Synthesis, characterisation and catalytic activity of metal complexes of neutral, unsymmetrical P/S ferrocenediyl ligands

Article abstract of DOI:10.1039/b205653d

Two asymmetric 1,1′-disubstituted ferrocenediyl ligands, 1-(diphenylphosphino)-1′-(methylthio)ferrocene (1) and 1-(diphenylphosphino)-1′-(mesitylthio)ferrocene (2) featuring phosphine and thioether substituents have been conveniently synthesised and the coordination chemistry of 1 probed by reaction with transition metal reagents. With Group 10 metal and Rh(I) species, chelating complexes are formed in high yield and a monodentate bis ligand complex with trans phosphorus ligation can also be synthesised using a Pd(II) species and three equivalents of 1. With [M(CO)₅thf] (M = Cr, Mo or W), 1 forms a mixture of monodentate, P-bound pentacarbonyl and P/S-chelating tetracarbonyl products. The monodentate pentacarbonyl product can be converted into the chelating tetracarbonyl species via prolonged reflux in toluene. Preliminary studies show that 1, in combination with Pd(0) precursors, can act as a catalyst for the Suzuki coupling reaction.

Full text of DOI:10.1039/b205653d

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis, characterisation and catalytic activity of metal
complexes of neutral, unsymmetrical P/S ferrocenediyl ligands
Vernon C. Gibson,a Nicholas J. Long,*a Andrew J. P. White,a Charlotte K. Williams,a
David J. Williams,a Marco Fontanib and Piero Zanellob
a Department of Chemistry, Imperial College of Science, Technology and Medicine,
South Kensington, London, UK SW7 2AY. E-mail: n.long@ic.ac.uk
b Dipartimento di Chimica dell’Università di Siena, Via Aldo Moro, 53100 Siena, Italy
Received 10th June 2002, Accepted 12th July 2002
First published as an Advance Article on the web 5th August 2002
Two asymmetric 1,1Ј-disubstituted ferrocenediyl ligands, 1-(diphenylphosphino)-1Ј-(methylthio)ferrocene (1) and
1-(diphenylphosphino)-1Ј-(mesitylthio)ferrocene (2) featuring phosphine and thioether substituents have been
conveniently synthesised and the coordination chemistry of 1 probed by reaction with transition metal reagents.
With Group 10 metal and Rh() species, chelating complexes are formed in high yield and a monodentate bis ligand
complex with trans phosphorus ligation can also be synthesised using a Pd() species and three equivalents of 1.
With [M(CO)5thf] (M = Cr, Mo or W), 1 forms a mixture of monodentate, P-bound pentacarbonyl and P/S-chelating
tetracarbonyl products. The monodentate pentacarbonyl product can be converted into the chelating tetracarbonyl
species via prolonged reflux in toluene. Preliminary studies show that 1, in combination with Pd() precursors, can act
as a catalyst for the Suzuki coupling reaction.
Introduction
The formation of asymmetric, multidentate ligands, focussing
on aspects of hemilability,1 has been of great interest. In par-
ticular, ferrocenyl and ferrocenediyl ligands are sought after
due to their extensive coordination chemistry and applications
within catalysis.2,3 Whilst hemilabile P/SϪ,4 P/S,5 P/O6,7 and
N/OϪ 8 ligand systems are well-known and important in
catalysis, examples of analogous asymmetric 1,1Ј-disubstituted
ferrocenes9,10 are rare. The first 1-(PR2), 1Ј-(SR)ferrocenediyl
species (R = Ph) was published in 199211 followed by an
alternative five-step synthesis of the same ligand reported by
Dong and co-workers in 1998.12 Butler et al.13 have formed
[{η5-C5H4SMe}Fe{η5-C5H3(PPh2)(SMe)-1,2}] via the ortho-
lithiation of dibromoferrocene and recently, a chiral phos-
phine–thioether ferrocene ligand has been utilised to facilitate
the copper-catalysed asymmetric addition of diethylzinc to
alkylidene malonates.14 Our group first reported the mixed
phosphine–thioether ligand 1 in 1999,15 but herein the synthesis
has been improved and the synthetic methodology extended to
include the bulky mesityl substituent (ligand 2). To explore the
Scheme 1 The syntheses of compounds 1–6.
formation of new ferrocenophanes (species that feature linking
of the cyclopentadienyl rings by the introduction of a hetero-
annular bridge or bridges) and to probe the hemilability of
asymmetric, 1,1Ј-disubstituted ligands, 1 has been treated with
a range of transition metal reagents and the coordination
chemistry, spectroscopic, crystallographic and electrochemical
characterisation and preliminary catalytic studies are reported.
improved yields (ca. 40 %) via the reaction of 1,1Ј-dilithio-
ferrocene with a 1 : 1 mixture of dimethyl disulfide and
chlorodiphenylphosphine at room temperature in hexane. The
reaction gives both the symmetrically and asymmetrically disub-
stituted products, 1,1Ј-bis(methylthio)ferrocene, 1-(diphenyl-
phosphino)-1Ј-(methylthio)ferrocene (1) and 1,1Ј-bis(diphenyl-
phosphino)ferrocene in a 1 : 2 : 1 ratio. 1 can be separated from
the other products by careful column chromatography (neutral
Results and discussion
grade II alumina, diethyl ether–hexane, 1 : 4) in 40% yield com-
pared to 21% obtained previously.15 The difference in yields is
attributed to the preparation and purification of the 1,1Ј-dilithio-
ferrocene precursor; distilled TMEDA and oxygen-free hexane
were used to maximise the yield of 1,1Ј-dilithioferrocene
and minimise the yield of lithioferrocene and the former was
also washed thoroughly with oxygen-free hexane, which
enabled separation of any unreacted materials before further
reaction.
Synthesis and characterisation of the ligands 1 and 2
The ligands 1-(diphenylphosphino)-1Ј-(methylthio)ferrocene
(1) and 1-(diphenylphosphino)-1Ј-(mesitylthio)ferrocene (2)
were synthesised using a one step route from ferrocene (Scheme
1). Initial work on the formation of 1 and preliminary coordin-
ation chemistry with CuI and AgI metal centres has been pub-
lished previously.15 However, 1 can now be synthesised in
3280
J. Chem. Soc., Dalton Trans., 2002, 3280–3289
This journal is © The Royal Society of Chemistry 2002
DOI: 10.1039/b205653d

View Article Online
Ligand 2 was synthesised in an analogous manner to 1 i.e. a
mixture of dimesityl disulfide and chlorodiphenylphosphine
were added to 1,1Ј-dilithioferrocene and the resulting suspen-
sion stirred for 15 h. After an aqueous work up, column chrom-
atography (neutral grade II alumina, CH2Cl2–hexane, 3 : 7)
enabled isolation of the pure ligand in 10% yield. Higher
proportions of monosubstituted products are isolated than
during the synthesis of 1, which could be due to the increased
steric hindrance of the mesityl groups. The ligands were charac-
terised by microanalysis, 1H NMR, 31P{1H} NMR (a singlet is
observed at ca. Ϫ16.5 ppm for both ligands) and FAB mass
spectroscopy (with molecular ions and the expected fragmen-
tation patterns observed).
Ni–P bond is ca. 0.02 Å [0.03 Å] shorter than that to sulfur
(Table 1). Distortions from tetrahedral geometry at phosphorus
are quite large with angles ranging from 100.7(3) to 118.9(2)Њ
[105.2(2) to 115.0(2)Њ] whilst those at sulfur range between
101.5(4) and 109.2(2)Њ [102.3(4) and 111.9(2)Њ]. The cyclopenta-
dienyl rings are essentially eclipsed (less than 1Њ [4Њ] stagger) the
rings being inclined by ca. 3Њ [6Њ]—the phosphorus and sulfur
atoms effectively overlay each other. The pseudo six-membered
chelate ring formed by Ni/P/C(10)/Fe/C(5)/S has a steeply
folded geometry with an out-of-plane bend of ca. 56Њ about the
Pؒ ؒ ؒS vector. The non-bonded Sؒ ؒ ؒP and Niؒ ؒ ؒFe distances are
3.46 Å [3.52 Å] and 4.00 Å [4.04 Å] respectively. There are no
intermolecular packing interactions of note. The most closely
related literature structure11 is that of 1-diphenylphosphino-1Ј-
phenylthioferrocenetricarbonyliron() which has non-bonded
Sؒ ؒ ؒP and Feؒ ؒ ؒFe separations of 3.16 and 4.12 Å respectively,
the ferrocenyl ring systems having a slightly increased degree of
stagger at ca. 13Њ.
Synthesis and characterisation of Group 10 metal complexes of 1
As a guide to the coordinating properties of the ligands, Group
10 metal precursors were reacted with 1 (the small amount of 2
obtained precluded coordination studies).
Dichloro[1-(diphenylphosphino)-1Ј-(methylthio)ferrocene]-
platinum() (4) was formed by the dissolution of 1 in toluene
followed by the addition of dichloro(1,5-cyclooctadiene)-
platinum() dissolved in CH2Cl2. The solution was stirred for
15 h at room temperature and the crude product was filtered off
and washed with hexane to remove any unreacted starting
materials. 4 was isolated as a yellow powder in 91% yield. In the
NMR spectra, the peaks are all shifted to lower field compared
to the free ligand due to the usual inductive effect of the
The Ni() complex (3) was synthesised in good yield (Scheme
1) via the reaction of (1,2-dimethoxyethane)dibromonickel()
and 1 in CH2Cl2. There was an immediate colour change from
pale pink to dark green, and after 40 h the green solution was
separated from the unreacted starting material and decom-
position products and evaporated to dryness to leave a green
powder. This was washed with heptane to remove any un-
reacted ligand, and recrystallisation by layering a CH2Cl2 solu-
tion of the complex with pentane yielded the air-sensitive, green
needle-like 3 in 75% yield. 3 did not display any distinct signals
1
platinum centre. In the H NMR spectrum, the methyl group
exhibits a singlet at 2.86 ppm with platinum satellites, the
ferrocenyl protons give only two broad peaks shifted to lower
field at 4.39 and 4.83 ppm and the phenyl ring protons display a
multiplet at 7.43 ppm, shifted slightly compared to the free
ligand. In the 31P{1H}NMR spectrum, a peak at 8.64 ppm is
observed with platinum satellites (1JPt–P = 3737 Hz), which is in
close agreement with the analogous dppf–PtCl2 species.16 The
mass spectrum shows a molecular ion peak (682 amu) and
fragmentation signals due to loss of chloride groups (647, 611
amu). The X-ray analysis of crystals of 4 shows (Fig. 2) it to
1
in H NMR and showed a very broad peak around 42 ppm in
the 31P{1H} NMR spectrum indicating a paramagnetic nature
and tetrahedral geometry. The positive ion FAB mass spectrum
shows the molecular ion (635 amu) and fragmentation peaks
due to loss of methylthio (588 amu), bromide (555 and 475
amu) and diphenylphosphino (450 amu) groups. The X-ray
analysis of 3 shows the complex (Fig. 1) to crystallise with two
Fig. 1 Molecular structure of 3.
independent molecules in the asymmetric unit for which the
only major difference is in the relative orientations of the
phenyl substituents. The geometry at nickel is distorted tetra-
hedral with angles in the range 96.77(6)–125.03(5)Њ [99.69(7)–
128.14(5)Њ], the most obtuse angle being associated with the two
bromide ligands (the values in square brackets refer to the
second independent molecule). A noticeable departure of the
nickel centre from tetrahedral geometry is a small square
planar-type distortion with the NiBr2 and NiSP planes being
twisted by ca. 11Њ [6Њ] from an orthogonal relationship. The
Fig. 2 Molecular structure of 4.
have a ligand structure very similar to that of 3 with an eclipsed
geometry for the ferrocenyl unit (less than 1Њ stagger, rings
inclined by ca. 2Њ) and non-bonded Sؒ ؒ ؒP and Ptؒ ؒ ؒFe dis-
tances of 3.15 and 3.88 Å respectively, the noticeable reduced
Sؒ ؒ ؒP distance cf. that in 3 reflecting an inclining of the P–C5H4
and S–C5H4 vectors towards the platinum centre in 4 cf. away
from the nickel atom in 3. The departures from tetrahedral
J. Chem. Soc., Dalton Trans., 2002, 3280–3289
3281

View Article Online
Table 1 Selected bond lengths (Å) and angles (Њ) for 3
Molecule A
Molecule B
Molecule A
Molecule B
Ni–P
Ni–Br(1)
S–C(5)
2.298(2)
2.3408(13)
1.764(7)
2.281(2)
2.3380(12)
1.750(8)
Ni–S
Ni–Br(2)
P–C(10)
2.319(2)
2.3497(13)
1.796(7)
2.310(2)
2.3298(13)
1.801(7)
P–Ni–S
97.22(7)
117.74(6)
96.77(6)
101.5(4)
104.9(3)
106.9(4)
112.2(2)
118.9(2)
100.20(7)
108.74(7)
99.69(7)
102.3(4)
102.5(3)
106.0(3)
112.9(2)
115.0(2)
P–Ni–Br(1)
P–Ni–Br(2)
Br(1)–Ni–Br(2)
C(5)–S–Ni
C(10)–P–C(16)
C(16)–P–C(22)
C(16)–P–Ni
104.30(6)
112.66(6)
125.03(5)
109.2(2)
105.4(3)
100.7(3)
111.5(2)
105.12(6)
111.45(6)
128.14(5)
111.9(2)
105.9(3)
105.2(2)
111.2(2)
S–Ni–Br(1)
S–Ni–Br(2)
C(5)–S–C
C–S–Ni
C(10)–P–C(22)
C(10)–P–Ni
C(22)–P–Ni
Table 2 Selected bond lengths (Å) and angles (Њ) for 4
Table 3 Selected bond lengths (Å) and angles (Њ) for 6
Pt–Cl(1)
Pt–S
S–C(1)
2.300(2)
2.293(2)
1.761(6)
Pt–Cl(2)
Pt–P
P–C(6)
2.364(2)
2.241(2)
1.807(7)
Pd–Cl
S–C(10)
2.3373(13)
1.746(7)
Pd–P
P–C(5)
2.3500(14)
1.805(6)
Cl–Pd–ClЈ
ClЈ–Pd–PЈ
ClЈ–Pd–P
C(10)–S–C(11)
C(5)–P–C(23)
C(5)–P–Pd
C(23)–P–Pd
180
Cl–Pd–PЈ
Cl–Pd–P
P–Pd–PЈ
C(5)–P–C(17)
C(17)–P–C(23)
C(17)–P–Pd
92.62(5)
87.38(5)
180
105.8(2)
102.4(2)
108.88(14)
P–Pt–S
88.04(5)
177.81(6)
93.82(6)
99.8(3)
112.9(3)
104.4(3)
107.2(2)
116.3(2)
P–Pt–Cl(1)
P–Pt–Cl(2)
Cl(1)–Pt–Cl(2)
C(1)–S–Pt
C(6)–P–C(17)
C(17)–P–C(23)
C(17)–P–Pt
90.29(6)
169.64(7)
88.08(7)
105.6(2)
107.8(3)
100.6(2)
119.37(14)
87.38(5)
92.62(5)
102.4(4)
102.1(3)
115.3(2)
120.69(14)
S–Pt–Cl(1)
S–Pt–Cl(2)
C(1)–S–C(11)
C(11)–S–Pt
C(6)–P–C(23)
C(6)–P–Pt
C(23)–P–Pt
isomer formed – presumably the trans species, as shown by
X-ray diffraction. The mass spectrum does not show the
molecular ion, but rather fragmentation peaks due to loss
of chloride (974 and 938 amu), methylthio (891 amu) and
diphenylphosphino (754 amu) groups. The failure to achieve
P,S chelation is illustrated by the X-ray crystallographic
geometry at phosphorus and sulfur are again quite large with
angles ranging from 100.6(2) to 119.37(14)Њ, and from 99.8(3)
to 112.9(3)Њ respectively (Table 2). The six-membered chelate
ring fold angle (vide supra) is here 69Њ. The platinum adopts an
only slightly distorted square planar geometry with cis angles in
the range 88.04(5)–93.82(6)Њ, the most noticeable deformation
being an out-of-plane deviation of ca. 0.43 Å of the phos-
phorus atom from the PtCl2S plane. The Pt–P distance is ca.
0.05 Å shorter than that to sulfur. There are no particularly
short intermolecular contacts, though it is interesting to note
that the coordination planes of C2 related complexes have a
non-bonded Ptؒ ؒ ؒPt separation of only 4.15 Å.
determination of the centrosymmetric complex L2PdCl2
(Fig. 3). The palladium is square planar with cis angles of
6
The palladium() complex (5) can be synthesised in the same
way as the platinum() analogue 4 (Scheme 1) to give an orange
1
powder in 81% yield. In the H NMR spectrum, as with 4, all
the ligand peaks are shifted to lower field on coordination to the
metal. The 31P{1H} NMR spectrum shows a singlet at 31.82
ppm and the FAB mass spectrum does not show the molecular
ion but rather a peak due to loss of a chloride group at 558
amu.
During attempts to synthesise the chelating complex (5), a
palladium complex with two monodentate ligands bound
through their phosphorus atoms was noted (Scheme 1). This
complex (6) arose when there was an excess of ligand, leading
to the observation that in order to synthesise the chelating
complex in high yield an excess of palladium precursor should
be used. 6 can be synthesised as an orange-red powder in 80%
yield when an excess of the ligand 1 (ca. 3 : 1 ratio) in toluene is
reacted with Pd(COD)Cl2 in CH2Cl2. Clearly, the better donat-
ing properties of phosphorus compared to sulfur are utilised
here, leading to the bonding with palladium being exclusively
through phosphorus. The trans isomer is the only one observed,
presumably due to the steric bulk of the ligand precluding
formation of the cis species. The 1H NMR spectrum of 6 shows
a singlet at 2.25 ppm due to the methylthio protons but these
are not shifted compared to the free ligand, indicating that the
sulfur atoms are not bound to the metal and illustrating the
mono-dentate binding mode. The 31P{1H} NMR shows one
peak only at 15.57 ppm, which indicates that the P atom is
metal bound (there is a large downfield shift from the equiva-
lent signal in the free ligand) and that there is only a single
Fig. 3 Molecular structure of 6.
87.38(5) and 92.62(5)Њ, and Pd–P and Pd–Cl distances of
2.350(1) and 2.337(1) Å respectively; the non-bonded Pdؒ ؒ ؒFe
separation is 4.40 Å (Table 3). Departures from tetrahedral
geometry at phosphorus are comparable with those seen in 3
and 4, with angles in the range 102.1(3)–120.69(14)Њ; the angle
at sulfur is 102.4(4)Њ. The two cyclopentadienyl rings are essen-
tially eclipsed (ca. 3Њ stagger) but the P–C5H4 and S–C5H4
vectors are here rotated by ca. 74Њ with respect to each other.
There are no noteworthy intermolecular packing interactions.
Synthesis and characterisation of Rh(I) complex of 1
Two new classes of methanol carbonylation catalysts, featuring
P/S-substituted ligands and Rh() metal centre(s), which show
3282
J. Chem. Soc., Dalton Trans., 2002, 3280–3289

View Article Online
significant improvements in absolute rates, under industrial
conditions, over those obtained with [RhI2(CO)2]Ϫ have been
recently reported.17 An analogous Rh() complex (7) was
formed by dissolving two equivalents of 1 in the minimum
amount of toluene and adding this to a solution of tetra-
carbonyldichlorodirhodium() dissolved in methanol (Scheme
2). An orange precipitate formed immediately, which was
and an indication of the poorer donating ability of the
thioether group compared to the phosphine.
1
In the H NMR spectrum of 8, the chelating coordination
mode can be seen by the shifting to lower field of both the
methylthio and the phenyl group resonances. The methylthio
protons exhibit a singlet at 2.90 ppm and the ferrocenyl region
shows three peaks: a multiplet, with an integral value of four, at
4.22 ppm due to the β and βЈ protons and two pseudo triplets,
each integrating to two, at 4.40 and 4.58 ppm due to the α and
αЈ protons. The phenyl groups show two multiplets at 7.38 and
7.54 ppm. The 31P{1H} NMR spectrum also indicates that a
chelating complex has been formed as only one peak is seen at
19.53 ppm with tungsten satellites (1JW–P = 234 Hz).19 The posi-
tive ion FAB mass spectrum shows a molecular ion (712 amu)
and a fragmentation pattern due to loss of carbonyl groups
(684, 656, 628 and 600 amu). The carbonyl region of the IR
spectrum shows four peaks at 2016, 1905, 1890 and 1875 cmϪ1
,
this being the expected pattern for a cis chelate metal tetra-
1
carbonyl complex.19,20 The H NMR of 9 provides good evi-
dence for the monodentate nature of the ligand in this complex,
with the phosphorus atom bonding to tungsten and the sulfur
atom not coordinating. The signal for the methylthio protons at
2.15 ppm is hardly shifted from that of the free ligand. The
ferrocenyl region of the spectrum shows four pseudo triplets
and the phenyl groups exhibit one resonance at 7.42 ppm. In the
31P{1H} NMR spectrum, there is one peak in the spectrum at
11.15 ppm (1JW–P = 245 Hz) which is not as downfield shifted as
for 8. The positive ion FAB mass spectrum shows a peak for the
molecular ion (740 amu) and peaks due to loss of four and five
carbonyl groups respectively (628, 600 amu). The carbonyl
region of the IR spectrum shows three peaks at 2070, 1980 and
1934 cmϪ1, the expected number and position of peaks for a
pentacarbonyl complex.19
Scheme 2 The syntheses of compounds 7–13.
filtered off after 2 h and washed with diethyl ether and
methanol. Carbonylchloro[1-(diphenylphosphino)-1Ј-(methyl-
thio)ferrocene]rhodium() (7) was isolated in 79% yield. The
31P{1H} NMR spectrum shows a doublet at 37.42 ppm
due to coupling with the Rh nucleus, the coupling constant
1JRh–P being 168 Hz, which is in good agreement with other
rhodium ferrocenediyl phosphine chelating complexes.18
A
The X-ray structure of the tungsten complex 8 again reveals
(Fig. 4) a P,S ligand structure essentially unchanged from that
single metal carbonyl IR stretching frequency is observed
at 2006 cmϪ1 and the positive ion FAB mass spectrum shows
the molecular ion (583 amu) and fragmentation peaks due
to loss of carbonyl (555 amu) and chloride (547 amu)
groups.
Synthesis and characterisation of Group 6 metal complexes of 1
The reactivity of 1 with Group 6 metal carbonyl species was
explored via the addition of [M(CO)5thf] to a solution of the
ligand under inert atmospheres with the reactions being moni-
tored via IR spectroscopy. The [M(CO)5thf] starting material
(M = W, Mo, Cr) was prepared by photolysis of M(CO)6 in dry,
degassed thf for an appropriate period of time using a 400 W
UV lamp. The irradiation time varied for each of the metal
carbonyls employed (Cr, 35 min, W, 14 min, Mo, 6 min) and the
formation of the [M(CO)5thf] intermediate was monitored by
IR spectroscopy [the diagnostic peaks being ν(CO) (in thf )
M(CO)6 = ca. 1978 cmϪ1, M(CO)5(thf ) = ca. 2075, 1924, 1889
cmϪ1].
Tungsten complexes. The compounds 8 and 9 were syn-
thesised as shown in Scheme 2. An excess of pentacarbonyl-
(tetrahydrofuran)tungsten() was added to a solution of 1 in
toluene. The yellow solution was stirred at room temperature
for 20 h after which time the IR spectrum remained the same
and indicated the formation of a mixture of penta- and tetra-
carbonyl species. The crude reaction mixture was purified by
column chromatography (neutral grade II alumina, CH2Cl2–
hexane, starting with 1 : 9 and increasing the polarity to 1 : 5 to
remove 9 and 1 : 3 to remove 8) enabling isolation of tungsten
hexacarbonyl, 9 in 39% yield and 8 in 50% yield. The reason for
the isolation of both penta- and tetra-carbonyl products rather
than the expected compound 8 could be due to the relative
proportions of pentacarbonyl(tetrahydrofuran)tungsten and
tetracarbonyl(tetrahydrofuran)tungsten in the starting mixture
Fig. 4 Molecular structure of 8.
seen in 3 and 4 with eclipsed cyclopentadienyl ring systems (less
than 1Њ stagger, rings inclined by ca. 2Њ). The non-bonded Sؒ ؒ ؒP
and Wؒ ؒ ؒFe separations are 3.48 and 4.36 Å respectively, the
P–C5H4 and S–C5H4 vectors being tilted slightly away from
the tungsten centre. The fold of the six-membered chelate ring
(vide supra) is ca. 52Њ. The angles at phosphorus and sulfur
are in the ranges 102.2(2)–118.47(13)Њ and 101.0(3)–114.2(2)Њ
respectively (Table 4). The geometry at tungsten is slightly dis-
torted octahedral with cis angles in the range 85.4(3) to
97.7(2)Њ. The W–P bond length is ca. 0.02 Å shorter than that to
J. Chem. Soc., Dalton Trans., 2002, 3280–3289
3283

View Article Online
Table 4 Selected bond lengths (Å) and angles (Њ) for 8
W–S
2.555(2)
1.952(7)
2.008(7)
1.760(6)
W–P
2.5378(14)
1.986(6)
2.050(7)
1.820(6)
W–C(24)
W–C(27)
S–C(1)
W–C(25)
W–C(26)
P–C(6)
C(24)–W–C(25)
C(25)–W–C(27)
C(25)–W–C(26)
C(24)–W–P
C(27)–W–P
C(24)–W–S
C(27)–W–S
P–W–S
88.1(3)
89.4(3)
90.6(3)
92.8(2)
89.2(2)
176.6(2)
97.7(2)
86.17(5)
114.2(2)
102.2(2)
103.3(2)
118.47(13)
C(24)–W–C(27)
C(24)–W–C(26)
C(27)–W–C(26)
C(25)–W–P
C(26)–W–P
C(25)–W–S
C(26)–W–S
C(1)–S–C(11)
C(11)–S–W
C(6)–P–C(23)
C(6)–P–W
C(23)–P–W
85.4(3)
88.0(3)
173.4(3)
178.3(2)
90.9(2)
93.0(2)
88.8(2)
101.0(3)
111.5(3)
103.3(2)
116.1(2)
111.5(2)
Fig. 5 Cyclic voltammetric responses recorded at a platinum electrode
on CH2Cl2 solutions containing [NBu4]PF6 (0.2 mol dmϪ3) and: (a) 1
C(1)–S–W
C(6)–P–C(17)
C(17)–P–C(23)
C(17)–P–W
(1.6 × 10Ϫ3 mol dmϪ3); (b) 4 (1.6 × 10Ϫ3 mol dmϪ3). Scan rate 0.2 V sϪ1
.
also affords an irreversible reduction, which is confidently
attributed to the Pt()/Pt() process. As far as 3 is concerned,
the presence of the redox-active Ni() fragment complicates the
redox path, in that a first irreversible oxidation (Ep = ϩ0.89 V)
almost overlaps a partially chemically reversible second oxid-
ation (EЊЈ = ϩ1.00 V). For simplicity, we assign such a process
to the Ni()/Ni() oxidation, the irreversibility of which
destabilises the subsequent ferrocene-centred oxidation. It can-
not however, be ruled out that the process may arise from the
oxidation of the bromide ligand.21 An irreversible reduction is
also detected (Ep = Ϫ0.51 V).
A more subtle redox pattern is exhibited by the Pd()
analogue 5, in that it also exhibits a one-electron oxidation that
appears to be chemically reversible at very low scan rates
(ipc/ipa ratio close to unity at 0.02 V sϪ1), but the increase of the
scan rate results in a decrease of the current ratio. Such
behaviour is diagnostic of a first-order reversible chemical reac-
tion following a reversible electron transfer, in which the chem-
ical complication is neither too slow nor too fast.22 In fact,
exhaustive electrolysis once again confirms the instability of
[5]ϩ. On this basis, we assume that the oxidation of 4 might also
be complicated by a first-order reversible chemical reaction, but
that the rate of the chemical complication is so fast that the
cyclic voltammogram apparently looks like a non-complicated
reversible electron transfer.22
sulfur, and the W–C bond trans to phosphorus is longer than
that trans to sulfur. There are no intermolecular contacts of
note.
Molybdenum and chromium complexes. The molybdenum
complexes 10 (tetracarbonyl, 22%) and 11 (pentacarbonyl,
47%) and chromium complexes 12 (tetracarbonyl, 61%) and 13
(pentacarbonyl, 10%) were synthesised in an analogous manner
to the tungsten species (Scheme 2) with crude reaction mixtures
being purified by column chromatography (neutral grade II
alumina, under nitrogen, CH2Cl2–hexane). All the species are
air-sensitive and similar shifts in NMR signals are observed as
for 8 and 9.
Conversion of 11 to 10. A toluene solution of 11 was refluxed
and the reaction monitored by IR and 31P{1H} NMR spectro-
scopy, which indicated that complete conversion of 11 into 10
had occurred after 20 h. The solvent was removed in vacuo to
yield analytically pure 10 in virtually quantitative yield. This
reaction provides evidence that 11 is the kinetic product of the
reaction and 10 is the thermodynamic product and that the
formation of penta- and tetra-carbonyl metal complexes is
also a function of (i) the purity of the starting materials
([M(CO)5thf] or [M(CO)4(thf )2]) and (ii) the reaction condi-
tions i.e. at higher temperatures, greater quantities of tetra-
carbonyl complexes are formed.
Complexes 7, 8 and 10 display an anodic pattern similar to
that of 5 (in fact, the ipc/ipa ratio for the ferrocene oxidation is
close to unity at low scan rate and decreases at high scan rates).
Comparison of the oxidation paths of 10 and 11 (Fig. 6) is
Electrochemistry
It has been reported previously that the ferrocene ligand 1
exhibits a one-electron oxidation in dichloromethane solution,
which appears to be chemically reversible on the cyclic voltam-
metric time scale.15 As expected, the related ligand 2 displays
very similar voltammetric behaviour. Nevertheless, the two
derivatives differ in the stability of the respective monocations,
in that exhaustive electrolysis shows that [1]ϩ tends to reorgan-
ize to a new compound (probably a dimer, in that it affords two
slightly separated reversible oxidations at EЊЈ = ϩ0.62 V and
ϩ0.78 V, respectively (attempts to determine its nature are
under study), whereas [2]ϩ proves to be stable. The redox ability
of some of the metal complexes of ligand 1 was also investi-
gated. Fig. 5 shows the cyclic voltammetric behaviour of com-
plex 4 with respect to that of the free ligand 1. The insertion of
a PtCl2 bridge causes the ferrocene oxidation to shift towards
more positive potential values by about 0.6 V, thus indicating
that the platinum fragment exerts a rather strong electron with-
drawing effect with respect to the ferrocendiyl unit. Despite the
apparent chemical reversibility of the ferrocenophane oxid-
ation on the cyclic voltammetric time scale (the ipc/ipa ratio
approaches unity even at the low scan rate of 0.02 V sϪ1),
exhaustive one-electron oxidation reveals that [4]ϩ is unstable
i.e. cyclic voltammetric tests on the resulting solution only
detect traces of the original peak system. As shown in Fig. 5, 4
Fig. 6 Cyclic voltammetric responses recorded at a platinum electrode
on CH2Cl2 solutions containing [NBu4]PF6 (0.2 mol dmϪ3) and: (a) 10
(0.8 × 10Ϫ3 mol dmϪ3); (b) 11 (0.8 × 10Ϫ3 mol dmϪ3). Scan rate 0.2 V sϪ1
.
worthwhile as the molybdoferrocenophane 10 displays a
partially chemically reversible one-electron oxidation, whereas
the molybdoferrocene 11 undergoes a (coulometrically-tested)
chemically reversible one-electron oxidation followed by an
irreversible (molybdenum-centred) oxidation at high potential
values. This could account for the instability of the above-
3284
J. Chem. Soc., Dalton Trans., 2002, 3280–3289

View Article Online
Table 5 Electrochemical characteristics for the oxidation processes exhibited by compounds 1–12 in CH2Cl2 solution
a
Complex
EЊЈFe()/Fe()/V
∆Ep a/mV
ipc/ipa
EЊЈM()/M()/V
EЊЈM()/M()/V
1
2
ϩ0.46
ϩ0.43
ϩ1.00
ϩ1.04
ϩ0.92
ϩ0.88
ϩ0.79
ϩ0.79
ϩ0.63
ϩ0.65
69
90
140
74
140
130
170
140
70
1.0
1.0
b
3
ϩ0.89a, c
4
1.0
0.8
0.8
0.7
0.6
1.0
1.0
5
7
8
10
11
12
ϩ1.46a, c
ϩ1.17d
90
a Measured at 0.1 V sϪ1 b Measurement not possible. c Peak-potential value. d Measured at 5.12 V sϪ1
. .
Scheme 3 Suzuki coupling reaction used to carry out preliminary studies into the catalytic capabilities of the ligands.
mentioned ferrocenium congeners. It is plausible that the struc-
tural constraints imposed by such ferrocenophane assemblies
are not sufficiently flexible to support the expected elongation
of the Fe–cyclopentadienyl ring distances consequent to the
Fe()/Fe() process.23 Despite the fact that most metalloferro-
cenophanes exhibit ferrocene-centred oxidations displaying
features of chemical reversibility on the cyclic voltammetric
time scale (see for example24–27), they are commonly unstable in
the oxidised ferrocenium state,23–25 or at least in most cases their
stability during the long times of exhaustive electrolysis has not
been tested. Obviously this is not a general rule, in that, as
illustrated in Fig. 7, the chromoferrocenophane 12 not only is
stable in the Fe()–Cr() oxidation state (the exhaustive one-
electron oxidation at ϩ0.8 V proves that [12]ϩ is quite stable),
but also the subsequent Cr()/Cr() oxidation manifests some
degree of chemical reversibility, thus indicating that the Fe()–
Cr() oxidation state is not totally unaccessible. Table 5 sum-
marizes the electrode potentials of the above discussed redox
changes.
Catalysis studies
Fig. 7 Cyclic voltammetric responses recorded at a platinum electrode
on a CH2Cl2 solution of 12 (0.9 × 10Ϫ3 mol dmϪ3). [NBu4]PF6 (0.2 mol
The Suzuki reaction. The ligands 1,1Ј-bis(mesitylthio)-
ferrocene,28 1Ј-(mesitylthio)ferrocene-1-thiol10 and
were
1
dmϪ3) supporting electrolyte. Scan rates: (a) 0.2 VsϪ1; (b) 5.12 V sϪ1
.
investigated with Pd() reagents as possible systems for the
Suzuki C–C bond forming reaction.29 The studies are prelimin-
ary but do give an indication into their catalytic activities. 1,1Ј-
bis(diphenylphosphino)ferrocene and triphenylphosphine were
chosen as suitable ligands to benchmark any activity for the
Suzuki reaction.30 The reaction studied is shown in Scheme 3,
along with the catalytic species tested and the yields of product
obtained.
assumed that the ligands form 1 : 1 complexes with the pal-
ladium() precursor, although this still needs to be proven. The
solutions were heated at 70 ЊC for 5 h and the work up carried
out as described later. The isolated yields of 4-methylbiphenyl
are shown in Scheme 3. The results are reproducible (the
experiment was carried out three times), and these prelimin-
ary studies show that there is encouraging catalysis with lig-
and 1 in the synthesis of biaryls via the Suzuki coupling
reaction. The ligand gives double the yield obtained with
The reactants were mixed with two equivalents of base and
suspended in toluene. To this solution, a one mole percent solu-
tion of the ligands and palladium() precursor were added. It is
J. Chem. Soc., Dalton Trans., 2002, 3280–3289
3285

View Article Online
1,1Ј-bis(diphenylphosphino)ferrocene and close to the value
obtained with triphenylphosphine, under these experimental
conditions.Ligands1,1Ј-bis(mesitylthio)ferroceneand1Ј-(mesityl-
thio)ferrocene-1-thiol are not useful for this coupling reaction,
in the case of the latter this is presumably due to its charge and
that it would almost certainly oxidise the palladium() centre
prior to any reaction with the substrates, thus hindering the
reductive elimination step in the catalytic cycle.
ferrocene.15) Found: C 66.38, H 5.09%; Anal. calc. for
C23H21FePS: C 66.35, H 5.05 %. H NMR δ (CDCl3): 2.23 (s,
3H, SCH3), 4.04 (q, 2H, C5H4), 4.11 (q, 2H, C5H4), 4.19 (t,
2H, C5H4), 4.38 (t, 2H, C5H4), 7.31 (m, 10 H, C6H5); 31P{1H}
NMR δ (CDCl3): Ϫ16.78; m/z: 416 (M)ϩ, 401 (M Ϫ Me)ϩ, 369
(M Ϫ SMe)ϩ, 339 (M Ϫ Ph)ϩ, 232 (M Ϫ PPh2)ϩ.
1
1-(Diphenylphosphino)-1Ј-(mesitylthio)ferrocene (2)
1,1Ј-Dilithioferrocene (generated from 5 g of ferrocene) was
suspended in dry hexane (100 mL). To the dropping funnel was
added dimesityl disulfide (6.50 g, 21.5 mmol, 0.8 equiv.),
chlorodiphenylphosphine (3.85 g, 21.5 mL, 0.8 equiv.) and dry
toluene (50 mL). This solution was added dropwise over around
10 min to the suspension of 1,1Ј-dilithioferrocene causing
warming and darkening of the mixture which was stirred for
15 h. Water (20 mL) was added and the suspension stirred for a
further 1 h. A solid was formed which was filtered off and the
supernatant layers were separated and the aqueous layer
washed with dry hexane (2 × 10 mL). The combined organic
extracts were dried (MgSO4) and solvent was removed in vacuo
to give an orange oil. Column chromatography (neutral grade
II alumina, CH2Cl2–hexane, 3 : 7) enabled separation of
1-(diphenylphosphino)-1Ј-(mesitylthio)ferrocene (1.40 g, 2.70
mmol, 10% from ferrocene). Found: C 71.44, H 5.82 %; Anal.
Conclusions
The improved synthesis of 1 and also a new route to 1-(diphenyl-
phosphino)-1Ј-(mesitylthio)ferrocene 2 have been reported. The
coordination chemistry of 1 has been explored with Group 10
metal and Rh() halides and the ligand forms chelating com-
plexes in high yield. A monodentate bis ligand complex can be
formed using Pd() and three equivalents of 1 and the complex
shows exclusively trans phosphorus ligation. The coordination
chemistry of 1 with M(CO)5THF (M is Cr, Mo and W) shows
that the ligand forms a mixture of monodentate, P-bound
pentacarbonyl and P/S-chelating tetracarbonyl products. The
monodentate pentacarbonyl product can be converted into the
chelating tetracarbonyl via prolonged reflux in toluene and this
is thought to be the kinetic product of the complexation reac-
tion, whilst the tetracarbonyl is the thermodynamic product.
Preliminary studies into the efficacy of ligands synthesised
recently in the Long group, including 1, in combination with
Pd() precursors as catalysts for the Suzuki coupling reaction
show that 1 is a good ligand for this reaction and future studies
will focus on this.
1
calc. for C31H29FePS: C 71.54, H 5.58%. H NMR δ (CDCl3):
2.30 (s, 3H, CH3), 2.55 (s, 6H, CH3), 4.05 (t, 2H, C5H4), 4.25 (m,
4H, C5H4), 4.53 (t, 2H, C5H4), 7.33 (m, 6 H, C6H5), 7.50 (m,
6 H, C6H5); 31P{1H} NMR δ (CDCl3): Ϫ16.50 ppm; m/z: 520
(M)ϩ, 401 (M Ϫ Mes)ϩ, 370 (M Ϫ SMes)ϩ, 336 (M Ϫ PPh2)ϩ.
Dibromo[1-(diphenylphosphino)-1Ј-(methylthio)ferrocene]-
nickel(II) (3)
Experimental
(1,2-Dimethoxyethane)dibromonickel() (0.12 g, 0.4 mmol, 1
equiv.) was added to a dry Schlenk tube containing 1-(diphenyl-
phosphino)-1Ј-(methylthio)ferrocene (0.20 g, 0.5 mmol, 1.25
equiv.). Dry, deoxygenated CH2Cl2 (50 mL) was injected into
the Schlenk tube, under nitrogen, causing the solution to turn
green within a few minutes. The reaction was stirred for 40 h at
room temperature. A brown precipitate was visible in the green
solution; the solution was removed by cannula and evaporated
to dryness, then washed (20 mL heptane) to remove any un-
reacted ligand. Analysis of the brown precipitate showed it to
be insoluble in all solvents tested, microanalysis showed it to be
mostly inorganic in composition. The green powder isolated
from the filtrate was the required product (0.18 g, 0.3 mmol,
75%). Crystals suitable for X-ray studies were grown by layering
a CH2Cl2 solution of the complex with pentane. Found: C
44.09, H 3.05%; Anal. calc. for C23H21Br2FeNiPS: C 43.46 H
3.31%. m/z: 635 (M)ϩ, 588 (M Ϫ SMe)ϩ, 555 (M Ϫ Br)ϩ, 475 (M
Ϫ 2Br)ϩ, 450 (M Ϫ PPh2)ϩ.
All preparations were carried out using standard Schlenk tech-
niques.31 All solvents were distilled over standard drying agents
under nitrogen directly before use and all reactions were carried
out under an atmosphere of nitrogen. Alumina gel (neutral –
grade II) was used for chromatographic separations unless
specified otherwise.
All NMR spectra were recorded using a Delta upgrade on a
Jeol EX270 MHz spectrometer operating at 250.1 MHz (1H)
and 101.3 MHz (31P{1H}) respectively. Chemical shifts are
1
reported in δ using CDCl3 (1H, δ 7.25) as the reference for H
spectra, while the 31P{1H} spectra were referenced to H3PO4.
Infrared spectra were recorded using NaCl solution cells
(CH2Cl2 or thf ) on a Mattson Polaris Fourier Transform IR
spectrometer. Mass spectra were recorded using positive FAB
methods, on an Autospec Q mass spectrometer. Microanalyses
were carried out by SACS, University of North London.
Material and apparatus for electrochemistry have been
described elsewhere.32
Dichloro[1-(diphenylphosphino)-1Ј-(methylthio)ferrocene]-
platinum(II) (4)
1-(Diphenylphosphino)-1Ј-(methylthio)ferrocene (1)
1,1Ј-Dilithioferrocene (generated from 5 g of ferrocene) was
suspended in dry hexane (70 mL). To the dropping funnel was
added dimethyl disulfide (1.93 mL, 21.5 mmol, 0.8 equiv.),
chlorodiphenylphosphine (3.85 g, 21.5 mL, 0.8 equiv.) and
dry hexane (10 mL). This solution was added dropwise over
ca. 10 min to the suspension of 1,1Ј-dilithioferrocene causing
warming and darkening of the mixture and this was stirred for
15 h. Water (20 mL) was added and the suspension stirred
for further 1 h. A solid was formed which was filtered off and
the supernatant layers were separated and the aqueous layer
washed with dry hexane (2 × 10 mL). The combined organic
extracts were dried (MgSO4) and solvent removed in vacuo
to give a foul-smelling orange oil. Column chromatography
(neutral grade II alumina, diethyl ether–hexane, 1 : 4) enabled
separation of 1-(diphenylphosphino)-1Ј-(methylthio)ferrocene
(4.48 g, 10.77 mmol, 40% from ferrocene, cf. 21% from
1-(Diphenylphosphino)-1Ј-(methylthio)ferrocene (0.25 g, 0.6
mmol) was dissolved in dry toluene (3 mL) and added via can-
nula to dichloro(1,5-cyclooctadiene)platinum() (0.20 g, 0.53
mmol) dissolved in dry CH2Cl2 (7 mL). The colour of the solu-
tion changed from pale yellow to orange on completion of the
addition and a precipitate formed and the reaction mixture was
stirred for 15 h. The yellow precipitate was filtered off and
washed with dry hexane (10 mL) and dry CH2Cl2 (2 mL).
Recrystallisation of the precipitate was achieved by dissolution
in CH2Cl2 and layering with hexane enabling isolation of the
product as yellow needles (0.33 g, 0.48 mmol, 91%). Found: C
39.93, H 2.95%; Anal.calc. for C23H21Cl2FePPtS: C 40.47, H
3.08%. 1H NMR δ (CDCl3): 2.86 (s, 3H, SCH3, 2JPt–H 23.25Hz),
4.39 (br, 4H, C5H4), 4.83 (br, 4H, C5H4), 7.43 (br m, 10H,
1
C6H5); 31P{1H} NMR δ (CDCl3): 8.64 (s, JPt–P 3737 Hz); m/z:
682 (M)ϩ, 647(M Ϫ Cl)ϩ, 611 (M Ϫ 2Cl)ϩ.
3286
J. Chem. Soc., Dalton Trans., 2002, 3280–3289

View Article Online
Dichloro[1-(diphenylphosphino)-1Ј-(methylthio)ferrocene]-
palladium(II) (5)
phosphino)-1Ј-(methylthio)ferrocene (0.15 g, 0.36 mmol, 1
equiv.) dissolved in toluene (50 mL). The yellow solution was
stirred at room temperature for 20 h and then evaporated to
dryness. The crude product was purified by column chromato-
graphy (neutral grade II alumina, CH2Cl2–hexane, 1 : 9 to
remove hexacarbonyl, 1 : 5 to remove 9 and 1 : 3 to remove 8)
which enabled isolation of compounds 9 (0.10 g, 0.14 mmol,
39%) and 8 (0.13 g, 0.18 mmol, 50%). Crystals of 8 suitable for
X-ray crystallography were grown by layering a CH2Cl2 solu-
tion with hexane. 8 Found: C 45.12, H 2.64%; Anal. calc. for
C27H21FeO4PSW: C 45.50, H 2.95%. 1H NMR δ (CDCl3): 2.90
(s, 3H, SCH3), 4.22 (m, 4H, C5H4), 4.40 (t, 2H, C5H4), 4.58 (t,
2H, C5H4), 7.38 (m, 6H, C6H5), 7.54 (m, 4H, C6H5); 31P{1H}
1-(Diphenylphosphino)-1Ј-(methylthio)ferrocene (0.25 g, 0.6
mmol) was dissolved in dry toluene (7 mL) and added via
cannula to dichloro(1,5-cyclooctadiene)palladium() (0.15 g,
0.53 mmol) dissolved in dry chloroform (18 mL). The colour of
the solution changed from yellow to red on completion of the
addition and the reaction was stirred for 15 h. The red solution
was concentrated slightly leading to precipitation of a red solid
which was filtered off and washed with dry hexane (10 mL).
Recrystallisation of the precipitate was attempted by dis-
solution in CH2Cl2 and layering with hexane. However, despite
repeated attempts an orange microcrystalline powder was
always obtained (0.26 g, 0.43 mmol, 81%). Found: C 46.40, H
3.57%; Anal. calc. for C23H21Cl2FePPdS: C 46.46, H 3.54%; 1H
NMR δ (CDCl3): 2.80 (s, 3H, SCH3), 4.37 (t, 2H, C5H4), 4.60 (t,
2H, C5H4), 4.82 (t, 2H, C5H4), 5.03 (t, 2H, C5H4), 7.42 (m, 6H,
C6H5), 7.70 (m, 4H, C6H5); 31P{1H} NMR δ (CDCl3): 31.82 (s);
m/z: 558 (M Ϫ Cl)ϩ, 523 (M Ϫ 2Cl)ϩ, 416 (PSF)ϩ.
1
NMR δ (CDCl3): 19.53 (d, JW–P 234 Hz); IR (THF): υ(CO)
2016, 1905, 1890, 1875 cmϪ1; m/z: 712 (M)ϩ, 684 (M Ϫ CO)ϩ,
656 (M Ϫ 2CO)ϩ, 628 (M Ϫ 3CO)ϩ, 600 (M Ϫ 4CO)ϩ. 9 Found:
C 45.33, H 2.76%; Anal. calc. for C28H21FeO5SPW: C 45.41, H
1
2.84%. H NMR δ(CDCl3): 2.15 (s, 3H, SCH3), 3.97 (t, 2H,
C5H4), 4.05 (t, 2H, C5H4), 4.31 (t, 2H, C5H4), 4.51 (t, 2H, C5H4),
7.42 (m, 10H, C6H5); 31P{1H} NMR δ (CDCl3): 11.15 (d, 1JW–P
245 Hz); IR (thf ): υ(CO) 2070, 1980, 1934 cmϪ1; m/z: 740 (M)ϩ,
628 (M Ϫ 4CO)ϩ, 600 (M Ϫ 5CO)ϩ.
Bis[1-(diphenylphosphino)-1Ј-(methylthio)ferrocene]dichloro-
palladium(II) (6)
1-(Diphenylphosphino)-1Ј-(methylthio)ferrocene (0.10 g, 0.24
mmol, 2.5 equiv.) was dissolved in toluene (10 mL) and added
[1-(Diphenylphosphino)-1Ј-(methylthio)ferrocene]tetracarbonyl-
molybdenum(0) (10) and [1-(diphenylphosphino)-1Ј-(methylthio)-
ferrocene]pentacarbonylmolybdenum(0) (11)
to
a solution of dichloro(1,5-cyclooctadiene)palladium()
(0.03 g, 0.095 mmol, 1 equiv.) dissolved in CH2Cl2 (5 mL). The
solution instantly changed colour from yellow to red, and was
stirred for 20 h at room temperature. The solvent was removed
and the red residue washed with hexane (3 × 10 mL) yielding 6
as an orange–red powder (0.08 g, 0.076 mmol, 80%). Crystals
suitable for X-ray diffraction were grown by slow evaporation
of a hexane–toluene solution of the complex. Found: C
54.67, H 4.93%; Anal. calc. for C46H42Cl2Fe2P2PdS2: C 54.81, H
Pentacarbonyl(tetrahydxrofuran)molybdenum() (53 mL of a
0.01 M solution in thf, 0.53 mmol, 1.5 equiv.), prepared by
UV photolysis, was injected into a solution of 1-(diphenyl-
phosphino)-1Ј-(methylthio)ferrocene (0.15 g, 0.36 mmol, 1
equiv.) dissolved in toluene (50 mL). The yellow solution was
stirred at room temperature for 20 h and then evaporated to
dryness. The crude product was purified by column chromato-
graphy (neutral grade II alumina, CH2Cl2–hexane, 1 : 9 to
remove hexacarbonyl, 1 : 5 to remove 11 and 1 : 3 to remove 10)
which enabled isolation of compounds 11 (0.11 g, 0.17 mmol,
47%) and 10 (0.05 g, 0.08 mmol, 22%). 10 Found: C 51.82, H
3.43%; Anal. calc. for C27H21FeMoO4SP: C 51.84, H 3.36%. 1H
NMR δ (CDCl3): 2.70 (s, 3H, SCH3), 4.22 (m, 4H, C5H4), 4.41
(t, 2H, C5H4), 4.57 (t, 2H, C5H4), 7.38 (m, 6H, C6H5), 7.55 (m,
4H, C6H5); 31P{1H} NMR δ (CDCl3): 30.07 (s); IR (thf ): υ(CO)
2020, 1902, 1877 cmϪ1; m/z: 625 (M)ϩ, 597 (M Ϫ CO)ϩ, 541
(M Ϫ 3CO)ϩ, 513 (M Ϫ 4CO)ϩ. 11 Found: C 51.72, H 3.19%;
1
5.00%. H NMR δ (CDCl3): 2.25 (s, 3H, SCH3), 4.39 (m, 2H,
C5H4), 4.43 (m, 2H, C5H4), 4.59 (m, 2H, C5H4), 4.63 (m, 2H,
C5H4), 7.39 (m, 6H, C6H5), 7.62 (m, 4H, C6H5); 31P{1H} NMR
δ (CDCl3): 15.57 (s); m/z: 974 (M Ϫ Cl)ϩ, 938 (M Ϫ 2Cl)ϩ, 891
(M Ϫ 2Cl Ϫ SMe)ϩ, 754 (M Ϫ 2Cl Ϫ PPh2)ϩ, 522 (M Ϫ 2Cl Ϫ
PSF)ϩ.
Carbonylchloro[1-(diphenylphosphino)-1Ј-(methylthio)-
ferrocene]rhodium(I) (7)
1-(Diphenylphosphino)-1Ј-(methylthio)ferrocene (0.16 g, 0.39
mmol, 2 equiv.) was dissolved in a minimum amount of toluene
(10 mL) and added to a solution of tetracarbonyldichloro-
dirhodium() (0.08 g, 0.19 mmol, 1 equiv.) in methanol (10 mL).
An orange precipitate formed immediately, which was filtered
off after 2 h. The precipitate was washed with diethyl ether
(20 mL) and methanol (20 mL), enabling isolation of 7 (0.09 g,
0.15 mmol, 79%). Found: C 49.46, H 3.42%; Anal. calc. for
1
Anal. calc. for C28H21FeMoO5SP: C 51.53, H 3.2 %. H NMR
δ (CDCl3): 2.15 (s, 3H, SCH3), 3.96 (t, 2H, C5H4), 4.05 (t, 2H,
C5H4), 4.31 (t, 2H, C5H4), 4.48 (t, 2H, C5H4), 7.42 (m, 10H,
C6H5); 31P{1H} NMR δ (CDCl3): 28.26 (s); IR (thf ): υ(CO)
2071, 1988, 1944 cmϪ1; m/z: 652 (M)ϩ, 540 (M Ϫ 4CO)ϩ, 512
(M Ϫ 5CO)ϩ.
1
C24H21ClFeOPRhS: C 49.40, H 3.60%. H NMR δ (CDCl3):
Chromium(0)[1-(diphenylphosphino)-1Ј-(methylthio)ferrocene]-
tetracarbonyl (12) and chromium(0)[1-(diphenylphosphino)-1Ј-
(methylthio)ferrocene]pentacarbonyl (13)
2.63 (s, 3H, SCH3), 4.21 (t, 2H, C5H4), 4.51 (t, 2H, C5H4), 4.55
(t, 2H, C5H4), 4.73 (t, 2H, C5H4), 7.37 (m, 6H, C6H5), 7.68 (m,
4H, C6H5); 31P{1H} NMR δ (CDCl3): 37.42 (d, 1JRh–P 168 Hz);
13C{1H} NMR δ (CDCl3): 24.03 (s, SCH3), 71.04 (s, C5H4),
72.59 (d, C5H4), 75.22 (d, C5H4), 75.43 (s, C5H4), 77.99 (s,
Chromium()pentacarbonyl(tetrahydrofuran) (53 mL of
a
0.01 M solution in thf, 0.53 mmol, 1.5 equiv.), prepared by
UV photolysis, was injected into a solution of 1-(diphenyl-
phosphino)-1Ј-(methylthio)ferrocene (0.15 g, 0.36 mmol, 1
equiv.) dissolved in toluene (50 mL). The yellow solution was
stirred at room temperature for 20 h and then evaporated to
dryness. The crude product was purified by column chromato-
graphy (neutral grade II alumina, CH2Cl2–hexane, 1 : 9 to
remove hexacarbonyl, 1 : 5 to remove 13 and 1 : 3 to remove 12)
enabled isolation of compounds 13 (0.02 g, 0.036 mmol, 10%)
and 12 (0.13 g, 0.22 mmol, 61%). 12 Found: C 55.79, H 3.74%;
2
C5H4), 81.91 (s, C5H4), 128.23 (d, o-C6H5, JP–C 10.38 Hz),
130.55 (s, m-C6H5), 131.42 (s, p-C6H5), 133.63 (d, ipso-C6H5,
1
1JP–C 12.46 Hz), 135.47 (d, CO, JRh–C 52 Hz); IR (CH2Cl2):
υ(CO) 2006 cmϪ1; m/z: 583 (M)ϩ, 555 (M Ϫ CO)ϩ, 547 (M
Ϫ Cl)ϩ, 519 (M Ϫ CO Ϫ Cl)ϩ.
[1-(Diphenylphosphino)-1Ј-(methylthio)ferrocene]tetracarbonyl-
tungsten(0) (8) and [1-(diphenylphosphino)-1Ј-(methylthio)-
ferrocene]pentacarbonyltungsten(0) (9)
1
Anal. calc. for C27H21CrFeO4PS: C 55.86, H 3.62%. H NMR
Pentacarbonyl(tetrahydrofuran)tungsten() (53 mL of
0.01 M solution in thf, 0.53 mmol, 1.5 equiv.), prepared by
UV photolysis, was injected into a solution of 1-(diphenyl-
a
δ (CDCl3): 2.65 (s, 3H, SCH3), 4.20 (t, 2H, C5H4), 4.27 (t, 2H,
C5H4), 4.38 (t, 2H, C5H4), 4.56 (t, 2H, C5H4), 7.38 (m, 6H,
C6H5), 7.58 (m, 4H, C6H5); 31P{1H} NMR δ (CDCl3): 46.22 (s);
J. Chem. Soc., Dalton Trans., 2002, 3280–3289
3287

View Article Online
Table 6 Crystal data, data collection and refinement parameters for compounds 3, 4, 6 and 8a
3
4
6
8
Formula
C23H21PSBr2FeNi
634.8
Dark olive green blocks
0.70 × 0.57 × 0.19
293
C23H21PSCl2FePt
682.3
Yellow prisms
0.17 × 0.13 × 0.13
293
Monoclinic
C2/c (no. 15)
11.450(1)
22.142(3)
18.201(2)
—
108.07(1)
—
4386.9(8)
8
2.066
7.45
C46H42P2S2Cl2Fe2Pd
1009.9
Orange prisms
0.32 × 0.13 × 0.07
293
Monoclinic
P21/c (no. 14)
9.434(1)
11.605(1)
18.583(1)
—
C27H21O4PSFeW
712.2
Orange rhombs
0.57 × 0.47 × 0.23
293
Monoclinic
P21/n (no. 14)
11.502(2)
17.863(2)
12.446(1)
—
95.60(1)
—
2545.0(6)
4
1.859
5.26
Formula weight
Colour, habit
Crystal size/mm
Temperature/K
Crystal system
Space group
a/Å
b/Å
c/Å
α/Њ
β/Њ
Triclinic
¯
P1 (no. 2)
9.928(1)
15.880(3)
17.155(2)
64.82(1)
78.56(1)
87.62(1)
2396.4(5)
4b
94.32(1)
—
γ/Њ
V/Å3
2028.7(2)
Z
2c
Dc/g cmϪ3
1.760
4.89
1.653
12.38
µ/mmϪ1
Radiation used
θ range/Њ
Mo-Kα
2.1–25.0
Mo-Kα
1.8–25.0
Cu-Kα
4.5–60.0
Mo-Kα
2.0–25.0
No. of unique reflections
measured
8109
5159
Gaussian
0.43, 0.08
488
3874
3145
Ellipsoidal
0.47, 0.35
239
3017
2389
Ellipsoidal
0.34, 0.18
227
4375
3683
Empirical
0.78, 0.36
293
observed, |Fo| > 4σ(|Fo|)
Absorption correction
Max., min. transmission
No. of variables
d
R1, wR2
0.049, 0.095
0.031, 0.060
0.048, 0.114
0.032, 0.071
a Details in common: graphite monochromated radiation, refinement based on F 2. b There are two crystallographically independent molecules in the
2
2
2
asymmetric unit. c The complex has crystallographic Ci symmetry. d R1 = Σ||Fo| – |Fc||/Σ|Fo|; wR2 = {Σ[w(Fo – Fc2)2]/Σ[w(Fo )2]}1/2;wϪ1 = σ2(Fo ) ϩ
(aP)2 ϩ bP.
IR (thf ): υ(CO) 2010, 1908, 1891, 1876 cmϪ1; m/z: 580 (M)ϩ,
Acknowledgements
468 (M Ϫ 4CO)ϩ. 13 Found: C 55.34, H 3.44%; Anal. calc. for
C28H21CrFeO5PS: C 55.26, H 3.45%. 1H NMR δ (CDCl3): 2.16
The EPSRC are thanked for funding a studentship (C. K. W.),
Johnson Matthey plc for the loan of precious metal salts and
BP Chemicals Ltd. for their assistance in the catalysis studies.
(s, 3H, SCH3), 3.94 (t, 2H, C5H4), 4.05 (t, 2H, C5H4), 4.31 (t,
2H, C5H4), 4.47 (t, 2H, C5H4), 7.43 (m, 10H, C6H5); 31P{1H}
P. Z. gratefully acknowledges the financial support of the
NMR δ (CDCl3): 47.22 (s); IR (thf ): υ(CO) 2061, 1988, 1939
cmϪ1; m/z: 608 (M)ϩ, 468 (M Ϫ 5CO)ϩ.
University of Siena (PAR 2001).
Suzuki coupling reactions
References
The experimental procedure used to test the ligands as catalysts
1 A. Bader and E. Lindner, Coord. Chem. Rev., 1991, 108, 27.
in the Suzuki coupling reaction: 4-bromotoluene (1.71 g, 10
mmol, 1 equiv.), phenyl boronic acid (1.34 g, 11 mmol,
1.1.equiv.) and potassium carbonate (2.76 g, 20 mmol, 2 equiv.)
were placed in a dry two-necked round bottomed flask fitted
with a reflux condenser and under nitrogen. The starting
materials were suspended in dry, deoxygenated toluene (30 mL).
To the suspension was added a solution of Pd2dba3 (0.046 g,
0.05 mmol, 0.005 equiv.) and the ligand (0.12 mmol, 0.012
mmol) in toluene (20 mL). The reaction was heated to 70 ЊC
and maintained at this temperature for 5 h. The solvent was
removed and the residue suspended in diethyl ether (40 mL),
washed with brine (3 × 30 mL) and evaporated to dryness. The
crude product was purified by column chromatography (silica
gel, hexane) and the solvent evaporated to give a white powder
(the yield of which is quoted for each of the ligand systems in
Scheme 3 and used to assess their catalytic activities). The
purity of the product was determined by GC.
2 (a) For a detailed literature review, see: Ferrocenes: Homogeneous
Catalysis – Organic Synthesis – Materials Science, ed. A. Togni and
T. Hayashi, VCH, Weinheim, 1995; (b) Metallocenes, ed. A. Togni
and R. L. Haltermann, Wiley-VCH, Weinheim, 1998.
3 For a comprehensive overview of ferrocene and other metallocene
chemistry see: N. J. Long, Metallocenes: An Introduction to
Sandwich Complexes, Blackwell Science, Oxford, 1998.
4 (a) Review article: J. R. Dilworth and N. Wheatley, Coord. Chem.
Rev., 2000, 199, 89; (b) For example: N. Brugat, A. Polo,
A. Alvarez-Larena, J. F. Piniella and J. Real, Inorg. Chem., 1999,
38, 4829; (c) E. Hauptman, P. J. Fagan and W. Marshall,
Organometallics, 1999, 18, 2061; (d ) R. Cao, M. Hong, F. Jiang,
B. Kang, X. Xie and H. Liu, Polyhedron, 1996, 15, 2661.
5 (a) For example: P. Perez-Lourido, J. Romero, J. A. Garcia-Vezquez,
A. Sousa, Y. Zheng, J. R. Dilworth and J. Zubieta, J. Chem. Soc.,
Dalton Trans., 2000, 769; (b) M. S. J. Foreman and J. D. Woollins,
J. Chem. Soc., Dalton Trans., 2000, 1533; (c) S. M. Aucott,
A. M. Z. Slawin and J. D. Woollins, J. Chem. Soc., Dalton Trans.,
2001, 2279; (d ) D. Morales-Morales, R. Redon, Y. Zheng and
J. R. Dilworth, Inorg. Chim. Acta, 2002, 328, 39; (e) A. Albinati,
J. Eckert, P. Pregosin, H. Ruegger, R. Salzmann and C. Stossel,
Organometallics, 1997, 16, 579.
6 For example: U. Klabunde and S. D. Ittel, J. Mol. Catal., 1987, 41,
123.
7 For example: W. Keim, R. Appel, S. Gruppe and F. Knoch,
Angew. Chem., Int. Ed. Engl., 1987, 26, 1012.
8 For example: S. Y. Desjardins, K. J. Cavell, J. L. Hoare, B. W.
Skelton, A. N. Sobolev, A. H. White and W. Keim, J. Organomet.
Chem., 1997, 544, 163.
9 (a) I. R. Butler, U. Griesbach, P. Zanello, M. Fontani, D. Hibbs,
M. B. Hursthouse and K. L. M. A. Malik, J. Organomet. Chem.,
1998, 565, 243; (b) I. R. Butler and S. C. Quayle, J. Organomet.
Crystallography
Table 6 provides a summary of the crystallographic data for
compounds 3, 4, 6 and 8. Data were collected on Siemens P4/
PC diffractometers using ω-scans. The structures were solved
by direct methods and they were refined based on F 2 using the
SHELXTL program system.33
CCDC reference numbers 187540–187543.
See http://www.rsc.org/suppdata/dt/b2/b205653d/ for crystal-
lographic data in CIF or other electronic format.
3288
J. Chem. Soc., Dalton Trans., 2002, 3280–3289

View Article Online
Chem., 1998, 552, 63; (c) I. R. Butler, M. G. B. Drew,
C. H. Greenwell, E. Lewis, M. Plath, S. Mussig and J. Szewezyk,
Inorg. Chem. Commun., 1999, 2, 576; (d ) W.-P. Deng, X.-L. Hou,
L.-X. Dai, Y.-H. Yu and W. Xia, Chem. Commun., 2000, 285;
(e) R. Broussier, S. Ninoreille, C. Bouron, O. Blacque, C. Ninoreille,
M. M. Kubicki and B. Gautheron, J. Organomet. Chem., 1998, 561,
85; ( f ) G. A. Balavoine, J.-C. Daran, G. Iftiem, E. Manoury
and C. Moveau-Bossuet, J. Organomet. Chem., 1998, 567, 191;
(g) V. C. Gibson, N. J. Long, A. J. P. White, C. K. Williams and
D. J. Williams, Organometallics, 2002, 21, 770.
20 I. R. Butler, W. R. Cullen, T.-J. Kim, S. J. Rettig and J. Trotter,
Organometallics, 1985, 4, 972.
21 I. V. Nelson and R. T. Iwamoto, J. Electroanal. Chem., 1964, 7, 218.
22 E. R. Brown and J. R. Sandifer, in Physical Methods of Chemistry,
Electrochemical Methods, eds. B. W. Rossiter and J. F. Hamilton,
Wiley, New York, 1986, vol. 2, ch. 4.
23 J. K. Pudelski, D. A. Foucher, C. H. Honeyman, A. J. Lough,
I. Manners, S. Barlow and D. O’Hare, Organometallics, 1995, 14,
2470 and references therein.
24 (a) C. K.-S. Gan T. S. A. Hor, in Ferrocenes, eds. A. Togni
and T. Hayashi, VCH, Weinheim, 1995, ch. 1, and references therein;
(b) P. Zanello, in Ferrocenes, eds. A. Togni and T. Hayashi, VCH,
Weinheim, 1995, ch. 7, and references therein.
10 V. C. Gibson, N. J. Long, A. J. P. White, C. K. Williams and
D. J. Williams, Chem. Commun., 2000, 2359.
11 (a) J. A. Adeleke, Y.-W. Chen and L.-K. Liu, Organometallics, 1992,
11, 2543; (b) J. A. Adeleke and L.-K. Liu, J. Chin. Chem. Soc., 1992,
39, 61.
25 S. Takemoto, S. Kuwata, Y. Nishibayashi and M. Hidai,
Organometallics, 2000, 19, 3249.
12 T. Y. Dong and C.-K. Chang, J. Chin. Chem. Soc., 1998, 45, 577.
13 I. R. Butler, S. Mussig and M. Plath, Inorg. Chem. Commun., 1999,
2, 424.
14 A. Alexakis and C. Benhaim, Tetrahedron: Asymmetry, 2001, 12,
1151.
26 N.-S. Choi and S. W. Lee, J. Organomet. Chem., 2001, 627, 226.
27 L. L. Maisela, A. M. Cruch, J. Darkwa and I. A. Guzei, Polyhedron,
2001, 20, 3189.
28 V. C. Gibson, N. J. Long, A. J. P. White, C. K. Williams and
D. J. Williams, Organometallics, 2000, 19, 4425.
15 N. J. Long, J. Martin, G. Opromolla, A. J. P. White, D. J.
Williams and P. Zanello, J. Chem. Soc., Dalton Trans., 1999,
1981.
16 B. McCulloch, D. L. Ward, J. D. Woollins and C. H. Brubaker, Jr.,
Organometallics, 1985, 4, 1425.
17 J. R. Dilworth, J. R. Miller, N. Wheatley, M. Baker and J. G. Sunley,
J. Chem. Soc., Chem. Commun., 1995, 1579.
18 R. Broussier, E. Bentabet, M. Laly, P. Richard, L. G. Kuz’mina,
P. Serp, N. Wheatley, P. Kalck and B. Gautheron, J. Organomet.
Chem., 2000, 613, 77.
29 (a) T. Oh-e, N. Miyaura and A. Suzuki, Synlett, 1990, 221;
(b) T. Oh-e, N. Miyaura and A. Suzuki, J. Org. Chem., 1993, 58,
2201.
30 T. J. Colacot, Platinum Met. Rev., 2001, 45, 22.
31 R. J. Errington, Advanced Practical Inorganic and Metalorganic
Chemistry, Blackie, London, 1997.
32 F. Fabrizi de Biani, F. Laschi, P. Zanello, G. Ferguson, J. Trotter,
G. M. O’Riordan and T. R. Spalding, J. Chem. Soc., Dalton Trans.,
2001, 1520.
33 SHELXTL PC version 5.03, Siemens Analytical X-Ray
Instruments, Inc., Madison, WI, 1994; SHELXTL PC version 5.1,
Bruker AXS, Madison, WI, 1997.
19 L.-T. Phang, S. C. F. Au-Yeung, T. S. A. Hor, S. B. Khoo, Z.-Y. Zhou
and T. C. W. Mak, J. Chem. Soc., Dalton Trans., 1993, 165.
J. Chem. Soc., Dalton Trans., 2002, 3280–3289
3289