetrahydro-
4H,9H-benz[f]indenyl))-0ꢇ5-cyclopentadienyl)-iron0II))-PdCl2 ꢀ1 Á PdCl2; C32H36Cl2FeNPPd)
To a degassed suspension of 36mg ꢀ0.14 mmol) of bisacetonitrilo palladium dichloride in 2.5 cm3 of
benzene, a degassed solution of 80 mg ꢀ0.15 mmol) of 1 in 2 cm3 of benzene was added, and the reac-
tion mixture was stirred at room temperature for 24h. The precipitate was ®ltered off and washed with
diethyl ether to yield 83mg ꢀ0.12 mmol, 86%) of 1 Á PdCl2.
M.p.: decomposition above 160ꢀC; 1H NMR: ꢈ ˆ 1.10±1.34 ꢀm, 2H), 1.35±1.77 ꢀm, 6H), 2.35 ꢀm,
1H), 2.44 ꢀd, Jꢀ1H,31P) ˆ 4.9 Hz, 3H), 2.48 ꢀm, 2H), 2.79 ꢀdd, JAB ˆ 17.8 Hz, JAX ˆ 13.7Hz, HA), 3.26
ꢀs, 3H), 3.95 ꢀs, 5H), 4.06 ꢀm, 1H), 4.56 ꢀd, J ˆ 2.5 Hz, 1H), 4.87 ꢀd, J ˆ 5.4Hz, 1H), 7.31±7.48 ꢀm,
6H), 7.77±7.94 ꢀm, 4H) ppm; 13C NMR: ꢈ ˆ 19.77 ꢀCH2), 22.60 ꢀCH2), 24.70 ꢀCH2), 25.67 ꢀCH2),
31.18 ꢀCH2), 34.05 ꢀCH), 39.89 ꢀCH), 50.72 ꢀd, J ˆ 5.3 Hz, NꢀCH3)), 53.96 ꢀNꢀCH3)), 70.57 ꢀd, J ˆ
6.1 Hz, Cp), 70.71, 71.76, 72.92 ꢀCp), 88.17 ꢀd, J ˆ 29.9Hz, Cp), 89.05 ꢀd, J ˆ 12.6Hz, Cp), 127.37 ꢀd,
J ˆ 18.9Hz, 2Ph), 128.80 ꢀd, J ˆ 17.3Hz, 2Ph), 130.30 ꢀd, J ˆ 77.2 Hz, Ph), 131.06 ꢀ2Ph), 131.28 ꢀd,
J ˆ 81.9Hz, Ph), 132.72 ꢀd, J ˆ 15.7 Hz, 2Ph), 135.51 ꢀd, J ˆ 17.3Hz, 2Ph) ppm; [ꢀ]20 ꢀnm)ˆ ‡ 32.3
ꢀ589), ‡ 32.3 ꢀ578)ꢀ ꢀc ˆ 0.099, CHCl3).
04S,4aR,8aS,Rm)-04-N,N-Dimethylamino-3-diphenylphosphino-0ꢇ5-5,6,7,8-tetrahydro-
4H,9H-benz[f]indenyl))-0ꢇ5-cyclopentadienyl)-iron0II) ꢀ2; C32H36FeNP)
To a degassed solution of 500 mg ꢀ1.5mmol) of 12 in 3 cm3 of diethyl ether, 1.5 cm3 ꢀ2.4mmol) of n-
BuLi ꢀ1.6M solution in hexane) were added at 0ꢀC. After stirring at room temperature for 20 h, 0.8 cm3

1002
M. Weiûenbacher et al.
ꢀ4.3mmol) of chlorodiphenylphosphine were added at 0ꢀC, and the solution was stirred for additional
4.5 h at room temperature and ®nally quenched with 10cm3 of saturated aqueous NaHCO3 solution.
The organic layer was separated, and the aqueous layer was extracted with CH2Cl2 ꢀ3Â 10cm3).
The combined organic phases were washed with H2O ꢀ3Â 15 cm3) and dried over MgSO4. A fter
evaporation of the solvent the residue was chromatographed on silica ꢀpetrol ether:ethylacetate:
triethylamine ˆ 79:20:1) to yield 590 mg ꢀ1.13 mmol, 76%) of 2.
1
M.p.: 57±59ꢀC; H NMR: ꢈ ˆ 1.0±1.4 ꢀm, 6H), 1.68±2.1 ꢀm, 11H), 2.3±2.39 ꢀm, 1H), 3.64±3.66
ꢀm, 1H), 3.88 ꢀs, 5H), 3.94 ꢀd, J ˆ 7.9 Hz, 1H), 4.13±4.15 ꢀm, 1H), 7.17±7.37 ꢀm, 8H), 7.55±7.61 ꢀm,
2H) ppm; 13C NMR: ꢈ ˆ 26.30 ꢀCH2), 26.33 ꢀCH2), 32.70 ꢀCH2), 34.17 ꢀCH2), 34.51 ꢀCH2), 39.76,
40.35, 66.15, 66.28, 69.55, 69.61, 70.69 ꢀCp), 73.66 ꢀd, J ˆ 30.3Hz, Cp), 89.76 ꢀd, J ˆ 15.2Hz, Cp),
93.12 ꢀd, J ˆ 75.8Hz, Cp), 127.3 ꢀPh), 127.46 ꢀd, J ˆ 30.3Hz, Ph), 127.64 ꢀd, J ˆ 27.3Hz, Ph), 128.20
ꢀPh), 132.47 ꢀd, J ˆ 81.9Hz, Ph), 134.94 ꢀd, J ˆ 85.0 Hz, Ph), 139.56 ꢀd, J ˆ 48.5Hz, Ph), 140.58 ꢀd,
‡
J ˆ 33.4Hz, Ph) ppm; MS: m=z ꢀ%) ˆ 521.3 ꢀ33.4, M ), 478.3 ꢀ35.7), 476.3 ꢀ22.6), 292.2 ꢀ37.0),
291.2 ꢀ28.2), 242.1 ꢀ41.6), 200.0 ꢀ42.1), 199.0 ꢀ100.0), 185.0 ꢀ13.4), 183.0 ꢀ43.1), 120.8 ꢀ13.0), 107.8
ꢀ29.7), 106.8 ꢀ13.9), 90.8 ꢀ19.3); [ꢀ]20 ꢀnm)ˆ ‡ 38.1 ꢀ589), ‡ 40.7 ꢀ578)ꢀ ꢀc ˆ 0.425, CHCl3).
04S,4aR,8aS,Rm)-004-N,N-Dimethylamino-3-diphenylphosphino-0ꢇ5-5,6,7,8-tetrahydro-
4H,9H-benz[f]indenyl))-0ꢇ5-cyclopentadienyl)-iron0II))-PdCl2 ꢀ2 Á PdCl2, C32H36Cl2FeNPPd)
To a degassed suspension of 65mg ꢀ0.26 mmol) of bisacetonitrilo palladium dichloride in 2.5 cm3
benzene, a solution of 140 mg ꢀ0.27 mmol) of 2 in 2.5 cm3 of benzene was added, and the reaction
mixture was stirred at room temperature for 24h. The precipitate was ®ltered off and washed with
diethyl ether. The product was recrystallized from CH2Cl2=hexane to yield 120 mg ꢀ0.17 mmol, 64%)
of 2 Á PdCl2.
M.p.: decomposition above 160ꢀC; 1H NMR: ꢈ ˆ 1.20±1.50 ꢀm, 6H), 1.80±1.92 ꢀm, 3H),
2.01±2.11 ꢀm, 1H), 2.21 ꢀd, J ˆ 9.4Hz, 1H), 2.27 ꢀdd, J1 ˆ 14.8 Hz, J2 ˆ 3.4 Hz, 1H), 2.33 ꢀs, 3H),
3.40 ꢀs, 3H), 3.63 ꢀs, 5H), 4.0±4.03 ꢀm, 1H), 4.19±4.21 ꢀm, 1H), 4.54 ꢀd, J ˆ 2.5 Hz, 1H), 7.30±7.36
ꢀm, 2H), 7.41±7.46 ꢀm, 1H), 7.51±7.62 ꢀm, 5H), 8.23±8.30 ꢀm, 2H) ppm; 13C NMR: ꢈ ˆ 26.40 ꢀCH2),
26.71 ꢀCH2), 30.06 ꢀCH2), 34.80 ꢀCH2), 36.92 ꢀCH2), 38.13, 41.90, 42.54, 51.40, 69.09 ꢀd,
J ˆ 27.3Hz), 70.89 ꢀd, J ˆ 15.2Hz), 71.38 ꢀCp), 71.56 ꢀd, J ˆ 9.1 Hz), 87.11, 92.49, 127.83 ꢀd,
J ˆ 48.5Hz, Ph), 128.29 ꢀd, J ˆ 45.5Hz, Ph), 130.63 ꢀd, J ˆ 12.1Hz, Ph), 131.48 ꢀd, J ˆ 45.5Hz,
Ph), 131.59 ꢀd, J ˆ 19.7Hz, Ph), 132.17 ꢀd, J ˆ 32.1 Hz, Ph), 133.73 ꢀd, J ˆ 36.4Hz, Ph), 135.41 ꢀd,
J ˆ 45.5Hz, Ph); [ꢀ]20 ꢀnm)ˆ À 38.8 ꢀ589), À 42.8 ꢀ578)ꢀ ꢀc ˆ 0.049, CHCl3); CD ꢀCH2Cl2): ꢉmax
ꢀÁ") ˆ 273 ꢀ ‡ 0.38), 313 ꢀ À 0.22), 376 ꢀ ‡ 0.09), 494 ꢀ À 0.08) nm.
Racemic cis-2-ferrocenoyl-cyclohexane-1-carboxylic acid ꢀ3; C18H20FeO3)
To a suspension of 34.7 g ꢀ260 mmol) AlCl3 in 300 cm3 of CH2Cl2, a solution of 44.3 g ꢀ240 mmol)
ferrocene and 19g ꢀ120mol) cis-cyclohexane-1,2-dicarboxylic acid anhydride in 300 cm3 CH2Cl2 was
added within 45 min at room temperature. The reaction mixture was stirred for additional 2 h at room
temperature and then poured on ice. The pH was adjusted to 4, and the organic layer was separated.
The aqueous layer was extracted with CH2Cl2 ꢀ4Â 100 cm3), and the combined organic layers were
concentrated to approximately 500 cm3 and extracted with a 2 N aqueous solution of NaOH
ꢀ4Â 250 cm3). Traces of organic solvents were removed from the aqueous phase in vacuo, and the
product was precipitated by slow addition of phosphoric acid ꢀ50%) at 5ꢀC. The dark orange pre-
cipitate was ®ltered off, washed with H2O, and dried over CaCl2 to yield 25.3 g ꢀ74 mmol, 62%) of 3.
M.p.: 51±54ꢀC; 1H NMR: ꢈ ˆ 1.17±1.53 ꢀm, 3H), 1.66±1.98 ꢀm, 3H), 2.09±2.30 ꢀm, 2H), 2.58 ꢀm,
1H), 3.49 ꢀm, 1H), 4.22 ꢀs, 5H), 4.44 ꢀs, 1H), 4.65 ꢀs, 1H), 4.86 ꢀs, 1H) ppm; 13C NMR: ꢈ ˆ 22.80
ꢀCH2), 24.39 ꢀCH2), 25.35 ꢀCH2), 29.37 ꢀCH2), 42.68 ꢀCH), 46.43 ꢀCH), 69.09 ꢀCp), 69.94 ꢀCp), 70.24
ꢀCp), 71.81 ꢀCp), 71.94 ꢀCp), 77.81 ꢀCp), 179.68 ꢀCO), 206.79 ꢀCO) ppm; MS: m=z ꢀ%) ˆ 340 ꢀ94,
‡
M ), 213 ꢀ98), 186 ꢀ87), 121 ꢀ82), 92 ꢀ100).

Novel Ferrocene-Based Aminophosphine Ligands
1003
01S,2R)-2-Ferrocenylmethyl-cyclohexane-1-carboxylic acid methylester ꢀ4; C19H24FeO2)
Method 1: To a solution of 6.9 g ꢀ20.3 mmol) of 3 in 220 cm3 of MeOH, 35cm3 of benzene, 35cm3 of
H2O, and 35 cm3 of concentrated HCl, freshly prepared zinc amalgam was added, and the mixture was
re¯uxed for 24h. After cooling to room temperature, the amalgam was ®ltered off and washed with
benzene. The organic layer was washed with brine ꢀ2Â 100 cm3) and dried over MgSO4. A fter
®ltration, a freshly prepared solution of diazomethane in diethyl ether was added, and the reaction
mixture was stirred at room temperature for 30min. The solvent was removed in vacuo, and the residue
was chromatographed twice on silica ꢀ®rst run: petrol ether:ethyl acetateˆ 95:5; second run: CH2Cl2:
petrol ether ˆ 75:25) to yield 6.1 g ꢀ17.9 mmol, 88%) of 4.
Method 2: Asuspension of 500 mg of PtO 2 in 40cm3 of anhydrous acetic acid was hydrogenated for
30min in a Parr apparatus at a pressure of 5 bar H2. After addition of 2.5 g ꢀ7.18 mmol) of 5 ꢀcaution:
Ptꢀ0) is highly pyrophoric!), the mixture was hydrogenated for 18h at a pressure of 5 bar, the colour of
the solution changing from dark red to orange during this time. The catalyst was ®ltered off, and the
acetic acid was removed in vacuo. The residue was dissolved in 50 cm3 of CH2Cl2, and the solution
was washed three times with saturated aqueous NH4Cl solution and 100 cm3 H2O each. After drying
over MgSO4 and removal of the solvent, the residue was chromatographed on alumina ꢀpetrol ether:
CH2Cl2 ˆ 25:75) to yield 1.6 g ꢀ4.7mmol, 66%) of 4.
M.p.: 63±65ꢀC; 1H NMR: ꢈ ˆ 1.22±1.45 ꢀm, 3H), 1.52±1.86 ꢀm, 6H), 2.32±2.44 ꢀm, 2H), 2.55 ꢀm,
1H), 3.68 ꢀs, 3H), 4.00 ꢀm, 1H), 4.03 ꢀm, 3H), 4.07 ꢀs, 5H) ppm; 13C NMR: ꢈ ˆ 23.27 ꢀCH2), 23.51
ꢀCH2), 26.36 ꢀCH2), 28.06 ꢀCH2), 31.68 ꢀCH2), 40.39, 44.45, 51.09, 67.11 ꢀCp), 67.20 ꢀCp), 68.51 ꢀCp),
‡
68.61 ꢀCp), 69.06 ꢀCp), 87.28 ꢀCp), 175.20 ꢀCO) ppm; MS: m=z ꢀ%) ˆ 340 ꢀ100, M ), 199 ꢀ91), 121
ꢀ48); [ꢀ]20 ꢀnm)ˆ 0.0 ꢀ589), 0.0 ꢀ587)ꢀ ꢀc ˆ 1.26, CHCl3); CD ꢀCH2Cl2, 1.05 10À 3 mol Á dmÀ 3):
ꢉ
max ꢀÁ") ˆ 333 ꢀ ‡ 0.015), 463 ꢀ À 0.006), 567 ꢀ ‡ 0.004) nm.
2-Carbomethoxybenzoyl ferrocene ꢀ5; C9H16FeO3)
To a solution of 12.0 g ꢀ64 mmol) of ferrocene in 120cm3 of p.a. CS2 in a three-necked ¯ask equipped
with a condenser and a mechanical stirrer, a solution of 12.8g ꢀ64 mmol) of 2-carbomethoxyben-
zoylchloride in 50 cm3 dry diethyl ether was added quickly. After addition of a solution of 17.2g
ꢀ129 mmol) AlCl3 in 50 cm3 dry diethyl ether ꢀcaution: dissolving AlCl3 in diethyl ether is a highly
exothermic reaction!), the reaction mixture was re¯uxed for 2 h during which time the product
separated as a highly viscous oil. After cooling the mixture to room temperature and separation of
the solvent, the residue was hydrolyzed with 500cm3 of diluted HCl ꢀpHˆ 3±4). The precipitated
product was ®ltered off, washed with 250 cm3 of H2O, recrystallized from MeOH:H2O ˆ 1:1 ꢀ10 cm3
of solvent per 1 g of product), and dried over CaCl2 to give 15.1g ꢀ43mmol, 68%) of pure 5.
M.p.: 139±141ꢀC; 1H NMR: ꢈ ˆ 3.62 ꢀs, 3H), 4.19 ꢀs, 5H), 4.50 ꢀs, 2H), 4.59 ꢀs, 2H), 7.54 ꢀm, 1H),
7.65 ꢀm, 2H), 7.93 ꢀm, 1H) ppm; 13C NMR: ꢈ ˆ 52.14 ꢀCH3), 69.81 ꢀCp), 70.00 ꢀCp), 72.40 ꢀCp),
79.83 ꢀCp), 127.62 ꢀPh), 129.22 ꢀPh), 129.55 ꢀPh), 129.75 ꢀPh), 131.99 ꢀPh), 142.11 ꢀPh), 166.88 ꢀCO),
‡
200.78 ꢀCO) ppm; MS: m=z ꢀ%) ˆ 348 ꢀ100, M ), 283 ꢀ17), 253 ꢀ48), 235 ꢀ4), 225 ꢀ8), 197 ꢀ15), 161
ꢀ32), 121 ꢀ11), 89 ꢀ3), 81 ꢀ6), 69 ꢀ7), 56 ꢀ44).
01S,2S)-2-Ferrocenylmethyl-cyclohexane-1-carboxylic acid ꢀ6; C18H22FeO2)
Asuspension of 5.75 g ꢀ16.9 mmol) of 4 in 100 cm3 of 25% aqueous KOH solution was re¯uxed under
Ar for 28 h. After ®ltration the solution was cooled with ice, and the product was precipitated with half-
concentrated H3PO4. The yellow powder was ®ltered off, washed with H2O, and dried over CaCl2 to
yield 3.9 g ꢀ12 mmol, 71%) of 6.
M.p.: 51±54ꢀC; 1H NMR ꢀDMSO-d6): ꢈ ˆ 0.95±1.17 ꢀm, 3H), 1.21±1.60 ꢀm, 6H), 2.10±2.26 ꢀm,
3H), 3.75±4.00 ꢀm, 9H) ppm; 13C NMR ꢀDMSO-d6): ꢈ ˆ 22.48 ꢀCH2), 23.53 ꢀCH2), 23.53 ꢀCH2),
25.90 ꢀCH2), 27.38 ꢀCH2), 30.33 ꢀCH2), 39.55 ꢀCH), 44.77 ꢀCH), 66.71 ꢀCp), 66.99 ꢀCp), 68.05 ꢀCp),

1004
M. Weiûenbacher et al.
‡
68.25 ꢀCp), 68.48 ꢀCp), 87.52 ꢀCp), 176.03 ꢀCO) ppm; MS: m=z ꢀ%) ˆ 326 ꢀ67, M ), 199 ꢀ100), 121
ꢀ99); [ꢀ]20 ꢀnm)ˆ 0.0 ꢀ589), 0.0 ꢀ578)ꢀ ꢀc ˆ 0.86, CHCl3).
04aS,8aS,Rm)-04-Oxo-0ꢇ5-5,6,7,8-tetrahydro-9H-benz[f]indenyl))-0ꢇ5-cyclopentadienyl)-iron0II)
ꢀ7; C18H20FeO) and 04aR,8aS,Sm)-04-oxo-0ꢇ5-5,6,7,8-tetrahydro-9H-benz[f]indenyl))-
0ꢇ5-cyclopentadienyl)-iron0II) ꢀ16; C18H20FeO)
Asolution of 1.3 g ꢀ4mmol) of 6 in 50cm3 CH2Cl2 was added dropwise to a solution of 1 g ꢀ4.8mmol)
tri¯uoracetic anhydride in 50 cm3 CH2Cl2 at 0ꢀC. The reaction mixture was stirred at this temperature
for additional 5 h. To the dark red solution, 30 cm3 of saturated aqueous NaHCO3 solution were added,
and the organic layer was separated. The aqueous layer was extracted with CH2Cl2 ꢀ2Â 25 cm3), and the
combined organic layers were washed with H2O ꢀ3Â 50cm3). After drying over MgSO4 and removal of
the solvent, the residue was chromatographed on silica ꢀCH2Cl2:petrol ether ˆ 95:5) to yield 0.8 g
ꢀ2.6mmol, 65%) of 7 and 75mg ꢀ0.24 mmol, 6%) of 16.
1
7: M.p.: 125±128ꢀC; H NMR: ꢈ ˆ 1.28±1.74 ꢀm, 8H), 2.40±2.75 ꢀm, 4H), 4.14 ꢀs, 5H), 4.44 ꢀs,
1H), 4.45 ꢀs, 1H), 4.81 ꢀm, 1H) ppm; 13C NMR: ꢈ ˆ 21.65 ꢀCH2), 24.94 ꢀCH2), 25.31 ꢀCH2), 29.99
ꢀCH2), 35.18 ꢀCH), 50.10 ꢀCH), 65.31 ꢀCp), 70.00 ꢀCp), 70.24 ꢀCp), 70.78 ꢀCp), 74.04 ꢀCp), 91.53
‡
ꢀCp), 207.91 ꢀCO) ppm; MS: m=z ꢀ%): 308 ꢀ100, M ), 199 ꢀ20), 121 ꢀ43); [ꢀ]20 ꢀnm)ˆ ‡ 69.9 ꢀ598),
‡ 73.8 ꢀ578)ꢀ ꢀc ˆ 0.101, CHCl3); CD ꢀCH2Cl2, 0.99 10À 3 mol Á dmÀ 3): ꢉmax ꢀÁ") ˆ 240 ꢀ ‡ 1.70),
267 ꢀ ‡ 0.53), 302 ꢀ À 0.59), 340 ꢀ ‡ 0.68), 456 ꢀ ‡ 0.36), 531 ꢀ ‡ 0.05) nm.
1
16: M.p.: 122±123ꢀC; H NMR: ꢈ ˆ 1.05±1.38 ꢀm, 4H), 1.56±1.92 ꢀm, 4H), 2.33±2.47 ꢀm, 2H),
2.61±2.75 ꢀm, 2H), 4.07 ꢀs, 5H), 4.33 ꢀm, 1H), 4.49 ꢀm, 1H), 4.73 ꢀm, 1H) ppm; 13C NMR: ꢈ ˆ 25.58
ꢀCH2), 25.82 ꢀCH2), 26.16 ꢀCH2), 32.55 ꢀCH2), 34.26 ꢀCH2), 41.35 ꢀCH), 50.85 ꢀCH), 64.51 ꢀCp), 70.02
‡
ꢀCp), 70.27 ꢀCp), 70.90 ꢀCp), 74.73 ꢀCp), 91.69 ꢀCp), 205.51 ꢀCO) ppm; MS: m=z ꢀ%) ˆ 308 ꢀ100, M ),
199 ꢀ19), 121 ꢀ41); [ꢀ]20 ꢀnm)ˆ ‡ 85.4 ꢀ589), ‡ 86.7 ꢀ578)ꢀ ꢀc ˆ 0.315, CHCl3); CD ꢀCH2Cl2,
0.98 10À 3 mol Á dmÀ 3): ꢉmax ꢀÁ") ˆ 271 ꢀ ‡ 1.24), 304 ꢀ À 2.05), 336 ꢀ ‡ 1.79), 450 ꢀ ‡ 0.73), 521
ꢀ À 0.19) nm.
04S,4aS,8aS,Rm)-04-Hydroxy-0ꢇ5-5,6,7,8-tetrahydro-4H,9H-benz[f]indenyl))-
0ꢇ5-cyclo-pentadienyl)-iron0II) ꢀ8; C18H22FeO)
To a suspension of 30mg ꢀ0.8mmol) of LiAlH4 in 5 cm3 of dry diethyl ether, a solution of 0.25g
ꢀ0.8mmol) of 7 in 25cm3 of dry diethyl ether was added dropwise at À 12 to À 15ꢀC. The colour
changed from dark red to yellow. The reaction mixture was stirred for 1 h at À 15ꢀC and then for one ad-
ditional hour at room temperature. After addition of 5 cm3 of H2O the solution was ®ltered over celite.
The organic layer was separated, washed with H2O ꢀ2Â 25cm3), and dried over MgSO4. After removal
of the solvent the residue was chromatographed on silica ꢀCH2Cl2) to yield 166 mg ꢀ0.54 mmol, 68%)
of 8.
M.p.: 149±151ꢀC; 1H NMR: ꢈ ˆ 0.78±0.92 ꢀm, 1H), 1.16±1.31 ꢀm, 1H), 1.38 ꢀOH, d, J ˆ 8.4 Hz,
1H), 1.43±1.55 ꢀm, 3H), 1.60±1.69 ꢀm, 2H), 1.71±1.81 ꢀm, 1H), 1.92±2.01 ꢀm, 1H), 2.28 ꢀHB), 2.52
ꢀHA) ꢀA B,JAB ˆ 14.8 Hz, JAX ˆ 11.8Hz, JBX ˆ 4.9 Hz, 2H), 2.38±2.47 ꢀm, 1H), 4.00 ꢀs, 7H), 4.43 ꢀs,
1H), 5.21 ꢀt, J ˆ 6.9 Hz, 1H) ppm; 13C NMR: ꢈ ˆ 18.01 ꢀCH2), 20.85 ꢀCH2), 23.86 ꢀCH2), 25.91 ꢀCH2),
30.93 ꢀCH2), 32.74 ꢀCH), 42.90 ꢀCH), 64.89 ꢀCp), 65.27 ꢀCp), 65.27 ꢀCp), 66.10 ꢀCp), 69.75 ꢀCp),
‡
71.31 ꢀCH), 84.06 ꢀCp), 86.18 ꢀCp) ppm; MS: m=z ꢀ%) ˆ 310 ꢀ70, M ), 121 ꢀ40), 49 ꢀ100); [ꢀ]20
ꢀnm) ˆ ‡ 11.3 ꢀ589), ‡ 11.5 ꢀ578)ꢀ ꢀc ˆ 0.93, CHCl3); CD ꢀCH2Cl2, 0.99  10À 3 mol Á dmÀ 3): ꢉmax
ꢀÁ") ˆ 269 ꢀ ‡ 0.12), 295 ꢀ À 0.10), 339 ꢀ ‡ 0.07), 445 ꢀ ‡ 0.05), 511 ꢀ À 0.01) nm.
04S,4aS,8aS,Rm)-04-N,N-Dimethylamino-0ꢇ5-5,6,7,8-tetrahydro-4H,9H-benz[f]indenyl))-
0ꢇ5-cyclopentadienyl)-iron0II) ꢀ9; C20H27FeN)
Asuspension of 70 mg ꢀ0.52 mmol) of AlCl in 25cm3 1,2-dichloroethane was cooled to 0ꢀC, and
3
gaseous dimethylamine was added until complete consumption of AlCl3 ꢀan excess of dimethylamine

Novel Ferrocene-Based Aminophosphine Ligands
1005
is recommended). Asolution of 0.14 g ꢀ0.45 mmol) of 8 in 25 cm3 1,2-dichloroethane was added
slowly to this clear solution. The reaction mixture was stirred at 0ꢀC for 30min and then for additional
2 h at 50ꢀC. After hydrolysis with 20 cm3 of H2O the organic layer was separated and the solvent was
removed in vacuo. The residue was suspended with 25 cm3 of 0.1 N KOH and extracted with diethyl
ether ꢀ3Â 50cm3). The combined organic layers were washed with H2O ꢀ3Â 50cm3) and dried over
MgSO4. The solvent was removed in vacuo, and the residue was chromatographed on silica ꢀpetrol
ether:ethyl acetate:triethylamine ˆ 79:20:1) to yield 85mg ꢀ0.25 mmol, 56%) of 9.
M.p.: 78±80ꢀC; 1H NMR: ꢈ ˆ 0.90±1.07 ꢀm, 1H), 1.14±1.30 ꢀm, 1H), 1.34±1.53 ꢀm, 3H),
1.57±1.76 ꢀm, 3H), 2.05 ꢀm, 1H), 2.28 ꢀHB), 2.50 ꢀHA) ꢀA B,JAB ˆ 13.0Hz, JAX ˆ 12.3 Hz,
JBX ˆ 4.9 Hz, 2H), 2.44 ꢀs, 6H), 3.98 ꢀs, 6H), 4.01 ꢀs, 1H), 4.24 ꢀd, J ˆ 5.4 Hz, 1H), 4.36 ꢀs, 1H)
ppm; 13C NMR: ꢈ ˆ 20.61 ꢀCH2), 21.42 ꢀCH2), 24.36 ꢀCH2), 26.35 ꢀCH2), 31.40 ꢀCH2), 33.85 ꢀCH),
42.27 ꢀCH), 42.99 ꢀNꢀCH3)2), 64.68 ꢀCp), 65.35 ꢀCp), 65.64 ꢀCH), 66.62 ꢀCp), 69.87 ꢀCp), 83.85 ꢀCp),
‡
85.05 ꢀCp) ppm; MS: m=z ꢀ%) ˆ 337 ꢀ70, M ), 291 ꢀ67), 121 ꢀ79), 56 ꢀ100); [ꢀ]20 ꢀnm)ˆ ‡ 12.7
ꢀ589), ‡ 13.0 ꢀ587)ꢀ ꢀc ˆ 0.955, CHCl3).
04aR,8aS,Rm)-04-Oxo-0ꢇ5-5,6,7,8-tetrahydro-9H-benz[f]indenyl))-
0ꢇ5-cyclopentadienyl)-iron0II) ꢀ10; C18H20FeO)
To a solution of 500 mg ꢀ1.6mmol) of 7 in 10 cm3 of THF, 1.25 cm3 ꢀ2.5mol) of a 2 M LDA solution in
THF were added slowly at À 78ꢀC, and the mixture was stirred at this temperature for 40min. After
quenching with 25 cm3 of H2O, the solution was allowed to warm to room temperature and extracted
with CH2Cl2 ꢀ3Â 25cm3). The combined organic layers were washed with 3% HCl ꢀ3Â 30cm3),
saturated NaHCO3 solution ꢀ3Â 30 cm3), and H2O ꢀ3Â 30 cm3) and dried over MgSO4. A fter evapora-
tion of the solvent the residue was chromatographed on silica ꢀCH2Cl2:petrol ether:ethyl acetate ˆ
4:1:5) to yield 418 mg ꢀ1.36 mmol, 83.6%) of 10.
1
M.p.: 153±155ꢀC; H NMR: ꢈ ˆ 1.16±1.40 ꢀm, 4H), 1.65±1.71 ꢀm, 1H), 1.76±1.82 ꢀm, 1H),
1.85±1.95 ꢀm, 1H), 2.09±2.20 ꢀm, 2H), 2.45±2.52 ꢀm, 1H), 2.54±2.63 ꢀm, 1H), 4.15 ꢀs, 5H), 4.39±
4.40 ꢀm, 1H), 4.40±4.43 ꢀm, 1H), 4.81±4.82 ꢀm, 1H) ppm; 13C NMR: ꢈ ˆ 25.97 ꢀCH2), 26.02 ꢀCH2),
26.16 ꢀCH2), 31.53 ꢀCH2), 34.28 ꢀCH2), 41.47 ꢀCH), 52.46 ꢀCH), 65.47 ꢀCp), 69.55 ꢀCp), 69.92 ꢀCp)
‡
70.35 ꢀCp), 75.20 ꢀCp), 91.06 ꢀCp), 205.50 ꢀCO) ppm; MS: m=z ꢀ%) ˆ 308.2 ꢀ100, M ), 198.9 ꢀ16.1),
120.8 ꢀ21.0), 55.8 ꢀ18.6); [ꢀ]20 ꢀnm) ˆ ‡ 37.0 ꢀ589), ‡ 38.3 ꢀ578)ꢀ ꢀc ˆ 0.081, CHCl3); CD ꢀCH2Cl2):
ꢉ
maxꢀÁ") ˆ 266 ꢀ ‡ 0.47), 302 ꢀ À 0.70), 338 ꢀ ‡ 0.59), 455 ꢀ ‡ 0.30) nm.
04S,4aR,8aS,Rm)-04-Hydroxy-0ꢇ5-5,6,7,8-tetrahydro-4H,9H-benz[f]indenyl))-
0ꢇ5-cyclo-pentadienyl)-iron0II) ꢀ11a; C18H22FeO) and 04R,4aR,8aS,Rm)-
04-hydroxy-0ꢇ5-5,6,7,8-tetrahydro-4H,9H-benz[f]indenyl))-
0ꢇ5-cyclopentadienyl)-iron0II) ꢀ11b; C18H22FeO)
To a suspension of 170 mg ꢀ4.5mmol) of LiAlH4 in 15cm3 of diethyl ether, a solution of 900 mg
ꢀ2.9mmol) of 10 in 35 cm3 of diethyl ether was added dropwise at 0ꢀC. The reaction mixture was
stirred at 0ꢀC for 1 h and then for one additional hour at room temperature. H2O ꢀ25 cm3) was added,
and the organic layer was separated. The aqueous layer was extracted with diethyl ether ꢀ2Â 20cm3),
and the combined organic layers were washed with H2O ꢀ3Â 25 cm3) and dried over MgSO4. A fter
evaporation of the solvent the residue was chromatographed on silica ꢀCH2Cl2), affording the two
isomeric alcohols 11a ꢀ505 mg, 1.63 mmol, 56.1%) and 11b ꢀ255mg, 0.82mmol, 28.3%).
11a: M.p.: 151±153ꢀC; 1H NMR: ꢈ ˆ 0.99±1.12 ꢀm, 2H), 1.18±1.38 ꢀm, 2H), 1.62±1.75 ꢀm, 4H),
1.79±1.94 ꢀm, 4H), 2.06 ꢀOH, d, J ˆ 6.9 Hz, 1H), 2.42±2.52 ꢀm, 1H), 3.94 ꢀdd, J1 ˆ 4.2 Hz, J2 ˆ
6.7 Hz, 1H) , 4.09±4.11 ꢀm, 1H), 4.14±4.16 ꢀm, 1H), 4.20 ꢀs, 5H) ppm; 13C NMR: ꢈ ˆ 26.09 ꢀCH2),
26.51 ꢀCH2), 27.94 ꢀCH2), 32.61 ꢀCH), 33.33 ꢀCH2), 34.24 ꢀCH2), 45.22 ꢀCH), 64.50, 64.81, 65.50,
‡
66.50 ꢀCp), 68.61, 87.01 ꢀCp), 94.69 ꢀCp) ppm; MS: m=z ꢀ%) ˆ 310.3 ꢀ100, M ), 292.2 ꢀ23.3), 291.2
ꢀ15.7), 225.1 ꢀ23.4), 223.0 ꢀ19.6), 221.0 ꢀ12.1), 172.1 ꢀ44.1), 129.0 ꢀ34.3), 128.8 ꢀ14.8), 120.9 ꢀ16.5),

1006
M. Weiûenbacher et al.
114.9 ꢀ12.5); [ꢀ]20 ꢀnm)ˆ ‡ 9.4 ꢀ589), ‡ 9.6 ꢀ578)ꢀ ꢀc ˆ 0.5, CHCl3); CD ꢀCH2Cl2): ꢉmax ꢀÁ") ˆ 246
ꢀ À 0.21), 288 ꢀ À 0.18), 475 ꢀ À 0.06) nm.
1
11b: M.p.: 115±117ꢀC; H NMR: ꢈ ˆ 0.9±1.0 ꢀm, 1H), 1.04±1.17 ꢀm, 2H), 1.19±1.48 ꢀm, 2H),
1.39 ꢀOH, d, J ˆ 2.5Hz, 1H), 1.73±1.99 ꢀm, 5H), 2.25±2.33 ꢀm, 1H), 2.49 ꢀdd, J1 ˆ 4.3 Hz, J2 ˆ
15.3Hz, 1H), 4.0±4.02 ꢀm, 1H) 4.03 ꢀs, 5H), 4.39±4.41 ꢀm, 1H), 4.54±4.60 ꢀm, 1H) ppm; 13C NMR:
ꢈ ˆ 26.04 ꢀCH2), 26.09 ꢀCH2), 30.99 ꢀCH2), 32.60 ꢀCH2), 34.14 ꢀCH2), 37.73 ꢀCH), 49.03 ꢀCH), 64.80,
65.27, 66.26, 69.48 ꢀCp), 74.31, 85.04 ꢀCp), 86.91 ꢀCp) ppm; MS: m=z ꢀ%): 310.3 ꢀ100, M ‡ ), 292.2
ꢀ16.6), 291.2 ꢀ16.5), 238.1 ꢀ15.3), 222.0 ꢀ20.9), 129.0 ꢀ32.2), 128.0 ꢀ11.8), 120.9 ꢀ16.8); [ꢀ]20
ꢀnm) ˆ ‡ 22.1 ꢀ589), ‡ 23.9 ꢀ578)ꢀ ꢀc ˆ 0.435, CHCl3); CD ꢀCH2Cl2): ꢉmax ꢀÁ") ˆ 264 ꢀ‡ 0.12),
287 ꢀ À 0.22), 444 ꢀ ‡ 0.04), 505 ꢀ À 0.03) nm.
04S,4aR,8aS,Rm)-04-N,N-Dimethylamino-0ꢇ5-5,6,7,8-tetrahydro-4H,9H-benz[f]indenyl))-
0ꢇ5-cyclopentadienyl)-iron0II) ꢀ12; C20H27FeN)
To a suspension of 165 mg ꢀ1.23 mmol) of AlCl3 in 25cm3 1,2-dichloroethane, gaseous dimethylamine
was added at 0ꢀC until the solution becomes clear ꢀan excess of dimethylamine is recommended). A
solution of 320 mg ꢀ1.03 mmol) 11a in 15cm3 1,2-dichloroethane was added dropwise to the AlCl3±
NHMe2 solution within 30min. The reaction mixture was stirred at 0ꢀC for 1 h; then the temperature
was increased to 40±45ꢀC, and stirring was continued at this temperature for additional 16h. The
proceeding of the reaction was monitored via TLC ꢀpetrol ether:ethyl acetate:triethylamine ˆ 79:20:1).
When the reaction was completed, 50 cm3 1% KOH were added, and the organic layer was separated.
The aqueous layer was extracted with CH2Cl2 ꢀ3Â 30cm3), and the combined organic layers were
washed with H2O ꢀ3Â 20cm3) and dried over MgSO4. Chromatography on silica ꢀpetrol ether:ethyl
acetate:triethylamineˆ 79:20:1) afforded 213 mg ꢀ0.63 mmol, 61.4%) of pure 12.
M.p.: 92±94ꢀC; 1H NMR: ꢈ ˆ 0.93±1.03 ꢀm, 1H), 1.05±1.17 ꢀm, 2H), 1.21±1.44 ꢀm, 3H), 1.73±1.95
ꢀm, 5H), 2.22 ꢀs, 6H), 2.31 ꢀdd, J1 ˆ 3.2 Hz, J2 ˆ 16.02 Hz, 1H), 3.55 ꢀd, J ˆ 8.83Hz, 1H), 3.99 ꢀt,
J ˆ 2.2 Hz, 1H), 4.0 ꢀs, 5H), 4.03 ꢀs, 1H), 4.25 ꢀs, 1H) ppm; 13C NMR: ꢈ ˆ 26.52 ꢀCH2), 26.82 ꢀCH2),
32.36 ꢀCH2), 33.15 ꢀCH2), 34.59 ꢀCH2), 39.84 ꢀCH), 41.67 ꢀCH3), 44.61 ꢀCH), 64.65, 64.81, 65.35,
‡
67.25, 69.37 ꢀCp), 87.53 ꢀCp), 91.81ꢀCp) ppm; MS m=z ꢀ%) ˆ 338.3 ꢀ24.1), 337.3 ꢀ100, M ), 294.1
ꢀ21.5), 293.2 ꢀ75.3), 292.2 ꢀ41.1), 291.2 ꢀ93.6), 255.0 ꢀ56.0), 253.0 ꢀ29.8), 225.0 ꢀ23.1), 221.0 ꢀ17.5),
164.9 ꢀ25.1); [ꢀ]20 ꢀnm)ˆ ‡ 56.6 ꢀ589), ‡ 61.3 ꢀ578)ꢀ ꢀc ˆ 0.865, CHCl3); CD ꢀCH2Cl2): ꢉmaxꢀÁ") ˆ
263 ꢀ ‡ 0.45), 293 ꢀ À 0.12), 330 ꢀ ‡ 0.05), 456 ꢀ ‡ 0.18) nm.
Racemic trans-2-ferrocenoyl-cyclohexane-1-carboxylic acid ꢀ13; C18H20FeO3)
Asolution of 11.0 g ꢀ59.1 mmol) of ferrocene and of 5.0 g ꢀ32.5 mmol) of trans-cyclohexane-1,2-
dicarboxylic acid anhydride in 100 cm3 of CH2Cl2 was added dropwise to a suspension of 8.0 g
ꢀ59.9 mmol) of AlCl3 in 100 cm3 of CH2Cl2. The mixture was stirred at room temperature for 2 h
and subsequently poured on ice. The pH was adjusted to 2±3 with phosphoric acid ꢀ50%), and the
organic layer was separated. The aqueous layer was extracted twice with 80cm3 CH2Cl2, and the
combined organic layers were washed twice with 100 cm3 of H2O. After the organic phase was reduced
to approximately 200 cm3 in vacuo it was extracted with a 2 N aqueous NaOH solution ꢀ4Â 250 cm3).
Traces of organic solvents were removed from the aqueous layers in vacuo, and the product was
precipitated by slow addition of H3PO4 ꢀ50%) at 5ꢀC. The dark orange product was ®ltered off, washed
with H2O, and dried over CaCl2 to yield 7.3 g ꢀ22.8mmol, 68%) of 13.
M.p.: decomposition above 175ꢀC; 1H NMR: ꢈ ˆ 1.10±1.60 ꢀm, 4H), 1.70±1.90 ꢀm, 2H),
2.20±2.20 ꢀm, 2H), 2.70±2.90 ꢀm, 1H), 2.90±3.10 ꢀm, 1H), 4.20 ꢀs, 5H), 4.50 ꢀs, 1H), 4.70 ꢀs, 1H),
4.90 ꢀs, 1H) ppm; 13C NMR: ꢈ ˆ 25.40 ꢀCH2), 25.80 ꢀCH2), 29.30 ꢀCH2), 31.00 ꢀCH2), 43.80, 67.90,
‡
70.0 ꢀCp), 71.80, 71.90, 77.80 ꢀCp), 180.90 ꢀCO), 207.00 ꢀCO) ppm; MS: m=z ꢀ%): 340.1 ꢀ100, M ),
294.1 ꢀ10.4), 213.0 ꢀ56.2), 199.1 ꢀ12.6), 186.0 ꢀ49.3), 165.1 ꢀ10.3), 129.0 ꢀ39.4), 121.0 ꢀ59.0), 109.0
ꢀ20.7), 94.9 ꢀ22.4), 81.0 ꢀ18.8), 55.9 ꢀ39.3), 48.9 ꢀ10.0), 41.0 ꢀ15.0).

Novel Ferrocene-Based Aminophosphine Ligands
1007
Racemic trans-2-ferrocenylmethyl-cyclohexane-1-carboxylic acid monomethylester
ꢀ14; C19H24FeO2)
To a solution of 3.4 g ꢀ10.6 mmol) of 13 in 110 cm3 of MeOH, 17cm3 of benzene, 17 cm3 of H2O, and
17cm3 of concentrated HCl, freshly prepared zinc amalgam was added, and the mixture was re¯uxed
for 15 h. After cooling to room temperature the amalgam was ®ltered off and washed with benzene.
The organic layer was washed with brine ꢀ2Â 100 cm3) and H2O ꢀ2Â 50cm3) and dried over MgSO4.
After ®ltration, a freshly prepared solution of diazomethane in diethyl ether was added, and the
reaction mixture was stirred at room temperature for 1 h. The solvent was removed in vacuo, and
the residue was chromatographed twice on silica ꢀ®rst run: petrol ether:ethyl acetate ˆ 95:5; second
run: CH2Cl2:petrol ether ˆ 75:25) to yield 2.09 g ꢀ6.14 mmol, 58%) of 14.
M.p.: 37±39ꢀC; 1H NMR: ꢈ ˆ 0.80±0.90 ꢀm, 1H), 1.10±1.20 ꢀm, 2H), 1.30±1.40 ꢀm, 1H),
1.50±1.70 ꢀm, 4H), 1.80±1.90 ꢀm, 1H), 2.00±2.10 ꢀm, 1H), 2.10 ꢀdd, J1 ˆ 8.1 Hz, J2 ˆ 14.1Hz,
1H), 2.40 ꢀdd, J1 ˆ 3.9 Hz, J2 ˆ 13.8Hz, 1H), 3.70 ꢀs, 3H), 4.00 ꢀs, 4H), 4.10 ꢀs, 5H) ppm; 13C NMR:
ꢈ ˆ 25.30 ꢀCH2), 25.50 ꢀCH2), 30.20 ꢀCH2), 30.70 ꢀCH2), 35.40 ꢀCH2), 40.60, 49.20, 51.40, 67.10
ꢀCp), 67.30 ꢀCp), 68.50 ꢀCp), 69.10 ꢀCp), 85.80 ꢀCp), 176.60 ꢀCO) ppm; MS: m=z ꢀ%) ˆ 340.2 ꢀ100,
‡
M ), 212.8 ꢀ16.5), 198.9 ꢀ70.0), 168.0 ꢀ10.6), 154.9 ꢀ14.0), 139.9 ꢀ41.1), 125.8 ꢀ30.1), 120.7 ꢀ57.5),
107.8 ꢀ45.0), 80.8 ꢀ81.8), 66.9 ꢀ25.2), 55.8 ꢀ20.6), 54.8 ꢀ16.1), 40.8 ꢀ20.4), 38.8 ꢀ12.7).
Racemic trans-2-ferrocenylmethyl-cyclohexane-1-carboxylic acid ꢀ15; C18H22FeO2)
Asuspension of 1.0g ꢀ2.94 mmol) of 14 in 50cm3 of aqueous KOH ꢀ20%) was re¯uxed under Ar for
25h. After ®ltration, the solution was cooled with ice, and the product was precipitated with H3PO4
ꢀ50%). The yellow powder was washed with H2O and dried over CaCl2 to yield 0.680g ꢀ2.08 mmol,
70.7%) of crude 15 which was used in the subsequent reaction without further puri®cation.
5
04aS ,8aR ,Rm)-04-Oxo-0ꢇ -5,6,7,8-tetrahydro-9H-benz[f]indenyl))-
Ã
à Ã
5
Ã
à Ã
0ꢇ -cyclopenta-dienyl)-ion0II) ꢀ16; C18H20FeO) and 04aR ,8aR ,Sm)-
04-oxo-0ꢇ5-5,6,7,8-tetrahydro-9H-benz[f]indenyl))-0ꢇ5-cyclopentadienyl)-ion0II)
ꢀ7; C18H20FeO)
To a solution of 1 cm3 ꢀ7.2mmol) of tri¯uoroacetic acid anhydride in 60 cm3 CH2Cl2, 7 g of molecular
3
Ê
sieve ꢀ4A) and ꢀdropwise) a suspension of 650mg ꢀ1.99 mmol) of 14 in 60cm CH2Cl2 were added at
À 15 to À 5ꢀC. After stirring at room temperature for 4 h the reaction mixture was ®ltered over celite.
The mixture was hydrolyzed with saturated aqueous NaHCO3 solution, and the organic layer was
separated. The aqueous layer was extracted twice with 50cm3 of CH2Cl2, and the combined organic
layers were washed four times with 70cm3 of H2O. After drying over MgSO4 and removal of the
solvent in vacuo, the residue was chromatographed on silica ꢀCH2Cl2:petrol ether ˆ 95:5) to afford
481 mg ꢀ1.56 mmol, 78%) of 16 and 25 mg ꢀ0.08 mmol, 4%) of 7. For physical and spectroscopic data
of 7 and 16, see above.
5
Ã
Ã
à Ã
04R ,4aS ,8aR ,Rm)-04-Hydroxy-0ꢇ -5,6,7,8-tetrahydro-4H,9H-benz[f]-indenyl))-
0ꢇ5-cyclopentadienyl)-iron0II) ꢀ17; C18H22FeO)
To a suspension of 35 mg ꢀ0.9 mmol) of LiAlH4 in 5 cm3 dry diethyl ether, a solution of 0.19g
ꢀ0.6mmol) of 16 in 15cm3 dry diethyl ether was added dropwise at À 10 to À 15ꢀC. The mixture
was stirred at this temperature for an additional hour and then for 1 h at room temperature. The excess
of hydride was destroyed by slow addition of 5 cm3 of H2O, and the reaction mixture was ®ltered over
celite. The organic layer was separated and washed twice with 25cm3 of H2O. After drying over
MgSO4 and removal of the solvent in vacuo, the residue was chromatographed on silica ꢀCH2Cl2) to
afford 147 mg ꢀ0.47 mmol, 78%) of 17.

1008
M. Weiûenbacher et al.
1
M.p.: 142±144ꢀC; H NMR: ꢈ ˆ 0.93±1.13 ꢀm, 2H), 1.14±1.34 ꢀm, 3H), 1.40±1.51 ꢀm, 1H),
1.64±1.85 ꢀm, 3H), 1.76 ꢀd, J ˆ 9.3 Hz, OH), 2.25±2.39 ꢀm, 2H), 2.39±2.46 ꢀm, 1H), 3.76 ꢀt, J ˆ
9.3 Hz, 1H), 3.99 ꢀm, 1H), 4.05 ꢀm, 1H), 4.11 ꢀs, 5H), 4.18 ꢀm, 1H) ppm; 13C NMR: ꢈ ˆ 25.93 ꢀCH2),
25.99 ꢀCH2), 30.09 ꢀCH2), 31.99 ꢀCH2), 34.04 ꢀCH2), 37.78 ꢀCH), 48.27 ꢀCH), 61.96 ꢀCp), 65.94 ꢀCp),
‡
66.26 ꢀCp), 69.12 ꢀCp), 71.55 ꢀCH), 86.37 ꢀCp), 91.45 ꢀCp) ppm; MS: m=z ꢀ%) ˆ 310 ꢀ100, M ), 292
ꢀ39), 263 ꢀ3), 225 ꢀ24), 199 ꢀ7), 172 ꢀ39), 165 ꢀ16), 129 ꢀ46), 121 ꢀ14), 91 ꢀ15), 81 ꢀ6), 56 ꢀ9), 43 ꢀ2).
01S,2R)-Cyclohexane-1,2-dicarboxylic acid monomethylester ꢀ19; C9H14FeO4)
Asuspension of 1 g of PtO in 50 cm3 MeOH was hydrogenated for 40min in a Parr apparatus at a
2
pressure of 4 bar H2. After addition of 4.8 g ꢀ26 mmol) of ꢀ1S,2R)-cyclohex-4-ene-1,2-dicarboxylic
acid monomethylester ꢀcaution: Ptꢀ0) is highly pyrophoric!), the mixture was hydrogenated for 18h at
a pressure of 5 bar. The catalyst was ®ltered off, and MeOH was removed in vacuo to afford a yellow
oil which was dried in vacuo to yield 4.4 g ꢀ24mmol, 92%) of 19.
M.p.: 63±65ꢀC; 1H NMR: ꢈ ˆ 1.32±1.61 ꢀm, 4H), 1.68±1.83 ꢀm, 2H), 1.93±2.07 ꢀm, 2H),
2.76±2.88 ꢀm, 2H), 3.66 ꢀs, 3H) ppm; 13C NMR: ꢈ ˆ 23.59 ꢀCH2), 23.72 ꢀCH2), 25.91 ꢀCH2), 26.23
ꢀCH2), 42.31 ꢀCH), 42.51 ꢀCH), 51.66 ꢀCH3), 174.02 ꢀCO), 180.13 ꢀCO) ppm; MS: m=z ꢀ%) ˆ 186
‡
ꢀ2, M ), 168 ꢀ54), 155 ꢀ71), 140 ꢀ84), 126 ꢀ85), 99 ꢀ16), 95 ꢀ10), 87 ꢀ32), 81 ꢀ100), 67 ꢀ75), 45 ꢀ13);
[ꢀ]20 ꢀnm)ˆ ‡ 4.9 ꢀ589), ‡ 5.2 ꢀ578), ‡ 21.5 ꢀ365)ꢀ ꢀc ˆ 2.37, EtOH).
01S,2R)-2-Ferrocenoyl-cyclohexane-1-carboxylic acid monomethylester ꢀ20; C19H22FeO3)
To a solution of 3.5g ꢀ19 mmol) of 19 in 25cm3 of dry benzene, 1.7 cm3 ꢀ21 mmol) of pyridine and
4.2 cm3 ꢀ56 mmol) of SOCl2 were added. After stirring the solution for 2 h at room temperature and for
additional 18 h at 40ꢀC the solvent was removed in vacuo. In order to get rid of SOCl2, the residue was
dissolved twice in 50cm3 benzene which was subsequently removed under reduced pressure. Amix-
ture of an oil and of colourless crystals ꢀpyridinium hydrochloride) was obtained. The product was
dissolved in 40 cm3 of diethyl ether and stored at 0ꢀC.
The solution of the acid chloride was ®ltered under exclusion of H2O over celite into a three necked
¯ask; then, a solution of 3.5 g ꢀ19 mmol) of ferrocene in 40 cm3 of CS2 was added. After addition of a
solution of 5 g ꢀ38 mmol) of AlCl3 in 30 cm3 of dry diethyl ether ꢀcaution: dissolving AlCl3 in diethyl
ether is a highly exothermic reaction!) the reaction mixture was re¯uxed for 4 h, and the product was
separated as a highly viscous oil. After cooling the mixture to room temperature and separation of the
solvent, the residue was hydrolyzed with 150 cm3 of diluted HCl ꢀpHˆ 3±4). The product precipitates
as an orange oil. The aqueous layer was extracted twice with 100 cm3 of H2O, and the combined
organic layers were washed three times with 100 cm3 of H2O. After drying over MgSO4 and evapora-
tion of the solvent the residue was chromatographed on silica ꢀpetrol ether:ethyl acetateˆ 9:1) to afford
2.2 g ꢀ6.2mmol, 33%) of an orange oil.
1H NMR: ꢈ ˆ 1.23±1.49 ꢀm, 3H), 1.72±1.83 ꢀm, 2H), 1.86±1.95 ꢀm, 1H), 2.08±2.20 ꢀm, 1H),
2.23±2.32 ꢀm, 1H), 2.58 ꢀm, 1H), 3.47 ꢀm, 1H), 3.67 ꢀs, 3H), 4.24 ꢀs, 5H), 4.45 ꢀm, 1H), 4.48 ꢀm, 1H),
4.69 ꢀm, 1H), 4.83 ꢀm, 1H) ppm; 13C NMR: ꢈ ˆ 22.91 ꢀCH2), 24.31 ꢀCH2), 25.46 ꢀCH2), 28.95 ꢀCH2),
42.75 ꢀCH), 46.57 ꢀCH), 51.54 ꢀCH3), 69.02 ꢀCp), 69.83 ꢀCp), 70.10 ꢀCp), 71.64 ꢀCp), 71.75 ꢀCp),
‡
78.16 ꢀCp), 174.76 ꢀCO), 206.00 ꢀCO) ppm; MS: m=z ꢀ%) ˆ 354 ꢀ100, M ), 323 ꢀ21), 289 ꢀ5), 257
ꢀ
ꢀ16), 213 ꢀ82), 185 ꢀ76), 149 ꢀ20), 129 ꢀ66), 121 ꢀ66), 92 ꢀ19), 81 ꢀ18), 56 ꢀ27), 43 ꢀ13); [ꢀ]20
ꢀnm) ˆ ‡ 27.6 ꢀ589), ‡ 27.8 ꢀ578)ꢀ ꢀc ˆ 1.375, CHCl3).
Grignard cross coupling reactions 0general procedure)
Asolution of 0.03mmol of the catalyst precursor ꢀ 1 Á PdCl2 or 2 Á PdCl2) in diethyl ether was degassed,
and 0.7 cm3 ꢀ10 mmol) of vinylbromide and 21mmol of a solution of phenylethylmagnesium chloride
in diethyl ether were added at À 78ꢀC. After stirring for 24h at 0ꢀC, 1.5 g of dry ice were added, and

Novel Ferrocene-Based Aminophosphine Ligands
1009
the reaction mixture was stirred for additional 15min. The solution was extracted with H2O, and the
organic layer was separated. The aqueous layer was extracted with diethyl ether, and the combined
organic layers were washed once with saturated NaHCO3 solution and twice with H2O. After drying
over MgSO4 the solvent was evaporated, and the residue was puri®ed by bulb-to-bulb distillation
1
ꢀ100ꢀC, 14torr). The chemical purity was determined by H NMR spectroscopy, the enantiomeric
purity by GC on FS-Lipodex C, 50 m  0.25 mm ꢀcarrier: H2, column temperature: 29ꢀC, split: 1:100,
injector temperature: 175ꢀC, sample: 30mg=cm3 in CH2Cl2, injected 0.5mm3).
X-Ray analyses
Crystallographic data were collected on a modi®ed STOE diffractometer using graphite monochro-
Ê
mated MoKꢀ radiation ꢀꢉ ˆ 0.71069A). Data collections performed at low temperatures involved the
use of a home-built coldstream low-temperature device. Structures were solved with direct methods
and re®ned with least-squares, including anisotropic atomic displacement parameters for all non-
hydrogen atoms. No absorption or extinction corrections were applied, and H atoms were included
at calculated positions. Data pertaining to the crystallographic characterization, data collection, and
structure re®nement are summarized in Table 1, computer programs used for structure solution and
re®nement are listed in Ref [9]. Atomic coordinates have been deposited at the Cambridge Crystal-
lographic Data Centre ꢀdeposition numbers CCDC 180283±180289).
Acknowledgments
È
This work was kindly supported by the Fonds zur Fo rderung der wissenschaftlichen Forschung
ꢀProject No. 9859-CHE).
References
[1] a) Jedlicka B, Kratky C, Weissensteiner W, Widhalm M ꢀ1993) J Chem Soc Chem Commun 1329;

b) Wally H, Schlogl K, Weissensteiner W, Widhalm M ꢀ1993) Tetrahedron Asymm 4: 285; c)
Wally H, Kratky C, Weissensteiner W, Widhalm M, Schlogl K ꢀ1993) J Organomet Chem 450: 185

[2] Hayashi T ꢀ1995) In: Togni A, Hayashi T ꢀeds) Ferrocenes. VCH, Weinheim, p 113; b) Schwink L,
Knochel P ꢀ1998) Chem Eur J 4: 950; c) Ogasawara M, Hayashi T ꢀ2000) In: Ojima I ꢀed)
Catalytic Asymmetric Synthesis. Wiley-VCH, NY, p 651 and references cited therein
[3] Hayashi T, Konishi M, Fukushima M, Mise T, Kagotani M, Tajika M, Kumada M ꢀ1982) J Am
Chem Soc 104: 180
[4] Ohno A, Yamane M, Hayashi T, Oguni N, Hayashi M ꢀ1995) Tetrahedron Asymm 6: 2495
[5] For analogous reactions see: a) Hill EA, Richards JH ꢀ1961) J Am Chem Soc 83: 4216; b) Schlogl

K, Fried M, Falk H ꢀ1964) Monatsh Chem 95: 576; c) Schlogl K, Falk H ꢀ1964) Angew Chem 76:



570; d) Falk H, Schlogl K ꢀ1965) Monatsh Chem 96: 266; e) Falk H, Schlogl K ꢀ1965) Monatsh
Chem 96: 1065
[6] For analogous reactions see: a) Dixneuf P ꢀ1971) Tetrahedron Lett 19: 1561; b) Dixneuf P, Dabard
R ꢀ1972) Bull Soc Chim Fr 7: 2847
[7] Wally H, Nettekoven U, Weissensteiner W, Werner A, Widhalm M ꢀ1998) Enantiomer 2: 441
[8] PCMODEL, Serena Software, Box 3076, Bloomington, USA
[9] a) Sheldrick GM ꢀ1986) SHELXS-86, AComputer Program for Crystal Structure Solution, Univ

Gottingen, Germany; b) Sheldrick GM ꢀ1997) SHELXL-97, AProgram for the Re®nement of
Crystal Structures from Diffraction Data, Univ Gottingen, Germany; c) SHELXTL ꢀ1994) Version

5, Siemens Analytical Instruments, Inc.
Received January 4, 2002. Accepted January 16, 2002