The highly regiospecific synthesis and crystal structure determination of 1,1′-2,5′ substituted ring-locked ferrocenes

Article abstract of DOI:10.1016/j.jorganchem.2009.01.049

1,1′-Ferrocene biscarboxaldehyde (1) has been prepared and the aldehyde groups were subsequently protected with acetal groups to produce 1,1′-bisacetalferrocene (2). A ring-locked ferrocene was synthesised by further derivatisation of the cyclopentadiene

Full text of DOI:10.1016/j.jorganchem.2009.01.049

Journal of Organometallic Chemistry 694 (2009) 2020–2028
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
The highly regiospecific synthesis and crystal structure determination
of 1,10-2,50 substituted ring-locked ferrocenes
a
a
b
b
b
Arthur Connell a, , Peter J. Holliman , Ian R. Butler , Louise Male , Simon J. Coles , Peter N. Horton ,
*
Michael B. Hursthouse b, William Clegg c, Luca Russo c
a School of Chemistry, Bangor University, Gwynedd LL57 2UW, UK
b EPSRC National Crystallography Service, School of Chemistry, University of Southampton, Southampton SO17 1BJ, UK
c School of Chemistry, Newcastle University, Newcastle upon Tyne NE1 7RU, UK
a r t i c l e i n f o
a b s t r a c t
1,10-Ferrocene biscarboxaldehyde (1) has been prepared and the aldehyde groups were subsequently
protected with acetal groups to produce 1,10-bisacetalferrocene (2). A ring-locked ferrocene was synthes-
ised by further derivatisation of the cyclopentadiene rings at the 2,20 positions with phosphine substitu-
ents to produce 2,20-bis-(acetal)-1,10-diphenylphosphinoferrocene (3), which was subsequently
coordinated to either a nickel chloride (5) or nickel bromide (6) metal centre. The ring-locked ferrocene
complexes produced 2,50-bis-(acetal)-1,10-diphenylphosphinoferrocene substitution patterns. The acetal
protecting groups of 2,20-bis-(acetal)-1,10-diphenylphosphinoferrocene were removed to produce 1,10-
bis-carboxaldehyde-2,20-diphenylphosphinoferrocene (4). The Cp rings of 1,10-bisacetalferrocene were
also further derivatised at the 2,20 positions with a silane to produce the ring-locked 1,10-siloxane-2,50-
bisacetalferrocenophane (7). The acetal protecting groups were removed from this to produce 1,10-silox-
ane-2,50-ferrocenophanecarboxaldehyde (8). For both the phosphine and siloxane electrophiles, the sub-
stitution on the Cp rings gives chiral products (obtained as racemic mixtures). Due to the highly
regioselective nature of the reaction and diastereoselectivity in the products only C2-symmetric com-
pounds were observed without the presence of meso diastereoisomers. Subsequent ring-locking forced
the Cp rings to rotate, leading to 1,10-ring-locked ferrocenes with 2,50-arrangement of the acetal groups
(i.e. on opposite faces of the ferrocene unit).
Article history:
Received 6 November 2008
Received in revised form 20 January 2009
Accepted 28 January 2009
Available online 5 February 2009
Keywords:
1,10 2,50-Ferrocene derivatives
Ortholithiation
Ferroceneophane
Ring-locked
Regiospecific
Chiral
Ó 2009 Elsevier B.V. All rights reserved.
1. Introduction
cene. In related work, Ingham et al. reported the synthesis of
monomeric, dimeric and polymeric ferrocenylacetylides such as
Substituted ferrocenes are useful compounds in catalysis and
molecular electronics and this requires regioselective synthesis of
target ferrocenes. Ortholithiation is a versatile tool to achieve
1,10,2,20-substituted ferrocenes and ring-locked ferrocenes have
been widely studied [1,2]. Here, we have investigated combining
these two strategies for the regioselective derivatisation of ferro-
cene. This approach is important to produce key target molecules;
for example, important targets for conducting polymers and
molecular wires are poly-1,10-ethynylferrocenes. To produce such
targets, the synthesis of 1,10-ethynylferrocene must first be
achieved. However, previous attempts to produce this monomer
have suffered intramolecular dimerisation of the ethynyl groups,
generating a ferrocenophane [1]. To try to prevent this, Pudelski
et al. reported the synthesis and deprotection of 1,10-bis((trimeth-
ylsilyl)ethynyl)ferrocene precursors [1,2]. However, on deprotec-
tion, these molecules underwent rapid intramolecular coupling
to form ferrocenophanes rather than the desired bis-ethynyl ferro-
1,10-bis(phenylethynyl)ferrocene [3] by coupling 1,10-diiodoferro-
cene with a specific quantity of phenyltrimethylstannane deriva-
tives using Pd(PPh3)4 as a catalyst.
Butler et al. reported key monomers for poly-1,10-ethynylferro-
cene using 10–bromo- and 10-iodoferrocenylethyne in a selective
coupling sequence [4]. This did produce the desired polymer but
the synthesis was limited by the number of by-products including
oligomers and ferrocene dimeric diynes. Ethynyl groups have also
been used to attach ferrocene units to conducting polymers [5,6]
and alkyne groups have been used to link ferrocene to chromoph-
ores (e.g. ferrocenylanthracenes and ferrocenyl bipyridines) by pal-
ladium-catalysed coupling [7]. Other examples of ferrocenyl
linkages include the first bis-alkynylferrocene complexes with flu-
orene spacers [8] and dyads and polyads of ferrocenylanthraqui-
nones, acridones and acridines [9–11]. Ethynyl ferrocene spacer
groups have also been attached to terpyridyl units by coupling a
terminal acetylene group to 40-[[(trifluoromethyl)sulfinyl]oxy]-
2,20:6,200-terpyridine using a Pd(PPh3)4 catalyst, with these com-
pounds showing fast quantitative information transfer in work
aimed at developing molecular wires [12,13]. In a related study,
* Corresponding author. Tel.: +44 1248 351 151; fax: +44 1248 370528.
E-mail address: chs612@bangor.ac.uk (A. Connell).
0022-328X/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2009.01.049

A. Connell et al. / Journal of Organometallic Chemistry 694 (2009) 2020–2028
2021
Long et al. reported the synthesis of alkynylferrocene and biferro-
cene ligands and their platinum-containing dimers and oligomers
[14] to study the electronic interaction between organometallic
units and the donor/acceptor role of the platinum. Similar studies
have been reported for ferrocene units bridged by silane or sulfide
spacers [15].
In this paper, the synthetic strategy adopted has been to pre-
pare 1,10-ring-locked ferrocenes to prevent rotation of the cyclo-
pentadienyl groups and to achieve 2,50-substitution of the Cp
rings to produce substituents on the top and bottom rings on oppo-
site sides of the ring lock (Schemes 1 and 2). This type of approach
could, in the future, be used to prevent intramolecular dimerisa-
tion of 1,10-ethynylferrocene and hence lead to poly-1,10-ethynyl-
ferrocene. In this study, electrophilic substitution on the Cp rings
gives chiral products (obtained as racemic mixtures). Due to the
highly regioselective nature of the reaction and diastereoselectivity
in the products only C2-symmetric compounds were observed
without the presence of meso diastereoisomers. Most reports for
synthesising tetrasubstituted ferrocenes follow a diastereoselec-
tive ortholithiation strategy applied to ferrocenes bearing one chi-
ral directing group on each Cp ring. The chiral directing groups
used have included an amino group [17], an oxazoline [18], and
further work by Balavoine et al. who reported variations in the dia-
stereoselectivity after ortholithiation of the 1,10-bis-aminoalkoxide
[19,20] and 1,10-bisacetal [16]. Ortholithiation of the latter was
highly regioselective towards formation of the meso diastereoiso-
mer with an overall yield of 8% reported [16]. By changing the pro-
tecting group to an aminoalkoxide, the regioselective nature of the
reaction changes and the diastereoselective formation of C2-sym-
metric tetrasubstituted ferrocenes was favoured giving an overall
yield of 28% with an enantiomeric excess >99% [19,20]. In both
cases, the ferrocenophane was formed by deprotecting and reduc-
ing the aldehydes to alcohols, then dehydrating to lock the Cp rings
via an ether group. of the Such C2-symmetric tetrasubstituted fer-
rocenes have already found successful application in asymmetric
catalysis, e.g. palladium-catalysed asymmetric cross-coupling
[17] and asymmetric allylic substitution [21]. In this paper, alde-
hyde groups protected using 1,3-propanediol were chosen as pre-
O
CHO
1, 3-Propanediol,
n-BuLi/ TMEDA
TsOH, Benzene
O
O
Fe
Fe
Fe
-78oC, THF, DMF
Reflux
CHO
O
2
1
Scheme 1.
PPh2
TsOH, THF,
H2O, RT
CHO
Fe
CHO
PPh2
4
O
O
PPh2
O
-78 oC, t-BuLi, RT
Ph2
NiX2DME2,
CH2Cl2
-78 oC, ClPPh2, RT
P
5 X = Cl
6 X = Br
O
O
Fe
MX2
Fe
P
Diethyl ether
Ph2
O
O
PPh2
O
3
2
O
O
CHO
TsOH, THF,
H2O, RT
SiMe2
O
SiMe2
O
Fe
Fe
-78 oC, t-BuLi, RT
SiMe2
SiMe2
-78 oC, N,N’-
(Me2N)Me2SiCl, RT
CHO
O
O
7
8
6
7
Scheme 2.

2022
A. Connell et al. / Journal of Organometallic Chemistry 694 (2009) 2020–2028
cursors to the acetylene function. The protection of ferrocenecarb-
oxaldehyde with 1,3-propanediol has been reported during the
preparation of ethynylferrocenes including the first 1,2,3,4,5-pen-
taethynyl derivative [22–24].
Ferrocene is conveniently disubstituted using a range of precur-
sors such as 1,10-bishalomercurioferrocene or 1,10-ferrocenylbo-
ronic acid [25,26]; however, the dilithium salt formed in situ by
treatment with butyllithium and tetramethylethylenediamine
(TMEDA) is considered the most practical [27–29].
This paper reports the investigation of the regioselective syn-
thesis of 1,10-2,20-substituted ferrocenes by ortholithiation of ace-
tal-protected precursors, followed by ring-locking to produce
either a nickel halide complex with phosphine ligands or via a
siloxane-bridged ferrocenophane to produce 1,10-2,50-substituted
ferrocenes. Deprotection of the acetal protecting groups to form
the corresponding aldehyde-bearing compounds is also described.
three resonances in the ferrocene region and three resonances for
the acetal group carbons, which is consistent with the structure.
A clean molecular ion was observed in the mass spectrum. In addi-
tion, saturated C–H and C–O stretches for the acetal group are ob-
served by infrared spectroscopy. Finally, the structure was
confirmed using X-ray crystallography. The molecular structure
of (2) is shown in Fig. 1.
Compound (2) was then lithiated using similar conditions pre-
viously reported in the literature by Balavoine et al. for the 1,10
ortholithiation of the corresponding chiral 1,10- bisacetalferrocene
[16]. However, in this case, lithiation was followed by reaction
with ClPPh2 to give the chiral 1,10,2,20-tetrasubstituted product
(3) in 62% yield (obtained as a racemic mixture) with no trisubsti-
tuted ferrocenes being isolated from the reaction. Due to the highly
regioselective nature of the reaction and diastereoselectivity in the
products only C2-symmetric compounds were observed (i.e. the
acetal groups are oriented as shown in Scheme 2) without the pres-
ence of the meso diastereoisomer. By comparison, Balavoine et al.
reported that the major product of their reaction was the 1,2,10-tri-
substituted molecule; the desired 1,10,2,20-tetrasubstituted prod-
uct was diastereoselective towards the meso diastereomer,
which, was obtained in 8% yield [16]. Interestingly, it has been re-
ported recently that larger, methoxy-substituted (chiral) acetal
groups can cause problems during attempted ortholithiation of
the Cp ring [32]. It may be that the larger acetal group bulk caused
steric hindrance for the butyllithium base during attempted ortho-
lithiation. It has also been reported that there can be competition
between proton abstraction and nucleophilic attack with methoxy
groups attached to the acetal group [32]. However, this was not the
case for the simple non-chiral acetal group used in this work. In
addition, the acetal group remains intact during reaction, meaning
it should be possible to further derivatise the products to prepare
1,10,2,20,3,30-substituted ferrocenes.
2. Results and discussion
The strategy used in this work was to prepare a ring-locked fer-
rocenophane containing precursors to 1,10-ethynyl groups. Two
ring-locked ferrocenes have been synthesised; the first using metal
phosphines complexed to a metal centre (Scheme 1) and the sec-
ond using a siloxane bridge (Scheme 2). The overall aim, in each
case, has been to ensure the ethynyl-precursor functional groups
are orientated on opposite faces of the upper and lower Cp rings
of the ferrocene to prevent intramolecular coupling.
1,10-Ferrocenecarbaldehyde (1) was prepared using literature
methods [28] and the product identified using 1H and 13C NMR
and elemental analysis, which were consistent with the data previ-
ously reported in the literature [28,29] Compound (1) was con-
verted to 1,10-bisacetalferrocene (2) with 1,3-propanediol using
the conditions reported by Steffen et al. for the synthesis of mono-
acetalferrocene from ferrocenecarboxaldehyde [22]. To the best of
our knowledge (2) reported in this work has only been reported
once in the literature [30]. This is due to the preferred use of chiral
acetal groups during the preparation of formyl scaffolds, although
it has been reported that a similar acetal group (using 1,1-ethane-
diol) has been used for the partial protection of 1,10-ferrocenecarb-
aldehyde [31].
Compound (3) has not previously been reported in the primary
literature. Compound (3) was analysed by 1H and 13C NMR, mass
Compound (2) was obtained in 81% yield and it was crystallo-
graphically analysed. The yield is higher than the 63% quoted in
the literature [30]. Two broad singlets are present in the ferrocene
region corresponding to the
a and b hydrogens of each Cp ring [28].
Resonances for the acetal group protons were identified by com-
parison with the data reported by Steffen et al. for the monoacetal
prepared from ferrocenecarboxaldehyde [22]. 13C NMR data show
Fig. 1. Molecular structure of (2). In Figs. 1–7 displacement ellipsoids are drawn at
the 30% probability level and all hydrogen atoms are omitted for clarity.
Fig. 2. Molecular structure of (3).

A. Connell et al. / Journal of Organometallic Chemistry 694 (2009) 2020–2028
2023
spectrometry with accurate mass measurement, IR spectroscopy,
and elemental analysis. The relative orientation of the functional
groups on the Cp rings is confirmed by X-ray crystallography
(Fig. 2) as the desired C2 symmetrically substituted isomer. Trace
peaks in the 1H NMR spectrum are ascribed to small amounts of
oxidation of the phosphine ligands; this is supported by the pres-
ence of a second small peak in the 31P NMR for phosphine oxide.
Deprotection of (3) gives the phosphine aldehyde (4) with C2 sym-
metry (Fig. 3) but attempts to synthesise a ring-locked complex
from (4) by coordination to nickel(II) bromide proved unsuccessful.
However, coordination of (3) to nickel(II) chloride gave (5) and
with nickel(II) bromide gave (6); both gave crystal structures with
the orientation expected for the C2 symmetrically substituted iso-
mer (Figs. 4 and 5, respectively). A clean molecular ion of (6) was
observed in the mass spectrum whilst mass spectrometry of (5)
showed only a clean molecular ion of (m/z = 726.1749) which cor-
responds to the uncomplexed diphosphinoferrocene (3) in EI and
FAB modes implying a weakly-bound complex for (5). Character-
isation of compounds (5) and (6) was attempted using NMR and
elemental analysis. A very broad set of resonances was shown on
the 1H NMR spectrum, which, could not be assigned to the ferro-
cene or acetal hydrogens. In addition, 13C and 31P NMR spectra
gave no signals for either compound. A possible explanation for
the lack of useful data obtained when characterising the sample
using NMR could be due to paramagnetism caused by an unpaired
electron in the nickel atom. The elemental analysis data was not
presented here due to the large difference between calculated
and experimental carbon, hydrogen and nitrogen values. This could
be due to decomposition of the sample to a phosphine oxide and/or
weakly complexed metal ions detaching from the ligand over time.
The authors concluded that the crystal structure and mass spec-
trometry data were sufficiently accurate characterisation tech-
niques to confirm the identity of each compound and that the
crystal structures were by far the most important data required
to illustrate the ring twisting capability required for preparation
of compounds (7) and (8).
Fig. 4. Molecular structure of (5).
By comparison, ortholithiation of (2) and subsequent reaction
with N,N0-dimethylaminodimethylchlorosilane using the same
Fig. 5. Molecular structure of (6).
conditions as for the synthesis of (5) gives the siloxane-bridged
bisacetal ferrocenophane (7), which X-ray crystallography again
shows exhibits C2 symmetry (Fig. 6), therefore the nature of the
electrophile did not change the diastereoselectivity of the product
and a chiral product, (7), was obtained as a racemic mixture. The
consistent orientation of the bridged products (3) and (6) is good
evidence for the regioselective nature of this 1,10 ortholithiation
reaction. Finally, the siloxane-bridged acetal was deprotected un-
der mildly acidic conditions to give the corresponding aldehyde
Fig. 3. Molecular structure of (4).

2024
A. Connell et al. / Journal of Organometallic Chemistry 694 (2009) 2020–2028
of Keppler et al. in that it is a one-pot reaction, but here we used
N,N0-dimethylaminodimethylchlorosilane rather than dichlorodi-
methylsilane, obtaining a similar yield of 61% compared to the
58% previously reported.
3. Conclusion
The reaction of (2) with 2.2 equivalents of tert-butyllithium
followed by 2.2 equivalents of a phosphine electrophile results in
formation of orthosubstituted 2,20-bis(acetal)-1,10- diphenyl-
phosphinoferrocene. Complexation of this to a nickel halide effec-
tively ring locks the Cp rings of the ferrocene, creating a 1,10-diph-
osphino-2,50-bis(acetal) arrangement. A similar lithiation followed
by treatment with a silane electrophile directly yields ortho-substi-
tuted 1,10-siloxane-2,50-bis(acetal)ferrocene. Thus, these reaction
conditions and this acetal group consistently provide chiral C2-
symmetric products (as racemic mixtures). The diastereoselectivi-
ty in the product results from a regioselectivity which can be used
to synthesize ring-locked ferrocenes with the functional groups
oriented on opposite sides of the ring lock on the top and bottom
Cp rings. Hence, this acetal is an alternative protecting group to
aminoalkoxides for the regioselective synthesis of 1,10,2,20-tetra-
substituted ferrocenes. In addition, deprotection of the acetal
groups successfully yields aldehydes.
Fig. 6. Molecular structure of (7).
4. Experimental
(8). The aldehyde group appeared at 9.88 ppm in the 1H NMR and
at 192.86 ppm in the 13C NMR. In addition, resonances from the
acetal protecting group were no longer present in either spectrum.
Finally, the molecular ion of (8) was detected using mass spec-
trometry and a crystal structure was obtained to confirm the iden-
tity of the sample (Fig. 7). A final dehydration step should yield
1,10-siloxane-2,50-bisethynylferrocene with acetylene groups in
fixed position to prevent intramolecular coupling.
4.1. General procedures
All experiments were conducted under a dry nitrogen or argon
atmosphere using standard oven-dried Schlenk glassware, unless
otherwise indicated. NMR spectra were recorded on a Bruker
AC500 instrument operating at 500 MHz for 1H, 125 MHz for 13C
and 202.5 MHz for 31P. Chemical shifts (d) are given in ppm and
are relative to tetramethylsilane (for 1H and 13C) and triphenyl-
phosphine (for 31P). J values are given in Hz and refer to JH,H unless
otherwise indicated. Mass spectra were recorded using electron
impact, chemical ionisation (NH3), fast atom bombardment, elec-
trospray, matrix-assisted laser desorption ionisation and laser
desorption ionisation at the EPSRC National Mass Spectrometry
Service at the University of Swansea. Infrared spectra were re-
corded on a Perkin–Elmer 1600 series FTIR spectrometer. Elemen-
tal analysis was performed on a Carlo Erba EA1108 elemental
analyser. All dry solvents were purchased as anhydrous solvents
except THF which was dried using sodium wire.
Silicon-ferrocenophanes such as 1,1,3,3-tetramethyl-1,3-disila-
2-oxa[3]ferrocenophane are of interest in polymerisation reactions
[33–35]. This particular ferrocenophane is related to (7) and (8) but
without the additional 2,50-substitution we report here. Previous
methods to make 1,1,3,3-tetramethyl-1,3-disila-2-oxa[3]ferroce-
nophane have involved the solvolysis of precursors such as 1,10-
bis(chlorodimethylsilane)ferrocene
[36],
1,10-bis(dimethylsi-
lane)ferrocene [37] or Fe(
g
5-C5H4SiMe2XEt)2 (where X = O [37] or
N [38]. However, Keppler et al. have reported an improved synthe-
sis and a new crystal structure for 1,1,3,3-tetramethyl-1,3-disila-2-
oxa[3]ferrocenophane from 1,10-dilithioferrocene ꢀ TMEDA and
dichlorodimethylsilane in a one-pot reaction [39]. Our method to
produce the siloxane bridge of the ferrocenophane is similar to that
4.2. X-ray crystallography
Compounds (2) and (8): Suitable crystals were selected and
datasets were measured on a Bruker APEXII CCD diffractometer
at Station 9.8 of the Daresbury Synchrotron Radiation Source
((2): k = 0.6710 Å, (8): k = 0.6893 Å) at 120 K, because of the very
weak diffraction observed for these small crystals. The data collec-
tions were driven by the Bruker APEX2 software [40] and pro-
cessed by SAINT [41]. Absorption corrections were applied using
SADABS [42]. The structures were solved and refined by a full-ma-
trix least-squares procedure on F2 in the Bruker SHELXTL software
suite [43].
Compounds (3)–(7): Suitable crystals were selected and data-
sets for (3) and (7) were measured on a Bruker APEXII CCD diffrac-
tometer and for (4)–(6) on
a
Bruker-Nonius KappaCCD
diffractometer, both at the windows of a Bruker-Nonius FR591
rotating anode (kMo K = 0.71073 Å) at 120 K. The data collections
a
were driven by COLLECT [44] and processed by DENZO [45].
Absorption corrections were applied using SADABS [42]. The
Fig. 7. Molecular structure of (8). Both crystallographically independent molecules
are shown.

A. Connell et al. / Journal of Organometallic Chemistry 694 (2009) 2020–2028
2025
structures of (3), (4) and (7) were solved in SIR2004 [46] and those
of (5) and (6) were solved in SHELXS97 [43]; all five structures
were refined by a full-matrix least-squares procedure on F2 in
SHELXL97 [43].
For all structures (2)–(8) all non-hydrogen atoms were refined
with anisotropic displacement parameters. All hydrogen atoms
were added at calculated positions and refined by use of a riding
model with isotropic displacement parameters scaled to the equiv-
alent isotropic displacement parameter (Ueq) of the parent atom.
Tables 1–4 give crystal data, refinement information, and selected
molecular geometry parameters.
metry in the unit cell parameters, possibly with unresolved twin-
ning. The structure itself has been modelled satisfactorily.
Compound (6): The Rint is very high and the value of R (observed
reflections) is just over 0.1. The structure itself has been modelled
satisfactorily. The crystal was a fine needle and the diffracted
intensity was weak, leading to a low ratio of observed/unique
reflections. In addition, the shape of the crystal and the highly
absorbing nature of the compound made the process of correcting
for absorption difficult. We feel that these are adequate explana-
tions for the poor merging and refinement statistics.
Compound (7): The dataset is composed of two diffraction pat-
terns which are related by a 7.8° rotation. Attempts were made to
de-convolute the two reciprocal lattices but there were not enough
Compound (4): No obvious reason can be found in the data for
the high Rint value and it is suggested that it is due to pseudo-sym-
Table 1
Crystal data and structural refinement for compounds 2, 3, 4, 5 and 6.
2
3
4
5
6
ssg0833
C18H22FeO4
358.21
2007src0239
C42H40FeO4P2 0.75CH2Cl2
790.22
2007src0411
C36H28FeO2P2
610.37
2006src1354 (ACXT2)
C42H40Cl2FeNiO4P2
856.14
2006src1355/ACXT1
C42H40Br2FeNiO4P2 CH2Cl2
1029.99
Formula
Formula weight
Temperature (K)
Wavelength (Å)
Crystal colour and habit
Crystal size (mm)
Crystal system
Space group
a (Å)
120
120
0.71073
120
0.71073
120
0.71073
Black slab
120
0.71073
Synchrotron, 0.6710
Yellow plate
0.04 ꢁ 0.02 ꢁ 0.01
Monoclinic
P21/c
5.9067(18)
27.635(9)
9.645(3)
90
93.303(4)
90
1571.8(9)
4
Yellow slab
0.14 ꢁ 0.09 ꢁ 0.04
Monoclinic
P21/n
13.5058(7)
11.0080(4)
25.7852(12)
90
97.013(2)
90
3804.9(3)
4
Orange plate
0.22 ꢁ 0.12 ꢁ 0.01
Orthorhombic
Pbca
18.8060(7)
15.7812(7)
18.9241(9)
90
90
90
5616.3(4)
8
Green needle
0.46 ꢁ 0.03 ꢁ 0.02
Monoclinic
P21/c
22.6916(11)
9.3444(4)
20.5597(10)
90
109.871(2)
90
4099.9(3)
4
0.28 ꢁ 0.20 ꢁ 0.10
Triclinic
ꢀ
P1
9.4907(5)
10.5668(4)
21.0227(11)
76.361(3)
89.822(3)
64.009(3)
1829.46(15)
2
b (Å)
c (Å)
a
(°)
b (°)
c
(°)
V (Å3)
Z
h range (°)
Reflections collected/unique
2.78–23.54
11,598/2761
0.0737
1971
2761/0/209
1.030
2.98–25.03
30,788/6547
0.0941
4586
6547/0/469
1.110
3.00–27.48
48,635/6443
0.1959
3723
6443/0/370
1.028
2.95–27.49
32,185/8189
0.0954
5754
8189/24/469
1.086
2.92–27.48
43,976/9380
0.2396
4077
9380/0/497
1.025
Rint
Observed reflections (F2 > 2
Data/restraints/parameters
Goodness of fit (F2)
r)
R, (obs) Rw (all data)
0.0475, 0.1188
0.60, ꢂ0.55
0.0985, 0.2282
1.18, ꢂ0.58
0.0753, 0.1403
0.58, ꢂ0.46
0.0900, 0.2566
1.20, ꢂ0.78
0.1071, 0.1762
1.01, ꢂ0.85
Largest diff. peak and hole (e Åꢂ3
)
Table 2
Selected bond lengths (Å) and angles (°) for (2), (3), (4), (5) and (6).
2
3
4
5
6
C(1)–C(11)
1.481 (5) C(1)–P(1)
1.399 (4) P(1)–C(11)
1.423 (4) P(1)–C(17)
1.443 (5) C(2)–C(23)
1.429 (5) C(6)–P(2)
1.497 (5) P(2)–C(27)
1.382 (5) P(2)–C(33)
1.389 (5) C(7)–C(39)
1.446 (6)
1.815 (8)
1.849 (8)
1.831 (8)
C(1)–P(1)
P(1)–C(12)
P(1)–C(18)
1.830 (4) C(1)–P(1)
1.831 (4) P(1)–C(19)
1.838 (4) P(1)–C(25)
1.458 (6) C(5)–C(11)
1.211 (5) C(6)–P(2)
1.834 (4) P(2)–C(31)
1.837 (4) P(2)–C(37)
1.840 (5) C(7)–C(15)
1.458 (6) Ni(1)–P(1)
1.216 (5) Ni(1)–P(2)
1.814 (8)
1.825 (8)
1.819 (9)
C(1)–P(1)
P(1)–C(19)
P(1)–C(25)
1.810 (8)
C(11)–O(1)
C(11)–O(2)
O(1)–C(12)
O(2)–C(14)
C(6)–C(15)
C(15)–O(3)
C(15)–O(4)
O(3)–C(16)
O(4)–C(18)
C(1)–C(11)–O(1)
C(1)–C(11)–O(2)
C(11)–O(1)–C(12) 112.6 (3)
C(11)–O(2)–C(14) 110.9 (3)
C(6)–C(15)–O(3)
C(6)–C(15)–O(4)
C(15)–O(3)–C(16) 109.6 (4)
C(15)–O(4)–C(18) 110.4 (4)
1.808 (9)
1.831 (9)
1.479 (11)
1.805 (9)
1.812 (9)
1.812 (9)
1.494 (11)
2.353 (3)
2.311 (3)
2.346 (2)
2.349 (2)
1.510 (11) C(2)–C(11)
1.481 (11) C(2)–C(11)
1.815 (8)
1.839 (8)
1.845 (8)
1.504 (11) P(2)–C(31)
C(7)–C(24)
C(11)–O(1)
C(6)–P(2)
P(2)–C(25)
1.791 (9)
1.804 (9)
1.830 (8)
C(6)–P(2)
P(2)–C(31)
P(2)–C(37)
1.482 (11) C(7)–C(15)
2.321 (2)
2.301 (2)
2.197 (2)
2.198 (2)
Ni(1)–P(1)
Ni(1)–P(2)
Ni(1)–Br(1)
Ni(1)–Br(2)
1.446 (6)
C(24)–O(2)
108.0 (3)
107.9 (3)
C(1)–P(1)–C(11)
99.7 (3)
C(1)–P(1)–C(12)
98.8 (2)
C(1)–P(1)–C(18) 101.0 (2)
C(2)–C(11)–O(1) 124.2 (5)
C(6)–P(2)–C(25) 100.5 (2)
Ni(1)–Cl(1)
Ni(1)–Cl(2)
C(1)–P(1)–C(17) 103.8 (3)
C(2)–C(23)–O(1) 109.4 (6)
C(2)–C(23)–O(2) 104.7 (6)
C(6)–P(2)–C(27) 104.5 (4)
C(1)–P(1)–C(19)
C(1)–P(1)–C(25)
C(5)–C(11)–O(1)
C(5)–C(11)–O(2)
C(6)–P(2)–C(31)
C(6)–P(2)–C(37)
C(7)–C(15)–O(3)
C(7)–C(15)–O(4)
C(1)–P(1)–Ni(1)
C(6)–P(2)–Ni(1)
P(1)–Ni(1)–P(2)
104.3 (4)
C(1)–P(1)–C(19)
C(1)–P(1)–C(25)
C(2)–C(11)–O(1)
C(2)–C(11)–O(2)
C(6)–P(2)–C(31)
C(6)–P(2)–C(37)
C(7)–C(15)–O(3)
C(7)–C(15)–O(4)
C(1)–P(1)–Ni(1)
C(6)–P(2)–Ni(1)
P(1)–Ni(1)–P(2)
106.7 (4)
106.8 (4)
110.2 (7)
106.3 (6)
108.6 (4)
106.1 (4)
108.9 (6)
108.0 (7)
113.1 (3)
112.7 (3)
103.56 (8)
103.6 (4)
107.2 (7)
109.5 (7)
109.9 (4)
102.2 (4)
108.2 (7)
108.0 (7)
116.8 (3)
112.4 (3)
103.2 (1)
109.1 (3)
110.2 (3)
C(6)–P(2)–C(31)
C(7)–C(24)–O(2) 124.8 (4)
99.3 (2)
C(6)–P(2)–C(33)
99.6 (3)
C(7)–C(39)–O(3) 106.3 (6)
C(7)–C(39)–O(4) 108.7 (6)
Cl(1)–Ni(1)–Cl(2) 124.9 (1)
Br(1)–Ni(1)–Br(2) 122.14 (6)

2026
A. Connell et al. / Journal of Organometallic Chemistry 694 (2009) 2020–2028
non-overlapped reflections to make this possible. Thus only the
major component has been integrated, with the overlap with the
second component leading to the high Rint value. Furthermore,
the structure has been refined as a mixture of enantiomers of ratio
52:48.
sion was added dropwise n-butyllithium (2.2 eq, 71 ml,
177 mmol). The mixture was stirred at room temperature over-
night. The reaction was cooled to ꢂ78 °C and THF (150 ml) fol-
lowed by anhydrous DMF (2.2 eq, 13.7 ml, 177 mmol) were
added. The mixture was stirred at ꢂ78 °C for 15 min followed by
1.5 h stirring at room temperature. The reaction was quenched
with brine (60 ml) and CH2Cl2 (50 ml) was added. The phases were
separated and the aqueous phase was extracted with CH2Cl2
(3 ꢁ 30 ml). The combined organic phases were dried (MgSO4)
and the solvents removed in vacuo. Column chromatography
(SiO2, hexane-ether 4:1 followed by hexane-ether-ethyl acetate
4:1:0.1 gradually increasing the ether and ethyl acetate) gives
1,10-ferrocenecarboxaldehyde as a bright red crystalline solid in
the third fraction. Yield (16.07 g, 82%), 1H NMR 9.96 (s, 2H, CHO),
4.3. Preparation of 1,10-ferrocenecarboxaldehyde (1) [28]
To a solution of ferrocene (15.00 g, 81 mmol) in hexane (300 ml)
was added tetramethylethylenediamine (TMEDA) (2.2 eq, 26.6 ml,
177 mmol) and the mixture was stirred for 5 min. To this suspen-
Table 3
4.89 (s, 4H,
a
-Cp-CH), 4.68 (s, 4H, b-Cp-CH); 13C NMR 192.81
-Cp-CH), 70.87
Crystal data and structural refinement for compounds 7 and 8.
(2 ꢁ CHO), 80.34 (2 ꢁ ipso Cp-C), 74.18 (4 ꢁ
a
7
8
(4 ꢁ b-Cp-CH); Anal. Calc. for C12H10O2Fe: C, 59.56; H, 4.17. Found:
2007src0319
C22H32FeO5Si2
488.51
120
0.71073
Orange block
0.14 ꢁ 0.08 ꢁ 0.04
Orthorhombic
Pna21
19.4040(13)
8.8831(4)
13.4774(9)
90
ssg1525
C16H20FeO3Si2
372.35
C, 59.67; H, 4.29%.
Formula
Formula weight
Temperature (K)
Wavelength (Å)
Crystal colour and habit
Crystal size (mm)
Crystal system
Space group
a (Å)
4.4. Preparation of 1,10-bisacetalferrocene (2)
120
Synchrotron, 0.6893
Orange plate
0.08 ꢁ 0.04 ꢁ 0.02
Triclinic
To
a
solution of 1,10-ferrocenecarboxaldehyde (10.00 g,
41 mmol) in benzene (450 ml) 1,3-propanediol (9.9 eq, 29.3 ml,
406 mmol) and p-toluenesulfonic acid (0.16 eq, 1.25 g, 6.6 mmol)
were added. The solution was refluxed for 24 h over a Dean-Stark
trap with the exclusion of light. After addition of deactivated silica
gel (50 ml) (hexane + 10% NEt3) the solvent was removed in vacuo.
Column chromatography (SiO2, hexane + 10% NEt3) of the residue
yielded 1,10-bisacetalferrocene as a yellow powder in the first frac-
tion. Yield (11.89 g, 81%); 1H NMR 5.32 (s, 2H, acetal tertiary CH),
ꢀ
P1
7.9457(15)
14.656(3)
15.163(3)
79.463(2)
78.676(2)
89.908(2)
1701.1(6)
4
b (Å)
c (Å)
a
(°)
b (°)
90
90
2323.1(2)
4
c
(°)
V (Å3)
4.24 (brs, 4H,
a-Cp-CH), 4.12 (dd, 4H, acetal CH2, J = 4.5, 11 Hz),
Z
h range (°)
Reflections collected/unique
3.11–25.03
18,918/4035
0.1383
2947
4035/1/276
1.079
2.54–24.20
12,503/5935
0.0228
5054
5935/0/397
1.060
4.07 (brs 4H, b-Cp-CH), 3.84 (t, 4H, acetal CH2, J = 11 Hz), 2.05
(m, 2H, acetal CH2), 1.40 (d, 2H, acetal CH2, J = 13.5 Hz); 13C NMR
100.46 (2 ꢁ acetal tertiary CH), 86.55 (2 ꢁ ipso Cp-C), 68.93
Rint
Observed reflections (F2 > 2
Data/restraints/parameters
Goodness of fit (F2)
r
)
(4 ꢁ -Cp-CH), 67.37 (4 ꢁ b-Cp-CH), 67.23 (4 ꢁ acetal O-CH2),
a
25.84 (2 ꢁ acetal CH2); I.R. (KBr) 3049, 2959, 2852, 1476, 1432,
1367, 1241, 1148, 1114, 1015, 989, 824, 696 cmꢂ1; MS (LSIMS):
M+ calculated = 357, M+ found = 358, MS (ES +) m/e Accurate Mass
R, (obs) Rw (all data)
0.0720, 0.1763
0.46, ꢂ0.48
0.0558, 0.1580
1.38, ꢂ0.45
Largest diff. peak and hole (e Åꢂ3
)
Table 4
Selected bond lengths (Å) and angles (°) for (7) and (8).
7
8
Molecule (a)
Molecule (b)
C(1)–Si(1)
Si(1)–O(1)
Si(1)–C(11)
Si(1)–C(12)
C(2)–C(15)
C(6)–Si(2)
Si(2)–O(1)
Si(2)–C(13)
Si(2)–C(14)
C(7)–C(19)
1.847 (10)
1.628 (13)
1.875 (10)
1.873 (11)
1.493 (11)
1.886 (10)
1.640 (13)
1.844 (11)
1.833 (11)
1.488 (11)
C(2)–Si(1)
Si(1)–O(3)
Si(1)–C(13)
Si(1)–C(14)
C(1)–C(6)
C(6)–O(1)
C(8)–Si(2)
Si(2)–O(3)
Si(2)–C(15)
Si(2)–C(16)
C(7)–C(12)
C(12)–O(2)
C(2)–Si(1)–O(3)
C(2)–Si(1)–C(13)
C(2)–Si(1)–C(14)
C(13)–Si(1)–C(14)
C(1)–C(6)–O(1)
C(8)–Si(2)–O(3)
C(8)–Si(2)–C(15)
C(8)–Si(2)–C(16)
C(15)–Si(2)–C(16)
C(7)–C(12)–O(2)
Si(1)–O(3)–Si(2)
1.873 (4)
1.637 (3)
1.856 (5)
1.854 (4)
1.456 (6)
1.218 (6)
1.877 (4)
1.645 (3)
1.854 (5)
1.859 (5)
1.450 (6)
1.217 (6)
110.6 (2)
106.1 (2)
113.3 (2)
108.9 (2)
124.7 (4)
110.1 (2)
112.1 (2)
108.0 (2)
109.4 (2)
125.2 (4)
141.9 (2)
C(18)–Si(3)
Si(3)–O(6)
Si(3)–C(29)
Si(3)–C(30)
C(17)–C(22)
C(22)–O(4)
C(24)–Si(4)
Si(4)–O(6)
Si(4)–C(31)
Si(4)–C(32)
C(23)–C(28)
C(28)–O(5)
C(18)–Si(3)–O(6)
C(18)–Si(3)–C(29)
C(18)–Si(3)–C(30)
C(29)–Si(3)–C(30)
C(17)–C(22)–O(4)
C(24)–Si(4)–O(6)
C(24)–Si(4)–C(31)
C(24)–Si(4)–C(32)
C(31)–Si(4)–C(32)
C(23)–C(28)–O(5)
Si(3)–O(6)–Si(4)
1.870(4)
1.635 (3)
1.853 (4)
1.863 (5)
1.464 (7)
1.212 (6)
1.869 (4)
1.643 (3)
1.856 (5)
1.858 (5)
1.446 (7)
1.224 (6)
110.9 (2)
112.8 (2)
106.0 (2)
108.4 (2)
125.1 (5)
110.7 (2)
106.9 (2)
111.1 (2)
110.1 (3)
127.1 (4)
141.5 (2)
C(1)–Si(1)–O(1)
C(1)–Si(1)–C(11)
C(1)–Si(1)–C(12)
C(11)–Si(1)–C(12)
C(2)–C(15)–O(2)
C(2)–C(15)–O(3)
C(6)–Si(2)–O(1)
C(6)–Si(2)–C(13)
C(6)–Si(2)–C(14)
C(13)–Si(2)–C(14)
C(7)–C(19)–O(4)
C(7)–C(19)–O(5)
Si(1)–O(1)–Si(2)
110.4 (4)
110.4 (4)
109.1 (5)
107.8 (6)
108.7 (7)
109.9 (6)
110.2 (4)
108.5 (5)
111.3 (5)
110.9 (5)
109.4 (7)
109.4 (7)
142.6 (3)

A. Connell et al. / Journal of Organometallic Chemistry 694 (2009) 2020–2028
2027
Reference
lated = 359.0940, found = 359.0940.
compound:
Polyethylenimine
[M + H]+
Calcu-
24.05 (P-Ar); I.R. (KBr) 3054, 2818, 2794, 2774, 2744, 1682, 1478,
1432, 1361, 1249, 1163, 1089, 1069, 1040, 1026, 999, 842,
695 cmꢂ1; MS (LSIMS): M+ calculated = 610, M+ found = 610, MS
(ES +) m/e Accurate Mass Reference compound: Perfluorotributyl-
amine Calculated = 610.0908, found = 610.0912.
4.5. Preparation of 2,20-bis-(acetal)-1,10-diphenylphosphinoferrocene
(3)
31P NMR for oxidized phosphines 25.17 (O@P-Ar).
To a cooled (ꢂ78 °C) solution of 1,10-bisacetalferrocene (5.00 g,
14 mmol) in dry ether (250 ml) was added dropwise tert-butyllith-
ium (2.2 eq, 1.7 M in pentane, 18.1 ml, 31 mmol). The mixture was
stirred at ꢂ78 °C for 15 min and the cooling bath was removed and
stirring continued at room temperature for 2 h. The reaction was
cooled to ꢂ78 °C and chlorodiphenylphosphine (2.2 eq, 6.84 g,
31 mmol) was added. The mixture was stirred at ꢂ78 °C for
15 min followed by 1.5 h stirring at room temperature. The reac-
tion was treated with water (60 ml) and CH2Cl2 (50 ml) was added.
The phases were separated and the aqueous phase was extracted
CH2Cl2 (3 ꢁ 30 ml). The combined organic phases were dried
(MgSO4) and the solvents removed in vacuo. The crude product
was dissolved in dry ether and cooled to ꢂ20 °C. This resulted in
crystallisation of 2,20-bis-(acetal)-1,10-diphenylphosphinoferro-
cene as a dark yellow solid. Yield (6.3 g, 62%); 1H NMR 7.42–7.39
(m, 5H, Ar), 7.32–7.28 (m, 3H, Ar), 7.26–7.23 (m, 3H, Ar), 7.20–
7.19 (m, 5H, Ar), 7.15–7.12 (m, 4H, Ar), 5.59 (d, 2H, acetal tertiary
4.7. Preparation of the nickel(II) chloride complex of 1,10-bis-(acetal)-
2,20- diphenylphosphinoferrocene (5)
To a solution of NiCl2(DME)2 (0.094 g, 3.3 mmol) in anhydrous
CH2Cl2 (ca. 2 ml) was added 1,10-bis-(acetal)-2,20-diphenylphos-
phinoferrocene (0.45 eq, 0.110 g, 1.52 mmol). The solution was lay-
ered with anhydrous ether and stored in the dark under nitrogen.
After ca. 72 h, dark green crystals of the nickel chloride complex
were filtered off from a dark green opaque solution. The crystals
were washed with petrol and ether and stored in the dark under
nitrogen. Yield (0.55 g, 42%), MS (EI +): M+ calculated = 854.7, M+
found = 726.1, MS (EI +) m/e Accurate Mass Reference compound:
Perfluorotributylamine, Calculated = 854.7, found = 726.1749.
Note: mass spectrometry of (5) showed only a clean molecular
ion of (m/z = 726.1749) which corresponds to the uncomplexed
diphosphinoferrocene (3) in EI and FAB modes implying
a
CH, J = 1.0 Hz), 4.73 (bs, 2H,
a
-Cp-CH), 4.22 (bs, 2H, acetal CH),
weakly-bound complex for (5). Note: NMR is not reported due to
the lack of useful data caused by paramagnetism due to an un-
paired electron in the nickel atom. The elemental analysis shows
a large difference between calculated and experimental carbon,
hydrogen and nitrogen values and has therefore not been reported
here. This could be due to decomposition of the sample to a phos-
phine oxide and/or weakly complexed metal ions detaching from
the ligand over time.
4.07 (bs, 2H, b-Cp-CH), 3.93 (bs 2 ꢁ CH acetal), 3.86 (d, 1H, acetal
CH, J = 4.7 Hz), 3.83 (d, 1H, acetal CH, J = 4.7 Hz) 3.74 (bs, 2H, acetal
CH), 3.31 (s, 2H, a0-Cp-CH), 1.97 (m, 2H, acetal CH), 1.27 (d, 2H, ace-
tal CH, J = 13.2 Hz); 13C NMR 140.22 (d, 2C, Ar-CP, JC–P = 9.2 Hz),
137.61 (d, 2C, Ar-CP, JC–P = 9.2 Hz), 135.07 (d, 4C, meta P-Ar-CH,
JC–P = 21.1 Hz), 132.32 (d, 4C, meta P-Ar-CH, JC–P = 17.4 Hz),
129.02 (2C, para P-Ar-CH), 128.03 (d, 4C, ortho P-Ar-CH, JC–P
=
7.3 Hz), 127.56 (d, 4C, ortho P-Ar-CH, JC–P = 5.5 Hz), 127.26 (2C,
para P-Ar-CH), 99.43 (d, 2C, tertiary acetal C, JC–P = 7.3 Hz), 92.82
(2C, Cp-C-P, JC–P = 22.0 Hz), 75.47 (2C, Cp-C-acetal), 72.75 (d, 2C,
4.8. Preparation of the nickel(II) bromide complex of 1,10-bis-(acetal)-
2,20-diphenylphosphinoferrocene (6)
a0-Cp-CH, JC–P = 12.8 Hz), 72.42 (2C,
a-Cp-CH), 72.31 (d, 2C, b-Cp-
CH JC–P = 3.7 Hz), 67.18 (d, 2C, acetal O-CH2, JC–P = 4.6 Hz), 66.71
(d, 2C, acetal O-CH2, JC-P = 2.8 Hz), 25.67 (2C, acetal CH2); 31P
NMR ꢂ21.29 (P-Ar); I.R. (KBr) 3092, 2962, 2922, 2851, 1382,
To a solution of NiBr2(DME)2 (0.122 g, 3.3 mmol) in anhydrous
CH2Cl2 (ca. 2 ml) was added 1,10-bisacetal-2,20-diphenylphosphi-
noferrocene (0.45 eq, 0.110 g, 1.52 mmol). The solution was lay-
ered with anhydrous ether and stored in the dark under
nitrogen. After ca. 72 h, dark green crystals of the nickel bromide
complex were filtered off from a dark green opaque solution. The
crystals were washed with petrol and ether and stored in the dark
under nitrogen. Yield (0.56 g, 39%), MS (LSIMS): M+ calcu-
lated = 944, M+ found = 943, MS (ES +) m/e Accurate Mass Refer-
ence compound: Perfluorotributylamine Calculated = 941.9466,
found = 941.9462. Note: NMR and elemental analysis data have
not been reported for compound (6) due to the same reasons as
those given above for (5).
;
1283, 1239, 1104, 1001, 812 cmꢂ1 MS (LSIMS): M+ calcu-
lated = 726, M+ found = 726.1, MS (ES +) m/e Accurate Mass Refer-
ence compound: Perfluorotributylamine; Calculated = 726.1746,
found = 726.1752.
31P NMR for oxidized phosphines 28.17 (O@P-Ar).
4.6. Preparation of 1,10-bis-carboxaldehyde-2,20-
diphenylphosphinoferrocene (4)
To a solution of 1,10-bis-(acetal)-2,20-diphenylphosphinoferro-
cene (1.00 g, 1.4 mmol) in (THF) (20 ml) and water (ca. 1 ml) was
added p-toluenesulfonic acid (1 eq, 0.260 g, 1.4 mmol). The mix-
ture was stirred with the exclusion of light for 3 h. After 3 h, the or-
ganic phase was extracted using CH2Cl2 (30 ml) and the combined
organic phases were dried over magnesium sulfate followed by re-
moval of the solvent in vacuo. Column chromatography of the res-
idue with ether (to remove 1,3-propanediol and other impurities)
followed by CH2Cl2 gave 1,10-bis-carboxaldehyde-2,20-diphenyl-
phosphinoferrocene in the second fraction as a red solid. Yield
(0.720 g, 84%), 1H NMR 10.05 (d, 2 ꢁ CHO, JP–H = 2.2 Hz), 7.31–
7.28 (m, 6H, Ar), 7.21–7.15 (m, 8H, Ar), 7.04–7.01 (pseudo t, 6H,
4.9. Preparation of 1,10-siloxane-2,50-bisacetalferrocenophane (7)
To a cooled (ꢂ78 °C) solution of 1,10-bisacetalferrocene (1.00 g,
2.8 mmol) in dry ether (75 ml) tert-butyllithium (2.2 eq, 1.7 M in
pentane, 3.6 ml, 6.2 mmol) was added. The mixture was stirred
at ꢂ78 °C for 15 min; the cooling bath was removed and stirring
continued at room temperature for 2 h. The reaction was cooled
to ꢂ78 °C and N,N0-dimethylaminodimethylchlorosilane (1.04 g,
8.4 mmol) was added; the mixture was stirred at ꢂ78 °C for
15 min followed by 1.5 h stirring at room temperature. The reac-
tion was quenched with water (60 ml) and diethyl ether (50 ml)
was added. The phases were separated and the aqueous phase
was extracted with CH2Cl2 (3 ꢁ 30 ml). The combined organic
phases were dried (MgSO4) and the solvents removed in vacuo.
1,10-Siloxane-2,50-bisacetalferrocenophane (7) was obtained as or-
ange crystals after recrystallisation from n-hexane. Yield (0.83 g,
Ar) 5.15 (bs, 2H,
a(CHO)-Cp-CH), 4.65 (pseudo t, 2H, b-Cp-CH,
J = 4.7 Hz), 3.63 (bs, 2H, a0(PPh2)-Cp-CH); 13C NMR 192.49 (d, 2C,
CHO, JC–P = 9.2 Hz), 135.54 (d, 4C ꢁ Ar-CP, JC–P = 11.0 Hz), 134.66
(d, 4C, meta P-Ar-CH, JC–P = 21.1 Hz), 132.15 (d, 4C, meta P-Ar-CH,
JC–P = 19.3 Hz), 129.75 (4C, para P-Ar-CH), 128.48 (m, 8C, ortho P-
Ar-CH), 84.68 (2C, Cp-C-P-Ar), 84.58 (2C, Cp-C-CHO), 82.54 (2C,
a
-Cp-CH), 77.69 (2C, b-Cp-CH), 75.20 (2C, a0-Cp-CH); 31P NMR -
61%) 1H NMR 5.28 (s, 2H, acetal tertiary CH), 4.54 (ps, 2H,
a-Cp-

2028
A. Connell et al. / Journal of Organometallic Chemistry 694 (2009) 2020–2028
CH), 4.41 (ps, 2H, a0-Cp-CH), 4.25 (m, 2H, acetal CH), 4.23 (ps, 2H,
b-Cp-CH) 4.17 (bs, 2H, acetal CH), 3.90 (m, 4H, acetal CH), 2.18 (m,
2H, acetal CH), 1.40 (d, 2H, acetal CH, J = 13.6 Hz), 0.37 (s, 6H,
methyl CH), 0.32 (s, 6H, methyl CH); 13C NMR 99.66 (2C, acetal ter-
[3] S.L. Ingham, K.S. Muhammad, J. Lewis, N.J. Long, P.R. Raithby, J. Organomet.
Chem. 470 (1994) 153–159.
[4] I.R. Butler, A.L. Boyes, G. Kelly, S.C. Quayle, T. Herzig, J. Szewczyk, Inorg. Chem.
Commun. 2 (1999) 403–406.
[5] Y. Zhu, M.O. Wolf, J. Am. Chem. Soc. 122 (2000) 10121–10125.
[6] K.R.J. Thomas, J.T. Lin, Y.S. Wen, Organometallics 19 (2000) 1008–1012.
[7] L.J. Hobson, I.R. Butler, Synth. Met. 71 (1995) 2199–2200.
[8] W.Y. Wong, G.L. Lu, K.F. Ng, C.K. Wong, K.H. Choi, J. Organomet. Chem. 637–639
(2001) 159–166.
[9] E.M. McGale, B.H. Robinson, J. Simpson, Organometallics 22 (2003) 931–939.
[10] I.R. Butler, A.G. Callabero, G.A. Kelly, J.R. Amey, T. Kraemer, D.A. Thomas, M.E.
Light, T. Gelbrich, S.J. Coles, Tetrahedron Lett. 45 (2004) 467–472.
[11] I.R. Butler, A.G. Caballero, G.A. Kelly, Inorg. Chem. Comm. 6 (2003) 639–642.
[12] T.Y. Dong, S.W. Chang, S.F. Lin, M.C. Lin, Y.S.L.L. Wen, Organometallics 25
(2006) 2018–2024.
tiary CH), 89.87 (2C, ipso-Cp-C-acetal), 77.91 (2C,
a-Cp-CH), 70.27
(2C, b-Cp-CH), 69.40 (2C, a0-Cp-CH), 68.80 (2C, ipso Cp-C-Siloxane),
66.20 (4C, acetal O-CH2), 39.77 (2C, acetal O-CH2), 25.84 (2C, acetal
CH2), 0.35 (2C, Si-CH3), 0.00 (2C, Si-CH3); IR (KBr) 3081, 2958, 2925,
2843, 2721, 1469, 1455, 1393, 1378, 1364, 1273, 1251, 1146, 1106,
1078, 1019, 995, 826, 692 cmꢂ1; MS (LSIMS): M+ calculated = 488,
M+ found = 488.1, MS (ES +) m/e Accurate Mass Reference com-
pound:
Polyethylenimine
[M + H]+
Calculated = 487.1257,
[13] C. Engratkul, L.R. Sita, Nano. Letter 10 (2001) 541–549.
[14] N.J. Long, A.J. Martin, R. Vilar, A.J.P. White, D.J. Williams, M. Younus,
Organometallics 18 (1999) 4261–4269.
found = 487.1258.
[15] P.F. Brandt, T.B. Rauchfuss, J. Am. Chem. Soc. 114 (1992) 1926–1927.
[16] G. Iftime, J.-C. Daran, E. Manoury, G.G.A. Balavoine, J. Organomet. Chem. 565
(1998) 115–124.
4.10. Preparation of 1,10-siloxane-2,50-ferrocenophanecarboxaldehyde
(8)
[17] T. Hayashi, A. Yamamoto, M. Hojo, Y. Ito, J. Chem. Soc., Chem. Commun. (1989)
495–496.
To
a
solution of 1,10-siloxane-2,50-bisacetalferrocenophane
[18] W. Zhang, Y. Adachi, T. Hirao, I. Ikeda, Tetrahedron: Asymm.
451–460.
7 (1996)
(0.06 g, 0.123 mmol) in THF (20 ml) and water (ca.1 ml) was added
p-toluenesulfonic acid (1 eq, 0.07 g, 0.37 mmol). The mixture was
stirred with the exclusion of light for 3 h. After 3 h, the organic
phase was extracted using CH2Cl2 (30 ml), and the combined or-
ganic phases were dried over magnesium sulfate followed by re-
moval of the solvent in vacuo. 1,10-Siloxane-2,50-ferroceno
phanecarboxaldehyde was recrystallised from diethyl ether lay-
ered with hexane to give red crystals. Yield (0.035 g, 80%), 1H
[19] G. Iftime, J.-C. Daran, E. Manoury, G.G.A. Balavoine, Angew. Chem., Int. Ed. 37
(1998) 1698–1701.
[20] G.G.A. Balavoine, J.-C. Daran, G. Iftime, E. Manoury, C. Moreau-Bossuet, J.
Organomet. Chem. 567 (1998) 191–198.
[21] W. Zhang, T. Hirao, I. Ikeda, Tetrahedron Lett. 37 (1996) 4545–4548.
[22] W. Steffen, M. Laskoski, G. Collins, U.H.F. Bunz, J. Organomet. Chem. 630 (2001)
132–138.
[23] W. Steffen, M. Laskoski, J.G.M. Morton, U.H.F. Bunz, J. Organomet. Chem. 689
(2004) 4345–4356.
[24] U.H.F. Bunz, J. Organomet. Chem. 494 (1995) C8–C11.
[25] P.V. Roling, M.D. Rausch, J. Org. Chem. 39 (1974) 1420–1424.
[26] A.N. Nesmeyanov, V.A. Sazonova, V.N. Drozd, Akad. Nauk. S.S.S.R. 126 (1959)
1004–1010.
[27] I.R. Butler, W.R. Cullen, J. Ni, S.J. Rettig, Organometallics 4 (1985) 2196–2201.
[28] G.G.A. Balavoine, G. Doisneau, T. Fillebeen-Khan, J. Organomet. Chem. 412
(1991) 381–382.
NMR 9.88 (s, 2H, CHO), 4.73 (bs, 2H,
a -
-Cp-CH), 4.68 (brs 2H, a0
Cp-CH,), 4.44 (brs 2H, b-Cp-CH), 0.26 (s, 6H, methyl CH), 0.17 (s,
6H, methyl CH); 13C NMR 192.86 (s, 2C, CHO), 83.50 (2C, Cp-C-
CHO), 82.67 (2C, a
-Cp-CH), 75.44 (2C, b-Cp-CH), 74.89 (2C, a0-Cp-
CH), 73.71 (2C, Cp-C-Siloxane) 0.74 (2C, Si-CH3), 0.00 (2C, Si-
CH3); MS (LSIMS): M+ calculated = 356, M+ found = 357.
[29] U.T. Mueller-Westerhoff, Z. Yang, G. Ingram, J. Organomet. Chem. 463 (1993)
163–167.
[30] C.T. Hartinger, A.A. Nazarov, V. Chevchenko, V.B. Arion, M. Galanski, B.K.
Keppler, Dalton Trans. (2003) 3098–3102.
Acknowledgements
[31] J.-G. Rodríguez, S. Pleite, J. Organomet. Chem. 637–639 (2001) 230–239.
[32] W. Zhang, F. Xie, H. Yoshinaga, T. Kida, Y. Nakatsuji, I. Ikeda, Tetrahedron 62
(2006) 9038–9042.
[33] X.-H. Liu, D.W. Bruce, I. Manners, J. Organomet. Chem. 548 (1997) 49–56.
[34] P. Gómez-Elipe, R. Resendes, P.M. Macdonald, I. Manners, J. Am. Chem. Soc. 120
(1998) 8348–8356.
[35] R. Rulkens, A.J. Lough, I. Manners, J. Am. Chem. Soc. 116 (1994) 797–798.
[36] C. Angelakos, D.B. Zamble, D.A. Foucher, A.J. Lough, I. Manners, Inorg. Chem. 33
(1994) 1709–1718.
The authors sincerely thank the EPSRC, Corus Colors Ltd and the
EPSRC EngD Scheme in Steel Technology at Swansea University for
financial support for AC, the EPSRC National Mass Spectrometry
Service at Swansea for MS data, the EPSRC for funding of the Na-
tional Crystallography Service, and STFC for access to synchrotron
facilities.
[37] R.L. Schaaf, P.T. Kan, C.T. Lenk, J. Org. Chem. 26 (1961) 1790–1795.
[38] V.I. Buziashvili, L.M. Khananaahvili, B.K. Kabanov, T.N. Nikol’skaya, O.V.
Kuz’man, Zh. Obshch. Khim. 58 (1988) 88.
Appendix A. Supplementary material
[39] C.H. Hartinger, T.H. Brehmer, G. Giester, M. Galanski, A.A. Nazarov, S.M. Luther,
B.K. Keppler, Inorg. Chim. Acta 328 (2002) 237–240.
[40] APEX2, Bruker AXS Inc., Madison, WI, USA, 2005.
[41] SAINT, Bruker AXS Inc., Madison, WI, USA, 1995.
[42] G.M. Sheldrick, SADABS, Bruker AXS Inc.,, Madison, WI, USA, 2007.
[43] G.M. Sheldrick, Acta Crystallogr. Sect. A 64 (2008) 112–122.
[44] R.W.W. Hooft, COLLECTNonius BV, Delft, The Netherlands, 1998.
[45] Z. Otwinowski, W. Minor, in: C.W. Carter, R.M. Sweet (Eds.), Methods in
Enzymology, vol. 276, Academic Press, New York, 1997, pp. 307–326.
[46] M.C. Burla, R. Caliandro, M. Camalli, B. Carrozzini, G.L. Cascarano, L. De Caro, C.
Giacovazzo, G. Polidori, R. Spagna, J. Appl. Crystallogr. 38 (2005) 381–
388.
CCDC 695609–695615 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
http://www.ccdc.cam.ac.uk/data_request/cif.
References
[1] J.K. Pudelski, M.R. Callstrom, Organometallics 11 (1992) 2757–2759.
[2] R.W. Heo, T.R. Lee, J. Organomet. Chem. 578 (1999) 31–42.