Aminoalkylferrocenyldichlorophosphanes: Facile synthesis of versatile chiral starting materials

Article abstract of DOI:10.1039/b617257a

A series of racemic and optically pure aminoalkylferrocenyldichlorophosphanes has been prepared by reaction of phosphorus trichloride with the corresponding lithiated aminoalkylferrocene precursors. Crystal structures of racemic 1-dichlorophosphanyl-2-N,N- dimethylaminomethylferrocene, racemic 1-dichlorophosphanyl-2-N,N- dimethylaminomethyl-3-triphenylsilylferrocene and (S)-N,N-dimethyl-1-[(R)-2- (dichlorophosphanyl)ferrocenyl]ethylamine reveal short intramolecular N...P distances, which are suggestive of weak N → P dative bonds. The aminoalkylferrocenyldichlorophosphanes can be used for the preparation of the corresponding primary phosphanes, one of which was characterised by X-ray crystallography. Optically pure (R)-N,N-dimethyl-1-[(S)-2-(phosphanyl) ferrocenyl]ethylamine can easily be lithiated twice to give the first enantiomerically pure lithium-phosphorus closo cluster compound, which was also structurally characterised. The Royal Society of Chemistry.

Full text of DOI:10.1039/b617257a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Aminoalkylferrocenyldichlorophosphanes: facile synthesis of versatile chiral
starting materials†
Steffen Tschirschwitz, Peter Lo¨nnecke and Evamarie Hey-Hawkins*
Received 27th November 2006, Accepted 31st January 2007
First published as an Advance Article on the web 19th February 2007
DOI: 10.1039/b617257a
A series of racemic and optically pure aminoalkylferrocenyldichlorophosphanes has been prepared by
reaction of phosphorus trichloride with the corresponding lithiated aminoalkylferrocene precursors.
Crystal structures of racemic 1-dichlorophosphanyl-2-N,N-dimethylaminomethylferrocene, racemic
1-dichlorophosphanyl-2-N,N-dimethylaminomethyl-3-triphenylsilylferrocene and (S)-N,N-dimethyl-
1-[(R)-2-(dichlorophosphanyl)ferrocenyl]ethylamine reveal short intramolecular N · · · P distances,
which are suggestive of weak N → P dative bonds. The aminoalkylferrocenyldichlorophosphanes can
be used for the preparation of the corresponding primary phosphanes, one of which was characterised
by X-ray crystallography. Optically pure (R)-N,N-dimethyl-1-[(S)-2-(phosphanyl)ferrocenyl]ethylamine
can easily be lithiated twice to give the first enantiomerically pure lithium–phosphorus closo cluster
compound, which was also structurally characterised.
low coordination number. In addition to the known phosphanyl-
Introduction
ferrocene12 and 1,1ꢀ-bis(phosphanyl)ferrocene,13 we recently re-
Phosphane-based ligands with a ferrocene backbone have been
ported on the synthesis of primary aminoalkylferrocenylphos-
of interest in organometallic and coordination chemistry since
phanes by reduction of the corresponding ferrocenylphos-
the first synthesis of 1,1ꢀ-bis(diphenylphosphanyl)ferrocene in
the mid-1960s.1 Chiral ferrocenylphosphanes were first prepared
by Hayashi et al. in the 1970s.2 Their synthetic strategy was
based on the fact that enantiomerically pure N,N-dimethyl-1-
ferrocenylethylamine (7), first prepared and optically resolved by
Ugi et al.,3 can be functionalised diastereoselectively to obtain
planar-chiral ferrocene derivatives.4 Since this pioneering work,
many ferrocene-containing ligands have been developed and
coordinated to catalytically active metal centres.5 The number of
applications of these complexes in homogeneous catalysis has been
growing ever since. Initially used for asymmetric hydrogenation,6
Grignard reactions7 or hydrosilylations,2 further functionalisa-
tion, such as replacement of the dimethylamino group by another
phosphanyl group or variation of the substituents at the phospho-
rus atoms, have created specific ligands for highly active catalyst
systems, a well-known example being the JOSIPHOS ligand.8
Dichlorophosphanes have proved to be extremely versatile
precursors for the synthesis of organophosphorus compounds.
However, ferrocenyldichlorophosphanes, which are of interest due
to the reactivity of the P–Cl bonds, are not readily accessible.
While 1,1ꢀ-bis(dichlorophosphanyl)ferrocene, as well as its dihalo
homologues, can be obtained in reasonable yield,9,10 the synthesis
of the monosubstituted derivative, ferrocenyldichlorophosphane,
is associated with severe problems and gives only low yields.11
Primary phosphanes, in which a ferrocene unit is attached to the
phosphorus atom, are not very well investigated, though they are
promising precursors for phosphanido and phosphinidene com-
plexes, as well as other compounds containing phosphorus in a
phonates.14 We showed that racemic 2-(N,N-dimethylamino-
methyl)phosphanylferrocene (5) can easily be dilithiated. The
resulting dilithium phosphanediide crystallises as a hexamer
containing all-R- or all-S-configured units with a central Li12P6
cluster, which, according to theoretical studies, can be described
as a closo cluster according to Wade’s rules.15
We herein report the facile and convenient synthesis of
novel racemic and enantiomerically pure aminoalkylferrocenyl-
dichlorophosphanes. These compounds can easily be converted
to the corresponding primary phosphanes. Furthermore, we
present the structure of the first enantiomerically pure dilithium
phosphanediide, which forms an aggregate with a central Li12P6
closo cluster.
Results and discussion
Readily available aminoalkylferrocenylphosphanes were used as
starting materials for the syntheses of the ferrocenyldichlorophos-
phanes. As shown in Scheme 1, N,N-dimethylaminomethyl-
ferrocene (1)16 was further functionalised by ortho-selective lithia-
tion and subsequent treatment with an electrophile. Compound
1 was also used as a starting material for racemic 2-N,N-
dimethylaminomethyl-1-triphenylsilylferrocene (2), one of whose
ortho-positions is blocked with a sterically demanding triphenylsi-
lyl group, which was chosen to study steric influences on the
chemistry of the target compounds and to explore the possibility
of introducing potential silyl anchor groups for immobilization
on a solid support. Silane 2 is readily available by reaction of
ortho-lithiated 1 with triphenylsilyl chloride. It is a crystalline
compound whose crystal structure is shown in Fig. 1. It has not yet
appeared in the literature, although similar compounds, such as its
tin analogue, a para-fluorophenyl-substituted derivative and the
Institut fu¨r Anorganische Chemie der Universita¨t Leipzig, Johannisallee 29,
Leipzig, Germany. E-mail: hey@rz.uni-leipzig.de; Fax: +49-341-9739319
† Parts of these results were previously reported at the XXII ICOMC
in Zaragoza, Spain, 23–28 July, 2006, as oral (O35) and poster (P47)
presentation.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 1377–1382 | 1377
©

View Article Online
Scheme 1
2-pyridyl)bis(trimethylsilyl)methyl]phosphorus dichloride,20 but
rather short in comparison to those reported for ter-
tiary aminomethylarylphosphanes.21 The P–Cl bonds [P1–Cl1
2.1167(6), P1–Cl2 2.1887(7)] are longer than in structurally
characterised aryldichlorophosphanes22 and 1,1ꢀ-bis(dichloro-
phosphanyl)ferrocene,10,23 which in general lie between 2.05 and
˚
2.08 A. The unusually long P1–Cl2 distance (trans to the weak
N → P bond) is also indicative of a dative interaction between
P and N and in accordance with comparable compounds.20 1H
NMR measurements at room temperature show only one singlet
for the two methyl groups attached to nitrogen, that is, the dative
bond is not retained in solution.
The molecular structure of the dichlorophosphane resulting
from 2 as starting material is depicted in Fig. 3. Compared to
the structures of 2 and the corresponding primary phosphane
6 (Fig. 4), the dimethylamino group is bent into the plane of
the cyclopentadienyl ring towards the phosphorus atom, and this
leads to a weak N → P dative bond, similar to that observed for
Fig. 1 Molecular structure of 2 (only the R-enantiomer is shown) with
thermal ellipsoids at 50% probability level. H atoms have been omitted for
clarity.
corresponding enantiopure compound derived from (R)-7, were
reported.17–19
˚
3, with an N · · · P distance of 2.511(1) A. In this case P1, C1, C2,
C11, and N1 are not coplanar, but again the distance between the
The precursors 1 and racemic 2 were used as starting materials
for the syntheses of the corresponding ferrocenyldichlorophos-
phanes (Scheme 1). Both were lithiated ortho-selectively and
treated subsequently with a slight excess of PCl3 at low tempera-
ture. The resulting crystalline dichlorophosphanes (racemic 3 and
4) are stable in the solid state but decompose slowly in solution,
especially in non-polar solvents such as n-hexane.
phosphorus atom and the chlorine atom trans to the weak N → P
˚
bond [P1–Cl1 2.1635(6) A] is significantly longer than the distance
˚
to the second chlorine atom [P1–Cl2 2.0818(6) A].
The molecular structure of 3 (Fig. 2) reveals a distorted
tetrahedral environment at the P atom. A noteworthy feature is the
almost perfect coplanarity of P1, C1, C2, C11 and N1, which is due
to a weak N → P dative bond. Thus, the system can be regarded
as two annelated coplanar five-membered rings. The N · · · P
˚
distance of 2.443(2) A is longer than that reported for [(6-methyl-
Fig. 3 Molecular structure of 4 (only the S-enantiomer is shown) with
thermal ellipsoids at 50% probability level. H atoms have been omitted
◦
˚
for clarity. Selected bond lengths (A) and angles ( ): P1–Cl1 2.1635(6),
P1–Cl2 2.0818(6), P1–C1 1.795(2), P1 · · · N1 2.511(1), Si1–C3 1.864(2),
C1–P1–Cl1 96.39(5), C1–P1–Cl2 102.21(6), Cl1–P1–Cl2 94.43(3).
Starting from enantiomerically pure (S)–N,N-dimethyl-1-
ferrocenylethylamine ((S)-7)3 and following the same synthetic
pathway (Scheme 2), (S,R)-8‡ was prepared with a diastereomeric
excess of 92%. The minor diastereomer could easily be removed
Fig. 2 Molecular structure of 3 (only the R-enantiomer is shown) with
thermal ellipsoids at 50% probability level. H◦atoms have been omitted for
˚
clarity. Selected bond lengths (A) and angles ( ): P1–Cl1 2.1167(6), P1–Cl2
‡ In all compounds with C-centre and planar chirality, the first letter
describes the configuration at the carbon centre and the second the
configuration of the substituted cyclopentadienyl ring.1
2.1887(7), P1–C1 1.800(1), P1 · · · N1 2.443(2), C1–P1–Cl1 99.36(5),
C1–P1–Cl2 95.55(5), Cl1–P1–Cl2 94.50(3).
1378 | Dalton Trans., 2007, 1377–1382
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
than reported by us previously.14 Phosphane 6 is a crystalline solid
whose structure is depicted in Fig. 4. It represents, to the best of
our knowledge, the first structurally characterised free primary
phosphanylferrocene without any spacer between the phosphanyl
group and the ferrocene moiety. It is air-stable in the solid state
and it is also one of the very rare examples of primary phosphanes
that are stable in solution towards oxygen over a prolonged period
of time (at least more than two weeks).24 All protons, including
those at the phosphorus atom, could be located. The P–C bond
is within the same range as in other reported aromatic primary
phosphanes.25,26 Also, within the standard deviation, the P–H dis-
tances are in a similar range as those observed in other structurally
characterised primary phosphanes for which the protons were also
located, while the H–P–H angle is slightly bigger.25,27
The diastereomerically pure primary phosphane (S,R)-9 is
obtained analogously from (S,R)-8 and LiAlH4 (Scheme 2), as
is its enantiomer (R,S)-9 from (R,S)-8. We have already shown
that 5 can be lithiated twice and that the resulting dilithium
phosphanediide forms a hexameric aggregate with a central
Li12P6 unit which can be described as a closo cluster.14 To extend
this chemistry, a similar reaction was performed with (R,S)-9 as
starting material.
Fig. 4 Molecular structure of 6 (only the S-enantiomer is shown) with
thermal ellipsoids at 50% probability level. H atoms (except at P1) have
◦
˚
been omitted for clarity. Selected bond lengths (A) and angles ( ): P1–H1p
1.40(5), P1–H2p 1.33(5), P1–C1 1.809(3), H1p–P1–H2p 98(3).
Scheme 2 ‡.
by recrystallisation. The corresponding enantiomer (R,S)-8 could
be obtained analogously starting from (R)-7.
(1)
Enantiomerically pure 8 crystallises in the chiral space group
P212121 [Flack parameter x = −0.01(1)]. The structure (Fig. 5)
resembles that of 3, but due to steric repulsion between the
methyl group (C12) at the chiral carbon atom and the ferrocenyl
moiety, P1, C1, C2, C11 and N1 are not coplanar. Therefore, the
With two equivalents of n-butyllithium, deprotonation occurred
at 0 ◦C in n-hexane to give a deep red solution and some red precip-
itate, which dissolved upon adding a few drops of THF. Dark violet
octahedral crystals of (R,S)-10 slowly formed. The compound
crystallises in the chiral space group P212121 [Flack parameter x =
−0.02(1)] and reveals the expected hexameric structure of like-
configurated ferrocenylphosphanediide units associated through
phosphorus–lithium contacts to a central Li12P6 cluster with
almost the same geometrical properties as reported for dilithiated
5 (Fig. 6). To the best of our knowledge, (R,S)-10 is the first
enantiomerically pure LiP cluster, and it can be described as a
closo cluster according to Wade’s rules.15
˚
interaction between P and N is weaker [P1 · · · N1 2.574(2) A], but
may still be regarded as a dative bond. This is also reflected in
the different P–Cl bond lengths: again the P–Cl bond trans to
˚
P · · · N is significantly longer [P1–Cl1 2.1553(8) A]. Compound
8 is more stable in solution than 3, probably due to the weaker
P · · · N interaction and therefore slightly lower reactivity of the
P–Cl bonds.
Conclusions
We have reported a facile method of preparing three novel
racemic (3 and 4) and optically pure (8) aminoalkylferro-
cenyldichlorophosphanes, which were characterised by X-ray
crystallography and whose potential as starting materials for the
syntheses of a wide variety of ferrocene-containing phosphorus
compounds is presently being investigated further. We demon-
strated that these aminoalkylferrocenyldichlorophosphanes can
easily be converted to the corresponding primary phosphanes (5,
6 and 9), which are thus now conveniently accessible in good yields.
The structure of a rare example of an air-stable primary phosphane
(6) could be determined. Finally, the synthesis and structure of the
first chiral LiP closo cluster compound was presented.
Fig. 5 Molecular structure of (S,R)-8 with thermal ellipsoids at 50%
probability level. H atoms (except H11) have been omitted for clarity. Se-
◦
˚
lected bond lengths (A) and angles ( ): P1–Cl1 2.1553(8), P1–Cl2 2.0841(7),
P1–C1 1.787(2), P1 · · · N1 2.574(2), C1–P1–Cl1 96.38(6), C1–P1–Cl2
101.45(6), Cl1–P1–Cl2 94.64(3).
Experimental
The dichlorophosphanes are suitable starting materials for
the syntheses of the corresponding primary phosphanes. Thus,
compounds 5 and 6 were obtained by reaction of crude 3 or 4
with LiAlH4 (Scheme 1) in shorter time and much better yield
General considerations
All manipulations were carried out using standard Schlenk
techniques under an atmosphere of dry high-purity nitrogen.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 1377–1382 | 1379
©

View Article Online
58.8 (CH2N), 66.5 (CFcSi), 69.3 (C5H5), 70.4 (C5H3), 73.9 (C5H3),
76.5 (C5H3), 91.1 (CFcCH2N), 127.5 (m-CPh), 129.1 (p-CPh), 135.8
(i-CPh), 136.4 (o-CPh) ppm. FT-IR m˜/cm−1 (KBr): 3093 (w), 3069
(m), 3047 (w), 3013 (w), 2993 (w), 2973 (w), 2944 (w), 2854 (w),
2811 (m), 2757 (s), 1485 (w), 1455 (m), 1429 (s), 1400 (w), 1380 (w),
1360 (w), 1321 (w), 1258 (m), 1226 (w), 1188 (w), 1171 (m), 1154
(w), 1136 (m), 1107 (s), 1081 (w), 1042 (m), 1023 (m), 999 (m), 954
(w), 846 (w), 833 (w), 816 (s), 759 (w), 744 (s), 705 (s), 681 (m), 638
(w), 552 (m), 527 (m), 511 (s), 493 (s), 450 (w), 443 (w). MS (EI
pos., 14 eV) m/z: 501 [M+] (100%), 457 [(M–NMe2)+] (38.7%), 259
[(SiPh3)+] (25.5%). Elemental analysis for C31H31FeNSi (%): Calc.:
C 74.24, H 6.23, N 2.79; found: C 74.21, H 6.26, N 2.71.
Racemic 1-dichlorophosphanyl-2-N,N-dimethylaminomethyl-
ferrocene (3). A solution of n-BuLi (9.47 ml, 22.63 mmol,
2.39 M in n-hexane) was slowly added to a solution of 1 (5.0 g,
20.58 mmol) in diethyl ether (50 ml). The mixture was stirred
for 8 h and slowly (over a period of 2 h) added to a solution of
PCl3 (3.39 g, 24.69 mmol) in diethyl ether (20 ml) at −80 ◦C. The
mixture was warmed to room temperature overnight and filtered.
The solvent was evaporated and the residue dried in vacuo.
Recrystallisation from a mixture of toluene and n-hexane (1 :
10) gave pure 3 as red crystals. Yield 5.88 g (83.1%). Mp. 117 ◦C
(decomp.). H NMR (+20 ◦C, 400.13 MHz, C6D6/TMS): d =
1
Fig. 6 Molecular structure of (R,S)-10. H atoms (except at chiral centres)
have been omitted for clarity.
2
2.00 (s, 6H, N(CH3)2), 2.65 (d, JHH = 14.0 Hz, 1H, CH2), 3.37
(d, 2JHH = 14.0 Hz, 1H, CH2), 3.90 (br. s, 1H, C5H3), 3.96 (s, 5H,
1
C5H5), 4.12 (br. s, 1H, C5H3), 4.76 (br. s, 1H, C5H3) ppm. 13C{ H}
THF was distilled from sodium/benzophenone. Diethyl ether,
toluene, n-hexane and dichloromethane were taken from an
MBRAUN Solvent Purification System MB SPS-800 and
stored over potassium mirror. Ethanol was degassed in a
nitrogen stream in an ultrasonic bath. 1H, 7Li, 13C and 31P
NMR spectra were recorded on a Bruker Avance DRX 400
spectrometer and referenced to tetramethylsilane (TMS).28
Mass spectra were recorded on a VG Analytics ZAB-HSQ
spectrometer. FT-IR spectra were recorded on a Perkin-Elmer
Spectrum 2000 spectrometer. Melting points were determined
in sealed glass capillaries under nitrogen and are uncorrected.
N,N-dimethylaminomethylferrocene (1)16 and both enantiomers
of N,N-dimethyl-1-ferrocenylethylamine (7)3 were synthesised
according to literature procedures. Other chemicals were obtained
from commercial sources and used as supplied, except PCl3,
which was distilled and saturated with nitrogen prior to use.
◦
NMR (+20 C, 100.16 MHz, C6D6/TMS): d = 45.4 (N(CH3)2),
57.8 (d, 3JPC = 2.4 Hz, CH2), 70.0 (br. s, C5H3), 70.7 (br. s, C5H3),
1
70.8 (C5H5), 71.0 (br. s, C5H3), 81.5 (d, JPC = 63.3 Hz, CFcP),
◦
2
1
92.6 (d, JPC = 13.5 Hz, CFcCH2) ppm. 31P{ H} NMR (+20 C,
161.98 MHz, C6D6/TMS): d = 145.6 ppm. FT-IR m˜/cm−1 (KBr):
3095 (w), 2956 (w), 2865 (m), 2833 (m), 2786 (w), 1460 (m), 1409
(w), 1347 (w), 1261 (w), 1225 (m), 1165 (m), 1106 (w), 1077 (w),
1030 (s), 1002 (m), 954 (w), 825 (s), 503 (m), 488 (w), 458 (m),
438 (s). MS (EI pos. 70 eV) m/z: 343 [M+] (61.8%), 308 [(M–Cl)+]
(9.1%), 265 [(M–Cl–HNMe2)+] (14.1%), 187 [(Fe(C5H4)PCl +
H)+] (100%), 152 [(Fe(C5H4)P + H)+] (57.4%), 109 [(CH2(C5H3)P
+ H)+] (39.4%). Elemental analysis for C13H16Cl2FeNP (%): Calc.:
C 45.39, H 4.69, N 4.07; Found: C 45.31, H 4.74, N 3.96.
Racemic 1-dichlorophosphanyl-2-N,N-dimethylamino-methyl-3-
triphenylsilylferrocene (4). The procedure for the preparation of
3 was followed using 2 (2.00 g, 3.99 mmol), n-BuLi (1.70 ml,
4.06 mmol, 2.39 M in n-hexane) and PCl3 (0.57 g, 4.15 mmol).
Recrystallisation from a mixture of toluene and n-hexane (1 :
5) gave pure product. Crystals for X-ray analysis were obtained
at −20 ◦C from a dichloromethane solution which was layered
Synthetic procedures
Racemic 2-N,N-dimethylaminomethyl-1-triphenylsilylferrocene
(2). A solution of n-BuLi (9.47 ml; 22.63 mmol; 2.39 M in n-
hexane) was slowly added to a solution of 1 (5.00 g, 20.58 mmol)
in diethyl ether (50 ml). The mixture was stirred for 8 h and cooled
to 0 ◦C. A solution of triphenylsilyl chloride (6.80 g, 23.06 mmol)
in diethyl ether (50 ml) was then added over a period of 2 h.
After stirring overnight at room temperature, the mixture was
filtered and the filtrate evaporated to dryness. The orange solid
was recrystallised from diethyl ether. Yield 8.74 g (84.7%). Mp.
116–117 ◦C. 1H NMR (+20 ◦C, 400.13 MHz, CDCl3/TMS): d =
1.76 (s, 6H, N(CH3)2), 2.80 (d, 2JHH = 12.8 Hz, 1H, CH2N), 2.96
◦
1
with n-hexane. Yield 2.04 g (84.9%). Mp. 129–130 C. H NMR
◦
(+20 C, 400.13 MHz, C6D6/TMS): d = 1.83 (s, 6H, N(CH3)2),
2
4
2.79 (dd, JHH = 14.0 Hz, JPH = 1.6 Hz, 1H, CH2N), 3.58 (d,
3
2JHH = 14.0 Hz, 1H, CH2N), 4.04 (s, 5H, C5H5), 4.24 (d, JHH
=
3
2.4 Hz, 1H, C5H2 o to Si), 5.07 (d, JHH = 2.4 Hz, 1H, C5H2 o
to P), 7.19 (m, 9H, m- and p-H of C6H5), 7.64 (d, 3JHH = 6.4 Hz,
6H, o-H of C6H5) ppm. 13C{ H} NMR (+20 ◦C, 100.16 MHz,
1
C6D6/TMS): d = 45.1 (N(CH3)2), 58.7 (d, 3JPC = 3.9 Hz, CH2N),
68.7 (CFcSi), 71.8 (C5H5), 74.3 (C5H2 o to Si), 79.1 (C5H2 o to P),
87.0 (d, 1JPC = 65.6 Hz, CFcP), 97.7 (d, 2JPC = 8.6 Hz, CFcCH2N),
130.1 (p-CPh), 135.1 (i-CPh), 136.5 (o-CPh) ppm. The signal for◦m-
2
(d, JHH = 12.8 Hz, 1H, CH2N), 4.02 (s, 5H, C5H5), 4.11 (s, 1H,
C5H3), 4.36 (s, 1H, C5H3), 4.66 (s, 1H, C5H3), 7.38 (m, 9H, m- and
3
p-H of C6H5), 7.67 (d, JHH = 7.2 Hz, 6H, o-H of C6H5) ppm.
13C NMR (25 ◦C, 100.6 MHz, CDCl3/TMS): d = 44.8 (N(CH3)2),
1380 | Dalton Trans., 2007, 1377–1382
CPh is overlapped by the solvent signal. 31P{ H} NMR (+20 C,
1
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
161.98 MHz, C6D6/TMS): d = 134.0 ppm. FT-IR m˜/cm−1 (KBr):
3068 (m), 3048 (m), 2959 (m), 2866 (m), 2833 (m), 2787 (w), 1469
(m), 1428 (s), 1409 (m), 1338 (w), 1311 (w), 1261 (m), 1217 (m),
1188 (w), 1127 (m), 1108 (s), 1021 (m), 1004 (m), 934 (w), 826 (m),
740 (m), 704 (s), 591 (w), 533 (w), 506 (s), 467 (s). MS (EI pos.
14 eV) m/z: 529 [(M–2HCl)+] (15.7%), 407 [(M–2HCl–HNMe2–
Ph)+] (9.1%), 294 [(SiPh3Cl)+] (100%), 217 [(SiPh2Cl)+] (8.6%), 154
[(Ph2)+] (19.3%), 66 [(CpH)+] (15.5%). MS (EI pos. 70 eV, scan
from m/z = 540 to m/z = 660 only, due to short lifetime of M+)
m/z: 601 [M+] (100%), 566 [(M–Cl)+] (15.9%), 551 [(M–Cl–Me)+]
(29.1%) Elemental analysis for C31H30Cl2FeNPSi (%): Calc.: C
61.81, H 5.02, N 2.33; Found: C 61.90, H 4.99, N 2.27.
12.5 Hz, 1H, CH2N), 3.72 (dd, 1JPH = 199.9 Hz, 2JHH = 12.2 Hz,
1
2
1H, PH), 3.89 (dd, JPH = 198.0 Hz, JHH = 12.2 Hz, 1H, PH),
3
3.93 (s, 5H, C5H5), 4.25 (d, JHH = 2.2 Hz, 1H, C5H2 o to Si),
3
4.32 (d, JHH = 2.2 Hz, 1H, C5H2 o to P), 7.21 (m, 9H, m- and
1
p-H of C6H5), 7.84 (m, 6H, o-H of C6H5) ppm. 13C{ H} NMR
(+20 ◦C, 100.16 MHz, C6D6/TMS): d = 44.6 (N(CH3)2), 57.1 (d,
3
3JPC = 3.2 Hz, CH2N), 70.3 (d, JPC = 1.0 Hz, CFcSi), 70.8 (d,
3
1JPC = 8.6 Hz, CFcP), 70.9 (C5H5), 78.0 (d, JPC = 2.8 Hz, C5H2
2
2
o to Si), 78.9 (d, JPC = 6.9 Hz, C5H2 o to P), 95.9 (d, JPC
=
11.9 Hz, CFcCH2N), 127.9 (m-CPh), 129.5 (p-CPh), 136.3 (i-CPh),
136.9 (o-CPh) ppm. 31P NMR (+20 ◦C, 161.98 MHz, C6D6/TMS):
d = −149.2 (t, 1JPH = 198.4 Hz) ppm. FT-IR m˜/cm−1 (KBr): 3068
(m), 3047 (m), 2969 (m), 2939 (m), 2857 (m), 2811 (s), 2762 (s),
2297 (m, PH), 2255 (m, PH), 1483 (w), 1456 (m), 1427 (s), 1398
(w), 1342 (w), 1317 (w), 1257 (m), 1218 (w), 1178 (w), 1127 (s),
1106 (s), 1076 (m), 1044 (w), 1026 (m), 1003 (m), 843 (m), 818
(s), 741 (m), 703 (s), 681 (m), 638 (w), 583 (m), 499 (s), 456 (m).
MS (EI pos. 14 eV) m/z: 533 [M+] (10.6%), 488 [(M–HNMe2)+]
(100%). Elemental analysis for C31H32FeNPSi (%): Calc.: C 69.79,
H 6.05, N 2.63; Found: C 69.77, H 5.99, N 2.57.
(S)-N,N -Dimethyl-1-[(R)-2-(dichlorophosphanyl)-ferrocenyl]-
ethylamine ((S,R)-8). The same procedure as for 3 was followed
using (S)-7 (3.00 g, 11.67 mmol), n-BuLi (5.37 ml, 12.83 mmol,
2.39 M in n-hexane) and PCl3 (1.93 g, 14.05 mmol) giving orange-
red plates from toluene and n-hexa◦ne (1 : 10). Yield 3.30 g (79.0%).
◦
1
Mp. 121–122 C. H NMR (+20 C, 400.13 MHz, C6D6/TMS):
3
d = 0.72 (d, JHH = 6.8 Hz, 3H, CH3), 1.86 (s, 6H, N(CH3)2),
3.89 (m, 3H, 2H of C5H3 and CH–CH3), 3.99 (s, 5H, C5H5), 4.79
(br. s, 1H, C5H3) ppm. 13C{ H} NMR (+20 ◦C, 100.16 MHz,
1
(S)-N,N-Dimethyl-1-[(R)-2-(phosphanyl)ferrocenyl]ethylamine
((S,R)-9). The procedure for the synthesis of 5 was applied
starting from crude (S,R)-8 resulting from (S)-7 (1.00 g,
3.89 mmol). Yield 0.95 g (84.5% based on (S)-7).
C6D6/TMS): d = 8.5 (CH–CH3), 38.9 (N(CH3)2), 58.7 (CH–CH3),
68.9 (d, 3JPC = 1.1 Hz, C5H3 m to P and to CHN), 70.2 (br. s, C5H3
o to CHN), 70.9 (C5H5), 71.7 (d, 2JPC = 2.0 Hz, C5H3 o to P), 82.7
(d, 1JPC = 62.6 Hz, CFcP), 97.5 (d, 2JPC = 14.5 Hz, CFcCHN) ppm.
◦
1
31P{ H} NMR (+20 C, 161.98 MHz, C6D6/TMS): d = 141.9 ppm.
FT-IR m˜/cm−1 (KBr): 3110 (w), 2978 (m), 2946 (m), 2876 (m), 2837
(w), 2790 (w), 1469 (m), 1451 (s), 1410 (w), 1373 (m), 1335 (w),
1294 (w), 1244 (s), 1186 (m), 1170 (m), 1105 (m), 1074 (m), 1044
(w), 1002 (m), 933 (m), 839 (m), 816 (s), 776 (w), 556 (w), 513 (w),
494 (w), 465 (s), 449 (s). MS (EI pos. 70 eV) m/z: 357 [M+] (20.8%),
201 [(M–Fe–Cp–Cl)+] (22.1%), 166 [(M–Fe–Cp–2Cl)+] (100%),
(R,S)-10. (R,S)-9 (0.2 g, 0.69 mmol) was dissolved in n-hexane
(10 ml). At 0 ◦C n-BuLi (0.59 ml, 1.4 mmol, 2.39 M in n-hexane)
was slowly added. Immediately, a fluffy red precipitate formed
which redissolved almost completely upon further addition of n-
BuLi. The mixture was allowed to warm to room temperature
and stirred for 30 min. THF was slowly added until the solid was
completely dissolved (ca. 1 ml). Deep violet octahedral crystals
of (R,S)-10 grew at room temperature over a few days. Yield
+
+
=
134 [(M–Fe–Cp–PCl2) ] (15.2%), 72 [(CH3CH NMe2) ] (37.1%).
Elemental analysis for C14H18Cl2FeNP (%): Calc.: C 46.97, H 5.07,
N 3.91; Found: C 47.04, H 4.99, N 3.79.
◦
1
0.22 g (ca. 95%, crystals contain solvent). 31P{ H} NMR (+20 C,
161.98 MHz, THF/C6D6/TMS), d = −176.6 ppm. 31P{ H} NMR
1
(+20 ◦C, 161.98 MHz, n-hexane/C6D6/TMS), d = −219.0 (m, v.
br.) ppm. 7Li NMR (+20 ◦C, 155.50 MHz, THF/C6D6/TMS), d =
0.8 ppm. 7Li NMR (+20 ◦C, 155.50 MHz, n-hexane/C6D6/TMS),
d = 11.5 (m, v. br.) ppm. The spectroscopic data of (R,S)-10 are
in agreement with those reported previously.14
Racemic 1-phosphanyl-2-N,N-dimethylaminomethylferrocene
(5). Crude 3 can be used directly without further purification.
Crude 3, obtained from 1 (1.0 g, 4.12 mmol), was dissolved
in diethyl ether (40 ml, solution became slowly turbid due to
decomposition of 3) and was added to a slurry of LiAlH4 (0.15 g,
3.95 mmol) in diethyl ether (10 ml) at 0 ◦C. The mixture was
stirred at room temperature overnight and carefully hydrolysed
with slightly basic (one prill of KOH) water (25 ml) at 0 ◦C. After
stirring for 2 h at room temperature the organic layer was separated
and dried over MgSO4. After filtration the solvent was removed
and the oily residue dried in vacuo. 5 can be further purified by
bulb-to-bulb distillation at 80 ◦C bath temperature in vacuum
(ca. 10−3 torr). Yield 0.92 g (81.2% based on 1). The spectroscopic
data of 5 are in agreement with those reported previously.14
X-ray crystallography
Suitable crystals were mounted in perfluoropolyalkyl ether. Crys-
tallographic measurements were made using a Stoe-IPDS imaging
plate diffractometer for compounds 2, 3 and (R,S)-10, a Siemens
SMART CCD diffractometer for compounds 6 and (S,R)-8 and
an Oxford Diffraction Xcalibur S diffractometer for compound
4. The structures were solved by direct methods and refined on
F2 by full-matrix least-squares techniques (SHELX97).29 All non-
hydrogen atoms, except for some solvent molecules, were refined
anisotropically, hydrogen atoms were either located and refined
isotropically or included in a riding mode. A highly disordered
solvent molecule (n-hexane) in the structure of compound 4
and two highly disordered solvent molecules (n-hexane) in the
structure of compound (R,S)-10 were removed using the program
SQUEEZE.30 Crystal data and details of data collection and
refinement are given in Table 1.
Racemic 1-phosphanyl-2-N,N-dimethylaminomethyl-3-triphe-
nylsilylferrocene (6). The procedure for the synthesis of 5
was followed starting from crude 4 obtained from 2 (2.00 g,
3.99 mmol). The yellow-orange solid was recrystallised from
ethanol. Crystals were grown from a diethyl ether solution at
◦
−20 ◦C. Yield 1.77 g (83.1% based on 2). Mp. 143–144 C. H
1
NMR (+20 ◦C, 400.13 MHz, C6D6/TMS): d = 1.75 (s, 6H,
2
2
N(CH3)2), 3.14 (d, JHH = 12.5 Hz, 1H, CH2N), 3.21 (d, JHH
=
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 1377–1382 | 1381
©

View Article Online
CCDC reference numbers are 628375 (2), 617194 (3), 628376
(4), 628374 (6), 617193 ((S,R)-8) and 617195 ((R,S)-10).
For crystallographic data in CIF or other electronic format see
DOI: 10.1039/b617257a
Acknowledgements
The authors gratefully acknowledge support from the Deutsche
Forcschungsgemeinschaft (Graduate College 378).
Notes and references
1 For an overview see: A. Togni and T. Hayashi, Ferrocenes. Homogeneous
Catalysis, Organic Synthesis, Materials Science, VCH, Weinheim, 1995.
2 T. Hayashi, K. Yamamoto and M. Kumada, Tetrahedron Lett., 1974,
4405.
3 D. Marquading, H. Klusacek, G. W. Gokel, P. Hoffmann and I. K.
Ugi, J. Am. Chem. Soc., 1970, 92, 5389.
4 T. Hayashi, T. Mise, M. Fukushima, M. Kagotani, N. Nagashima, Y.
Hamada, A. Matsumoto, S. Kawakami, M. Konishi, K. Yamamoto
and M. Kumada, Bull. Chem. Soc. Jpn., 1980, 53, 1138.
5 W. R. Cullen and J. D. Woollins, Coord. Chem. Rev., 1981, 39, 1.
6 W. R. Cullen and J. D. Woollins, Can. J. Chem., 1982, 60, 1793.
7 T. Hayashi, M. Tajika, K. Tamao and M. Kumada, J. Am. Chem. Soc.,
1976, 98, 3718.
8 A. Togni, C. Breutel, A. Schnyder, F. Spindler, H. Landert and A.
Tijani, J. Am. Chem. Soc., 1994, 116, 4062.
9 M. T. Reetz, A. Gosberg, R. Goddard and S.-H. Kyung, Chem.
Commun., 1998, 2077.
10 C. Moser, A. Orthaber, M. Nieger, F. Belaj and R. Pietschnig, Dalton
Trans., 2006, 3879.
11 G. R. Knox, P. L. Pauson and D. Wilson, Organometallics, 1992, 11,
2930.
12 C. Spang, F. T. Edelmann, M. Noltemeyer and H. W. Roesky, Chem.
Ber., 1989, 122, 1247.
13 M. J. Burk and M. F. Gross, Tetrahedron Lett., 1994, 35, 9363.
14 S. Tschirschwitz, P. Lo¨nnecke and E. Hey-Hawkins, Angew. Chem.,
2005, 117, 3025, (Angew. Chem., Int. Ed., 2005, 44, 2965).
15 K. Wade, Adv. Inorg. Chem. Radiochem., 1976, 18, 1.
16 J. K. Lindsay and C. R. Hauser, J. Org. Chem., 1957, 22, 355.
17 J. Azizian, R. M. G. Roberts and J. Silver, J. Organomet. Chem., 1986,
303, 397.
18 C. Beyer, U. Bo¨hme, C. Pietzsch and G. Roewer, J. Organomet. Chem.,
2002, 654, 187.
19 M. Murakami, R. Minamida, K. Itami, M. Sawamura and Y. Ito,
Chem. Commun., 2000, 2293.
20 T. R. van den Ancker, P. C. Andrews, S. J. King, J. E. McGrady, C. L.
Raston, B. A. Roberts, B. W. Skelton and A. H. White, J. Organomet.
Chem., 2000, 607, 213.
21 F. Carre´, C. Chuit, R. J. P. Corriu, P. Monforte, C. Reye´, J. Declercq
and A. Dubourg, Angew. Chem., 1993, 105, 1529, (Angew. Chem., Int.
Ed., 1993, 32, 1430).
22 R. J. Wehmschulte, M. A. Khan and S. I. Hossain, Inorg. Chem., 2001,
40, 2756; P. Kilian, A. M. Z. Slawin and D. J. Woollins, Chem.–Eur. J.,
2003, 9, 215; S. A. Reiter, S. D. Nogai and H. Schmidbaur, Dalton
Trans., 2005, 247.
23 E. Niecke, M. Nieger and R. Pietschnig, Phosphorus, Sulfur Silicon
Relat. Elem., 1999, 147, 325.
24 R. M. Hiney, L. J. Higham, H. Mu¨ller-Bunz and D. G. Gilheany, Angew.
Chem., 2006, 118, 7406, (Angew. Chem., Int. Ed., 2006, 45, 7248).
25 S. A. Reiter, S. D. Nogai, K. Karaghiosoff and H. Schmidbaur, J. Am.
Chem. Soc., 2004, 126, 15833.
26 R. A. Bartlett, M. M. Olmstead, P. P. Power and G. A. Sigel, Inorg.
Chem., 1987, 26, 1941.
27 M. Brynda, M. Geoffroy and G. Bernardinelli, Chem. Commun., 1999,
961.
28 R. K. Harris, E. D. Becker, S. M. Cabral de Menezes, R. Goodfellow
and P. Granger, Pure Appl. Chem., 2001, 73, 1795.
29 SHELX97 (Includes SHELXS97, SHELXL97, SHELXH97): G. M.
Sheldrick, (1997); SHELX97. Programs for Crystal Structure Analysis
(Release 97-2), University of Go¨ttingen, Germany.
30 P. v. d. Sluis and A. L. Spek, Acta Crystallogr., Sect. A, 1990, 46, 194.
1382 | Dalton Trans., 2007, 1377–1382
This journal is
The Royal Society of Chemistry 2007
©