New unsymmetrical thioether- and thiolate-substituted ferrocenediyl ligands and their metal complexes

Article abstract of DOI:10.1039/b306982f

A series of new unsymmetrical 1,1′-disubstituted ferrocenediyl ligands featuring thioether or thiolate substituents have been conveniently synthesised. The chelating nature of the ligands is shown by coordination with a range of transition metal centres and bridged metal dimeric species are observed with platinum and palladium. Electrochemical studies show that the diferrocene species is a partially delocalised Robin-Day Class II system, whilst on coordination of metal centres to the monoferrocene ligands, the oxidation potential of the ferrocene subunit is greatly increased ranging from 0.25 to 0.60 V upon passing from platinum(II) to palladium(II) ions. Interestingly, the outer ferrocenediyl subunits of the dimeric complexes indicate electronic interaction through the diplatinum or dipalladium backbones. The chromium- and vanadium-containing species act as low activity, pre-catalysts for ethylene polymerisation.

Full text of DOI:10.1039/b306982f

View Article Online / Journal Homepage / Table of Contents for this issue
New unsymmetrical thioether- and thiolate-substituted
ferrocenediyl ligands and their metal complexes
Vernon C. Gibson,a Nicholas J. Long,*a Charlotte K. Williams,a Marco Fontanib
and Piero Zanellob
a Department of Chemistry, Imperial College London, South Kensington, London,
UK SW7 2AY. E-mail: n.long@imperial.ac.uk
b Dipartimento di Chimica dell’Università di Siena, Via Aldo Moro, 53100 Siena, Italy
Received 18th June 2003, Accepted 22nd July 2003
First published as an Advance Article on the web 15th August 2003
A series of new unsymmetrical 1,1Ј-disubstituted ferrocenediyl ligands featuring thioether or thiolate substituents
have been conveniently synthesised. The chelating nature of the ligands is shown by coordination with a range of
transition metal centres and bridged metal dimeric species are observed with platinum and palladium. Electro-
chemical studies show that the diferrocene species is a partially delocalised Robin–Day Class II system, whilst
on coordination of metal centres to the monoferrocene ligands, the oxidation potential of the ferrocene subunit is
greatly increased ranging from 0.25 to 0.60 V upon passing from platinum() to palladium() ions. Interestingly,
the outer ferrocenediyl subunits of the dimeric complexes indicate electronic interaction through the diplatinum or
dipalladium backbones. The chromium- and vanadium-containing species act as low activity, pre-catalysts for
ethylene polymerisation.
at room temperature for 20 h. After aqueous work-up,
Introduction
followed by column chromatography (neutral grade II alumina,
CH2Cl2:hexane 1:4) 2 was isolated in ca. 50% yield. This is the
maximum yield possible, as cleavage of the disulfide bond
by methyllithium generates one molecule of 2 and one of
1Ј-mesitylthio)ferrocene-1-thiolate (which is oxidised during
work-up to regenerate the starting material). It should be noted
that we have recently developed6 a higher yielding route to
a range of unsymmetrical thioethers, by utilising lithium
triethylborohydride to break the disulfide bond followed by
In coordination chemistry, the ferrocenyl moiety has played a
significant role as a backbone or a substituent in ancilliary
ligands due to (i) the specific and unique geometries that the
ferrocene provides and (ii) its electronic (redox) properties,
whereby the possibility of switching the redox active state of
the ferrocene backbone gives access to potential control of
reactivity at a metal centre.1,2 Ferrocene species with donor
heteroatoms (e.g. P, N, S, O) substituted onto the cyclo-
pentadienyl rings are well-known especially those featuring
homo-substitution (i.e. the same donor atom along with the
same alkyl or aryl substituents) but much less well-known are
unsymmetrical 1,1Ј-disubstituted ferrocenes featuring hetero-
combinations of N, P or S atoms3 and as far as we are aware
our communication in 20004 featured the first examples of
unsymmetrical purely S-substituted neutral and neutral-
anionic ferrocenediyl ligands. Complete syntheses of the new
‘neutral–neutral’ 1-(methylthio)-1Ј-(mesitylthio)ferrocene and
the ‘neutral–anionic’ 1Ј-(mesitylthio)ferrocene-1-thiol ligands
is detailed here, along with their Pt(), Pd(), Cr() and V()
coordination chemistry, and potential to polymerise olefins.
1
reaction with RX (X = halide). The H NMR spectrum of
2 shows three methyl signals in increasing δ values due to
para-mesityl methyl, sulfur methyl and ortho-mesityl methyls,
respectively, and in the cyclopentadienyl proton region there
are four pseudo-triplets emphasising the asymmetry of the
substitution.
1Ј-(Mesitylthio)ferrocene-1-thiol (3) was also synthesised in
high yield via the treatment of a THF solution of 1 at Ϫ78 ЊC,
with two equivalents of lithium triethylborohydride. After
stirring at room temperature for 2 h, the solvent was removed
in vacuo and the crude residue dissolved in dry, oxygen-
free diethyl ether. Under anaerobic conditions and utilising
nitrogen-saturated reagents, the air-sensitive orange 3 was
isolated in 88% yield. In the 1H NMR spectrum of 3, the
thiolate proton is observed at 2.98 ppm along with four pseudo-
triplets in the cyclopentadienyl region, along with characteristic
mesityl methyl and mesityl proton signals at 2.21, 2.48 and 6.86
ppm, respectively.
Results and discussion
(i) Synthesis of ligands
The new disulfide, di[1Ј-(mesitylthio)ferrocene]disulfide (1) was
formed via a modification of the synthetic methods developed
by Herberhold et al. (Scheme 1).5 A THF solution of 1,2,3-
trithia-[3]ferrocenophane was treated with two equivalents of
mesityllithium and the mixture stirred at room temperature for
20 h. Purification via column chromatography (neutral grade II
alumina, CH2Cl2–hexane (1 : 9)) enabled separation of 1 in ca.
30% yield (other products isolated were dimesityl disulfide and
a small amount of the original ferrocenophane). The proposed
mechanism is shown in Scheme 2, where the S–S bonds are
cleaved by mesityl ion leading to mesityl thiolate and 1Ј-(mesit-
ylthio)ferrocene-1-thiolate, which are both oxidised and S–S
bonds reformed on aerobic work-up of the reaction mixture.
From 1, both 1-(methylthio)-1Ј-(mesitylthio)ferrocene 2 and
1Ј-(mesitylthio)ferrocene-1-thiol 3 can be formed in good yield
(Scheme 1). For 2, a THF solution of 1 was treated with
2 equivalents of methyllithium and the reaction mixture stirred
Reaction of 1 with lithium triethylborohydride results in the
near quantitative yield of the lithium thiolate salt, which can be
used in situ in reaction with transition metal reagents (see later).
(ii) Metal complexes of 2
The reactivity of 2 was investigated with platinum() and
palladium() precursors and compared to analogous reactions
involving 1,1Ј-(bismethylthio)ferrocene7 and 1,1Ј-bis(mesityl-
thio)ferrocene.8
A slight excess of the ligand was added to a toluene solution
of bis(benzonitrile)dichlorometal() (where metal is PtII or
PdII) and the mixture heated to 60 ЊC. The yellow (for 4) and
purple (for 5) solutions were heated at 60 ЊC for 20 h and hot
filtered to remove any unreacted metal complex. Complexes
4 and 5 were isolated in 91 and 85% yields, respectively, via
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
D a l t o n T r a n s . , 2 0 0 3 , 3 5 9 9 – 3 6 0 5
3599

View Article Online
Scheme 2 Suggested mechanism for the formation of 1.
dichlorometal() (M = Pt or Pd) in toluene at 60 ЊC gave an
initial precipitate of lithium chloride which was filtered off and
after evaporation of the filtrate to dryness and washing with
hexane, enabled isolation of 6 and 7 as yellow (93%) and purple
(81%) powders, respectively. The bridging, dinuclear nature of
the complexes is unusual, especially in ferrocenyl systems and
only limited analogous examples are known.9
1
Due to excessive broadening of the signals, the H NMR
spectra were run at low temperatures (ca. Ϫ65 ЊC) and showed
striking similarities. For 6, the methyl groups on the mesityl
ring show separate peaks (rather than the 2:1 ratio normally
observed), with one resonance for a methyl group at 3.32 ppm –
this deshielding is presumably due to significant interaction
with the platinum metal centre. As expected for asymmetry
resulting from a slowing of the fluxional processes, the ferro-
cenyl region shows eight separate resonances with similar sized
3JH–H couplings, although due to slight broadening these could
not be accurately measured. Once again, all the signals are
shifted downfield with respect to the free ligand, due to the
effect of the platinum centres. The spectrum of 7 (at 65 ЊC) is
very similar to 6 in the shape, number and shift of the signals.
For 6, mass spectrometry indicates the presence of the molec-
ular ion (196 amu) and fragmentation peaks corresponding to
loss of chloride groups (1160 and 1125 amu). For 7, a molecular
ion is not observed, but instead peaks due to loss of mesityl
(897 amu) and mesitylthio (870) groups are apparent. The
X-ray structure of 7 confirms the bridging dinuclear struc-
ture and this has been published previously in our initial
communication.4
It should be noted that all attempts to synthesise a nickel()
complex with this new ligand failed to produce any tractable
product. We propose that this is due to steric factors, as to date
reactions with mesityl-containing neutral–neutral ferrocenediyl
thioether ligands have also failed.
Complexes 8 and 9 were formed using 1-(lithiumthiolato)-1Ј-
(mesitylthio)ferrocene which was syringed into a solution of
M(thf )3Cl3 (M = Cr, V) at Ϫ78 ЊC. Colour changes were
immediately observed, purple to yellow for chromium and
salmon-pink to dark red for vanadium. On stirring at room
temperature, both solutions became deep red and following
evaporation to dryness, washing, extraction and filtration, dark
purple solids were obtained in 48% yield for 8 and 43% yield for
9. The paramagnetic 8 and 9 are highly air-sensitive materials
but reasonable microanalysis results were obtained, and though
mass spectrometry was inconclusive, the magnetic susceptibility
Scheme 1 Formation of ligands 1–3 and complexes 4–9. Reagents and
conditions: (i) MesLi, THF, Ϫ78 ЊC, then stirring at 25 ЊC, 16 h; (ii)
MeLi (1.6 M), THF, stirring, 16 h; (iii) LiEt3BH, THF, stirring, 1.5 h
(stop here for lithio salt and use in situ) then for thiol formation KOH/
diethyl ether, HCl until pH 2; (iv) MCl2(PhCN)2 (M = Pt or Pd),
toluene, 60 ЊC, 20 h; (v) (from lithio salt) MCl2(PhCN)2 (M = Pt or Pd),
toluene, 60 ЊC, 16 h; (vi) (from lithio salt) MCl3(THF)3 (M = Cr or V),
THF, Ϫ78 ЊC, then stirring at 25 ЊC, 16 h.
washing and reprecipitation. In the positive ion FAB mass
spectra, neither complex showed a molecular ion but did
show peaks indicating loss of chlorine groups. In the room-
temperature 1H NMR spectra, both complexes gave signals
characteristically shifted to lower field compared to the free
ligand. Most of the signals in the spectra were broadened,
especially in the ferrocenyl region due to the fluxional nature of
the molecule via bridge reversal of the ferrocenophane linkage
and sulfur inversion of the R groups.7 Variable-temperature
spectra were not taken as it was assumed that the fluxional
processes would follow those observed previously and this is
not the focus of the present research.
(iii) Metal complexes of 3
1Ј-(Mesitylthio)ferrocene-1-thiol or its thiolate anion (gener-
ated in situ by reaction of 3 with two equivalents of lithium
triethylborohydride) were treated with a range of transition
metal centres. For example, reaction with bis(benzonitrile)-
D a l t o n T r a n s . , 2 0 0 3 , 3 5 9 9 – 3 6 0 5
3600

View Article Online
measurements obtained via Evans’ method NMR spectro-
scopy10 (with the samples being made up in a N2-filled glove
box) gave good results. The magnetic moments were calculated
to be 3.79 and 2.40 µB for chromium and vanadium,
respectively, and are in reasonable agreement with the spin-only
values (3.87 and 2.87 µB, respectively). The method of calcu-
lation used is an adaption of the one described by Evans10 and
enables calculation of the magnetic susceptibility (χexp) in cgs
units. The values are slightly lower than calculated by spin-only
methods. This is due to spin–orbit coupling which tends to
reduce the value of the magnetic moment for complexes with d1
and d2 electronic configurations. The magnetic susceptibility
measurements by Evans’ method NMR and microanalysis
support a simple chelating structure but a dimeric structure (as
seen for 6 and 7) cannot be conclusively ruled out.
ance of the above cited minor starred peak system, which corre-
sponds to the second oxidation of the diferrocene 1. During
the long periods of controlled potential electrolysis complete
conversion to [1]2ϩ occurs. In confirmation, an authentic sample
of electrogenerated [1]2ϩ affords a voltammetric profile identical
to that shown in Fig. 1b and exhibits a red–brown colour and a
visible spectrum quite coincident with that illustrated for the
species resulting from exhaustive oxidation of 3.
The appearance of two separate one-electron oxidations for
1 (∆EЊЈ = 0.17 V) suggests that the two ferrocenediyl units
are mutually interacting. Assuming that the value of the relative
comproportionation constant (Kcom = 7 × 102) might for the
most part reflect the degree of electronic communication in the
mixed-valent complex [3]ϩ,11 it suggests that the biferrocenium
ion [1]ϩ belongs to the partially delocalised Robin–Day
Class II.12
Considering the metallaferrocenes studied here, Figs. 2a and
b compare the cyclic voltammetric profile of the platinum
complex 4 with that of the corresponding free ligand 2. As seen,
the one electron oxidation of the ferrocene subunit in 4 occurs
at potential values about 0.25 V higher than that of 2. Con-
trolled potential coulometry (Ew = ϩ1.0 V) shows the consump-
tion of one electron per molecule of 4, but cyclic voltammetric
tests on the exhaustively oxidised solution prove that [4]ϩ
partially undergoes decomposition releasing the free ligand 2
(in its oxidised form [2]ϩ).
Electrochemistry
Complexes 2 and 3 in CH2Cl2 solution exhibit common one-
electron oxidation with features of chemical reversibility in
cyclic voltammetry, which is typical of monoferrocene mole-
cules. However, notably different is the behaviour of the two
complexes as far as their exhaustive oxidation is concerned. In
fact, the solution resulting from controlled potential electrolysis
of 2 (Ew = ϩ0.8 V) displays a cyclic voltammetric profile quite
complementary to the original one, which indicates complete
chemical stability of the redox couple 2/[2]ϩ. It is noted that as
a consequence of the exhaustive one electron removal, the
original yellow solution (λmax = 443 nm) does not appreciably
change colour, thus suggesting that the sulfur-based substit-
uents significantly contribute to the HOMO level (iron-centred
ferrocenium species are blue to green coloured). In contrast, as
illustrated in Fig. 1b, the solution resulting from controlled
potential electrolysis of 3 displays a cyclic voltammetric profile
characterised by two reversible reductions.
Fig. 2 Cyclic voltammograms recorded at a platinum electrode in
CH2Cl2 solutions containing [NBu4][PF6] (0.2 mol dmϪ3) and: (a) 4 (0.8
× 10Ϫ3 mol dmϪ3); (b) 2 (0.7 × 10Ϫ3 mol dmϪ3); (c) 11 (1.1 × 10Ϫ3 mol
dmϪ3); (d) 10 (1.0 × 10Ϫ3 mol dmϪ3). Scan rate 0.2 V sϪ1
.
It is interesting to compare the redox behaviour of the couple
2/4 with that of the related symmetric ligand 1,1Ј-bis(mesityl-
thio)ferrocene, 10, and its platinum complex 11 (Scheme 3).8 As
illustrated in Figs. 2c and d, in this case the separation between
the one-electron oxidation of the platinum complex and that of
the free ligand is about 0.6 V. This means that the symmetric
ligand 10 exhibits a higher coordinating ability towards the
PtCl2 fragment than that of the asymmetric ligand 2. Also in
this case, despite the features of chemical reversibility of the
one-electron oxidation of 11, exhaustive oxidation is again
accompanied by partial release of the free ligand 10 (in its
oxidised form [10]ϩ).
Notably different is the behaviour of the “hang-glider like”
cyclometalated platinum complex 14.8 The corresponding
monocation [14]ϩ is quite long-lived indicating that hexa-
coordination renders the platinum complex stable in its oxidised
form. The one-electron removal causes the original light yellow
solution to turn light green, suggesting that the HOMO level
is essentially iron centred. In contrast to the platinum complexes
4 and 11, the corresponding palladium complexes 5 and 12
afford an oxidation process which shows fast release of the
corresponding free ferrocene ligands (i.e. in the backscan, a
new peak coincident with that of the ferrocenium/ferrocene
couple arises). To illustrate this, Fig. 3 shows the voltammetric
response of 12. Finally, Figs. 4a and b show the cyclic voltam-
metric behaviour of the dinuclear platinum and palladium
complexes 6 and 7, respectively. Apart from the presence of
Fig. 1 Cyclic voltammograms recorded at a platinum electrode
in CH2Cl2 solutions containing [NBu4][PF6] (0.2 mol dmϪ3) and: (a) 3
(1.1 × 10Ϫ3 mol dmϪ3), initial; (b) 3 after exhaustive oxidation; (c) 1
(0.7 × 10Ϫ3 mol dmϪ3). Scan rate 0.2 V sϪ1
.
In fact, as shown in Fig. 1a, the main anodic process exhib-
ited by 3 is followed by a minor oxidation (starred peak-
system), which tends to disappear with increased scan rate (at
2.0 V sϪ1 no more traces are detected). This means that [3]ϩ
undergoes a slow chemical reaction to the by-product respon-
sible for the second anodic process. Exhaustive oxidation (Ew =
ϩ0.8 V) consumes about 1.7/1.8 electrons per molecule and
the original yellow solution (λmax = 443 nm) turns red–brown,
displaying a main absorption at 453 nm together with two
further broad absorptions at 670 and 845 nm, respectively.
As deduced from Fig. 1c, the voltammetric profile of the
diferrocene complex 1 is quite complementary to that of the
product resulting from exhaustive two-electron oxidation of 3.
This suggests that oxidation of 3 is accompanied by heterolytic
cleavage of the –S–H function of the monocation [3]ϩ and the
subsequent dimerisation of the thio radical leads to [1]ϩ. Within
the cyclic voltammetric time scale, this is proved by the appear-
D a l t o n T r a n s . , 2 0 0 3 , 3 5 9 9 – 3 6 0 5
3601

View Article Online
Scheme 3 Related compounds 10–14 used in electrochemical studies.
Table 1 Formal electrode potentials (V, vs. SCE) and peak-to-peak
separations (mV) for the ferrocene-centred processes exhibited by
complexes under study in CH2Cl2 solution
a
a
Complex
EЊЈ0/ϩ
∆Ep
EЊЈϩ/2ϩ
∆Ep
3
2
10
1
ϩ0.48
ϩ0.43
ϩ0.45
ϩ0.49
97
103
94
76
ϩ0.66
76
4
5
11
12
14
6
7
13
FcH
ϩ0.69
ϩ1.01b
ϩ1.06
ϩ1.07b
ϩ0.83
ϩ0.83c
ϩ0.75c
ϩ0.31
ϩ0.39
107
–
65
–
Fig. 3 Cyclic voltammogram recorded at a platinum electrode in
CH2Cl2 solution of 12 (0.8 × 10Ϫ3 mol dmϪ3). [NBu4][PF6] (0.2 mol
dmϪ3) supporting electrolyte. Scan rate 0.2 V sϪ1
.
67
d
d
d
ϩ0.94c
ϩ0.97c
ϩ0.57
d
73
60
65
a Measured at 0.1 V sϪ1 b Peak potential value for irreversible processes.
.
c Measured from DPV technique. d Difficult to measure.
mesitylthio group. As illustrated in Fig. 4c, in this case two well
separated ferrocene oxidations are present (∆EЊЈ = 0.26 V), thus
illustrating the delicate electronic effects played by partial
replacement of functional groups in complex molecules and
their unpredictable role on the intramolecular electronic
mobility. In addition, the red–brown mixed-valent monocation
[13]ϩ, which can be tentatively assigned as partially delocalised
(Kcom = 2 × 104), is quite stable.
The formal electrode potentials of the ferrocene centred
processes for all the derivatives studied are compiled in Table 1.
Finally, we note that all the complexes display an irreversible
anodic process in the range from ϩ1.5 to ϩ1.6 V, which is likely
centred on the sulfur substituents, and in addition, all the
platinum and palladium complexes display irreversible reduc-
tion processes in the range from Ϫ1.2 to Ϫ1.6 V, which are
attributed to the corresponding M()/M(0) process.
Fig. 4 Cyclic voltammograms recorded at a platinum electrode in
CH2Cl2 solution of: (a) 6 (0.9 × 0Ϫ3 mol dmϪ3); (b) 7 (1.0 × 10Ϫ3 mol
dmϪ3); 13 (1.1 × 10Ϫ3 mol dmϪ3). [NBu4][PF6] (0.2 mol dmϪ3) supporting
electrolyte. Scan rate 0.02 V sϪ1
.
small amounts of free ligand (starred peaks) two more or less
separated anodic processes are present, which we assign to the
two ferrocene ligands. It is quite evident that the palladium
centres allow a greater mutual interaction of the ferrocene
ligands than platinum centres (∆EЊЈ = 0.22 V vs. ∆EЊЈ = 0.11 V).
Despite the features of chemical reversibility within the cyclic
voltammetric time scale, decomposition occurs upon exhaustive
electrolysis even at the potentials of the first process.
Ethylene polymerisation
It is interesting to compare the behaviour of the platinum
complex 6 with that of the similar complex 13,13 where in the
ferrocene substituents a diphenylphosphino group replaces the
The results of the Schlenk tests for the chromium and
vanadium complexes 8 and 9 are shown in Table 2. It may be
noted that the catalyst structure has been assumed to be a
D a l t o n T r a n s . , 2 0 0 3 , 3 5 9 9 – 3 6 0 5
3602

View Article Online
Table 2 The pre-catalysts and their activities for ethylene polymerisation (L = 1Ј-(mesitylthio)ferrocene-1-thiolate)
Activating
agent (eq.)
Pressure
ethylene/bar
Time of
run/h
Yield of
polymer/g
Procatalyst activity/
Pre-catalyst (mg, mmol)
Solvent/mL
g mmolϪ1 hϪ1 barϪ1
CrL(THF)2Cl2 (25, 0.04)
CrL(THF)2Cl2 (25, 0.04)
CrL(THF)2Cl2 (25, 0.04)
CrL(THF)2Cl2 (25, 0.04)
VL(THF)2Cl2 (25, 0.04)
VL(THF)2Cl2 (25, 0.04)
VL(THF)2Cl2 (25, 0.04)
VL(THF)2Cl2 (25, 0.04)
Toluene (200)
Toluene (200)
Toluene (200)
Toluene (200)
Toluene (200)
Toluene (200)
Toluene (200)
Toluene (200)
MAO (100)
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0.345
0.315
0.010
0.115
0.100
0.156
1.120
0.185
8.6
7.9
0.3
2.9
2.5
3.9
28.0
4.6
Et2AlCl (100)
MeAlCl2 (100)
AlMe3 (100)
MAO (100)
Et2AlCl (100)
MeAlCl2(100)
AlMe3 (100)
mononuclear octahedral complex; if it were to exist in a dimeric
form under polymerisation conditions, the activities would be
almost exactly the same.
6.85 (s, 4H, C6H2); m/z: 734 (M)ϩ, 423 (M Ϫ 2SMes)ϩ, 367 (M
Ϫ S Ϫ SMes)ϩ, 334 (M Ϫ 2S Ϫ SMes)ϩ, 304 (M Ϫ 4S Ϫ Mes)ϩ.
There were no oligomers detected in any of the polymeris-
ations runs. The analysis of the polyethylene was carried out at
BP (Sunbury) and showed that it was all HDPE (high density
polyethylene), giving Mw in the range 330,000–990,000 with
polydispersities in the range 2–4. This is consistent with other
research into CrIII pre-catalysts and indicative of a single site
catalyst.14 The catalysts do not polymerise ethylene in the
absence of an initiator, but when activated show moderate to
low activity for ethylene polymerisation. The most active
catalyst was the vanadium complex, when activated with
1-(Methylthio)-1Ј-(mesitylthio)ferrocene (2)
Compound 1 (0.58 g, 0.79 mmol, 1 eq.) was dissolved in dry,
deoxygenated THF (40 mL). To this solution was added methyl-
lithium (1.6 M solution in diethyl ether, 1.1 mL, 1.74 mmol, 2
eq.) and the reaction stirred for 16 h. The solvent was removed,
the residue dissolved in diethyl ether and water added. The
layers were separated and the aqueous layer extracted (3 × 20
mL, diethyl ether). The combined organic layers were dried
(MgSO4) and evaporated to dryness. Separation of the product
was achieved by column chromatography (neutral grade II
alumina, CH2Cl2–hexane (1 : 4)). The product was eluted as the
first fraction and formed an orange oil on removal of the
solvent (0.15 g, 0.4 mmol, 50%). The only other product was
the starting disulfide, which was recovered in 25% yield.
Anal. Calc. for C20H22S2Fe: C 62.83, H 5.8%. Found: C 62.75,
H 5.82%; δH (CDCl3): 2.24 (s, 3H, CH3), 2.34 (s, 3H, SCH3),
2.51 (s, 6H, CH3), 4.13 (t, 2H, C5H4), 4.26 (m, 4H, C5H4),
4.34 (t, 2H, C5H4), 6.88 (s, 2H, C6H2); m/z: 382 (M)ϩ, 367
(M Ϫ Me)ϩ, 334 (M Ϫ SMe)ϩ, 271 (M Ϫ CpSMe)ϩ, 232
(M Ϫ SMe Ϫ Mes)ϩ.
MeAlCl2, which shows an activity of 28 g mmolϪ1 hϪ1 barϪ1
.
Some caution must be taken when analysing the results as
previous work in the Gibson group has shown that when
vanadium procatalysts are activated with aluminium alkyl
initiators, the ligand is displaced and a vanadium–aluminium
species forms.14
Experimental
All preparations were carried out using standard Schlenk
techniques.15 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
reported in δ using CDCl3 (1H, δ 7.25 ppm) as the reference for
1H 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.16
1Ј-(Mesitylthio)ferrocene-1-thiol (3)
Compound 1 (0.20 g, 0.27 mmol, 1 eq.) was dissolved in dry,
deoxygenated THF (20 mL). To this solution was added lithium
triethylborohydride (1 M solution in THF, 0.54 mL, 0.54 mmol,
2 eq.) and the reaction stirred for 1.5 h. The solvent was
removed in vacuo and the residue dissolved in diethyl ether
(20 mL). To this solution was added, by cannula, nitrogen
saturated 1% KOH solution (20 mL) and the solution was
stirred vigorously. The aqueous layer was removed by cannula
to a dry Schlenk tube under nitrogen and acidified with
nitrogen-saturated concentrated HCl added dropwise until the
pH reached 2 and at this point a precipitate formed. This was
extracted with diethyl ether (3 × 20 mL) and dried (Na2SO4).
The solvent was removed to give an orange solid. This was
recrystallised from pentane at Ϫ25 ЊC to give the pure thiol
(0.13 g, 0.47 mmol, 88%).
Anal. Calc. for C19H20S2Fe: C 61.80, H 5.47%. Found:
C 61.73, H 5.46%; δH (CDCl3): 2.21 (s, 3H, CH3), 2.48 (s, 6H,
2 CH3), 2.98 (s, 1H, SH) 4.11 (t, 2H, C5H4), 4.21 (m, 4H, C5H4),
4.33 (t, 2H, C5H4), 6.86 (s, 2H, C6H2); m/z: 368 (M)ϩ, 367
(M Ϫ H)ϩ, 334 (M Ϫ SH)ϩ.
Di[1Ј-mesitylthioferrocene]disulfide (1)
1,2,3-Trithia-[3]ferrocenophane (1 g, 3.6 mmol, 1 eq.) was
dissolved in THF (20 mL) and cooled to Ϫ78 ЊC. Mesityl-
lithium (2 eq.), generated in situ and dissolved in THF (20 mL),
was added to this by cannula. The reaction was stirred at room
temperature overnight. Water (20 mL) was added and the
layers separated. The aqueous layer was extracted with CH2Cl2
(2 × 20 mL) and the combined organics were dried (MgSO4)
and evaporated to dryness. Column chromatography (neutral
grade II alumina, CH2Cl2–hexane (2 : 8)) enabled separation of
the product (0.40 g, 0.54 mmol, 30%).
Lithium 1Ј-(mesitylthio)ferrocene-1-thiolate
Di[1Ј-(mesitylthio)ferrocene]disulfide (0.10 g, 0.135 mmol,
1 eq.) was dissolved in dry, deoxygenated THF. To this solution
was added lithium triethylborohydride (1 M solution in THF,
0.3 mL, 0.3 mmol, 2.2 eq.) and the reaction stirred for 1.5 h.
The volatile by-products were removed in vacuo and the residue
redissolved in THF (10 mL). The solution was reacted in situ
with a range of metal centres.
Anal. Calc. for C38H38S4Fe2: C 62.13, H 5.21%. Found: C
62.14, H 5.32%; δH (CDCl3): 2.21 (s, 6H, CH3), 2.47 (s, 12H,
CH3), 4.07 (s, 4H, C5H4), 4.21 (s, 4H, C5H4), 4.32 (d, 8H, C5H4),
D a l t o n T r a n s . , 2 0 0 3 , 3 5 9 9 – 3 6 0 5
3603

View Article Online
Dichloro[1-(methylthio)-1Ј-(mesitylthio)ferrocene]platinum(II)
(4)
with hot toluene (2 × 20 mL) and the filtrate was evaporated to
a black oil. The oil was suspended in hot hexane (2 × 40 mL)
causing precipitation of the product as a purple–black powder
(0.11 g, 0.11 mmol, 81%). Recrystallisation from CH2Cl2–
hexane (1:1) resulted in purple crystals suitable for spectroscopy
and microanalysis.
Anal. Calc. for C38H38S4Fe2Pd2Cl2: C 44.82, H 3.76%. Found:
C 44.66, H 3.77%; δH (CDCl3), Ϫ40 ЊC: 2.16 (s, 6H, CH3), 2.41
(s, 6H, CH3), 3.29 (s, 6H, CH3), 4.07 (m, 2H, C5H4), 4.28 (m,
2H, C5H4), 4.44 (m, 4H, C5H4), 4.57 (m, 2H, C5H4), 4.60 (m,
2H, C5H4), 4.66 (m, 2H, C5H4), 5.07 (m, 2H, C5H4), 6.80 (s, 2H,
C6H2), 6.93 (s, 2H, C6H2); m/z: 982 (M Ϫ Cl)ϩ, 937 (M Ϫ Cl Ϫ
3Me)ϩ, 897 (M Ϫ Mes)ϩ, 870 (M Ϫ SMes)ϩ, 754 (M Ϫ SMes Ϫ
Mes)ϩ, 647 (M Ϫ 2SMes Ϫ 2Cl)ϩ.
A solution of 2 (0.15 mmol, 17 mL of a 9 mM stock solution
in toluene, 1.2 eq.) in toluene was injected into a solution of
bis(benzonitrile)dichloroplatinum() (0.06 g, 0.13 mmol, 1 eq.)
dissolved in toluene (20 mL) at 60 ЊC. The solution was stirred
for 20 h at this temperature and the solvent removed in vacuo.
The residue was washed with hexane (30 mL), dissolved in
CH2Cl2 (5 mL) and reprecipitated using hexane (40 mL),
enabling isolation of the product as a yellow powder (0.10 g,
0.12 mmol, 91%).
Anal. Calc. for C20H22S2FePtCl2: C 37.05, H 3.42%. Found:
C 37.46, H 3.62%; δH (CDCl3): 2.22 (s, 3H, CH3), 2.46 (s, 6H,
CH3), 2.67 (s, 3H, SCH3), 4.34 (br m, C5H4), 4.37 (br m, C5H4),
4.44 (br m, C5H4), 5.00 (br m, C5H4) 6.87 (s, 2H, C6H2); m/z: 611
(M Ϫ Cl)ϩ, 577 (M Ϫ 2Cl)ϩ, 382 (L)ϩ.
Bis(tetrahydrofuran)dichloro[1Ј-(mesitylthio)ferrocene-1-
(thiolato)]chromium(III) (8)
Dichloro[1-(methylthio)-1Ј-(mesitylthio)ferrocene]palladium (5)
Trichlorotris(tetrahydrofuran)chromium() (0.24 g, 0.65 mmol,
1 eq.) was dissolved in THF (30 mL) and cooled to Ϫ78 ЊC. To
this solution was added the solution of lithium 1Ј-(mesitylthio)-
ferrocene-1-thiolate (0.65 mmol), resulting in immediate colour
change from purple to yellow. As the solution warmed to room
temperature the colour darkened to a deep red, this colour was
maintained over 16 h stirring at room temperature. The solvent
was removed and the extracts washed with heptane (100 mL)
and extracted with toluene (3 × 30 mL), leaving the precipitated
lithium chloride behind as an off-white solid. The toluene was
removed yielding a deep purple solid (0.20 g, 0.32 mmol, 48%).
Anal. Calc. for C27H35S2O2FeCrCl2: C 50.9, H 5.56%. Found:
C 48.4, H 5.67%; m/z: 456 (M Ϫ Cl Ϫ 2THF)ϩ, 419 (M Ϫ 2Cl Ϫ
2THF)ϩ, 368 (L)ϩ; Magnetic moment by Evans’ method NMR
spectroscopy (CDCl3, cyclohexane): 3.79 µB (theoretical value:
3.88 µB).
A solution of 2 (0.15 mmol, 17 mL of a 9 mM stock solution
in toluene, 1.2 eq.) in toluene was injected into a solution
of bis(benzonitrile)dichloropalladium() (0.05 g, 0.13 mmol,
1 eq.) dissolved in toluene (20 mL) at 60 ЊC. The solution was
stirred for 20 h at this temperature and the solvent removed
in vacuo. The residue was washed with hexane (30 mL),
dissolved in CH2Cl2 (5 mL) and reprecipitated using hexane
(40 mL), enabling isolation of the product as a purple powder
(0.06 g, 0.11 mmol, 85%).
Anal. Calc. for C20H22S2FePdCl2: C 42.92, H 3.96%. Found:
C 43.02, H 3.85%; δH (CDCl3): 2.22 (s, 3H, CH3), 2.44 (s, 6H,
CH3), 2.65 (s, 3H, SCH3), 4.28 (t, 2H, C5H4), 4.33 (t, 2H, C5H4),
4.39 (t, 2H, C5H4), 4.67 (br m, C5H4), 6.87 (s, 2H, C6H2); m/z:
525 (M Ϫ Cl)ϩ, 487 (M Ϫ 2Cl)ϩ, 382 (L)ϩ.
Bis[1Ј-(mesitylthio)ferrocene-1-(ꢀ-thiolato)]bisplatinum(II)-
dichloride (6)
Bis(tetrahydrofuran)dichloro[1Ј-(mesitylthio)ferrocene-1-
(thiolato)]vanadium(III) (9)
Lithium 1Ј-(mesitylthio)ferrocene-1-thiolate (0.27 mmol) was
dissolved in toluene (30 mL). Bis(benzonitrile)dichloro-
platinum() (0.13 g, 0.27 mmol, 1 eq.) was also dissolved in
toluene at 60 ЊC, giving a pale yellow solution. To this solution
was added the toluene solution of lithium 1Ј-(mesitylthio)ferro-
cene-1-thiolate, resulting in an immediate darkening of the
solution to deep red–orange. The solution was stirred at 60 ЊC
for 16 h. An off-white precipitate was deposited, which was
filtered off and washed with hot toluene (2 × 20 mL) and the
filtrate evaporated to a brown oil. The oil was suspended in hot
hexane (2 × 40 mL) enabling precipitation of the product as a
pale brown yellow powder (0.15 g, 0.13 mmol, 93%). Recrystal-
lisation from CH2Cl2–hexane (1:1) resulted in yellow crystals
suitable for spectroscopy and microanalysis.
Anal. Calc. for C38H38S4Fe2Pt2Cl2: C 38.17, H 3.2%. Found:
C 38.18, H 3.1%; δH (CDCl3), Ϫ30 ЊC: 2.19 (s, 6H, CH3), 2.45
(s, 6H, CH3), 3.32 (s, 6H, CH3), 4.11 (m, 2H, C5H4), 4.25
(m, 2H, C5H4), 4.42 (m, 2H, C5H4), 4.46 (m, 2H, C5H4), 4.55
(m, 2H, C5H4), 4.60 (m, 2H, C5H4), 4.67 (m, 2H, C5H4), 5.20
(m, 2H, C5H4), 6.80 (s, 2H, C6H2), 6.93 (s, 2H, C6H2); m/z: 1196
(M)ϩ, 1160 (M Ϫ Cl)ϩ, 1125 (M Ϫ 2Cl)ϩ, 1061 (M Ϫ Mes Ϫ
Me)ϩ, 821 (M Ϫ 2SMes Ϫ 2Cl)ϩ.
Trichlorotris(tetrahydrofuran)vanadium() (0.24 g, 0.65 mmol,
1 eq.) was dissolved in THF (30 mL) and cooled to Ϫ78 ЊC. To
this solution was added the solution of lithium 1Ј-(mesitylthio)-
ferrocene-1-thiolate (0.65 mmol), resulting in immediate colour
change form pink to red. As the solution warmed to room
temperature the colour darkened to purple, this colour was
maintained over 16 h stirring at room temperature. The solvent
was removed and the extracts washed with heptane (3 × 30 mL)
and extracted into toluene (3 × 30 mL), leaving the precipitated
lithium chloride behind as an off white solid. The toluene was
removed yielding a deep purple–black solid (0.18 g, 0.28 mmol,
43%).
Anal. Calc. for C27H35S2O2FeVCl2: C 51.26, H 5.58%. Found:
C 50.89, H 5.86%; Magnetic moment by Evans’ method NMR
spectroscopy (CDCl3, cyclohexane): 2.40 µB (theoretical value:
2.82 µB).
Ethylene polymerisations
All polymerisations were carried out according to the following
procedure. The pre-catalyst (0.04 mmol) was thoroughly dried
(drying piston overnight or in a Schlenk tube on a vacuum line
if the compound is air sensitive) and added to a clean, dry
Schlenk tube, charged with a stirrer bar, under nitrogen. The
pre-catalyst was suspended in dry, deoxygenated toluene (200
mL) and to this suspension was added methylaluminoxane
(10% solution in toluene, 0.4 mmol, 2.66 mL). The addition of
methylaluminoxane caused complete dissolution of all species
and often a colour change. The Schlenk tube was then trans-
ferred to the ethylene line (p = 1 bar), where it was flushed
through with ethylene several times and then stirred at the
fastest rate for 1 h. After this time acidified methanol (200 mL)
was added and effervescence was observed. The layers were
Bis[1Ј-(mesitylthio)ferrocene-1-(ꢀ-thiolato)]bispalladium(II)
dichloride (7)
Lithium 1Ј-(mesitylthio)ferrocene-1-thiolate (0.27 mmol) was
dissolved in toluene (30 mL). Bis(benzonitrile)dichloro-
palladium() (0.10 g, 0.27 mmol, 1 eq.) was dissolved in toluene
at 60 ЊC, giving a deep red solution. To this solution was added
the toluene solution of lithium 1Ј-(mesitylthio)ferrocene-1-
thiolate, resulting in immediate darkening of the solution to
dark purple. The solution was stirred at 60 ЊC for 16 h. A black
precipitate was deposited, which was filtered off and washed
D a l t o n T r a n s . , 2 0 0 3 , 3 5 9 9 – 3 6 0 5
3604

View Article Online
1997, 4705; U. Siemeling, U. Vorfeld, B. Neumann, H. G. Stammler,
separated and the organic layer dried (MgSO4) and a small
sample submitted for GC-MS. The remaining solvent was
P. Zanello and F. Fabrizi de Biani, Eur. J. Inorg. Chem., 1999, 38, 1;
S. Bradley, P. C. McGowan, K. A. Oughton, M. Thornton-Pett and
M. E. Walsh, Chem. Commun., 1999, 77; S. Bradley, K. D. Camm,
X. Liu, P. C. McGowan, R. Mumtaz, K. A. Oughton, T. J. Podesta
and M. Thornton-Pett, Inorg. Chem., 2002, 41, 715; J. D. Carr,
S. J. Coles, M. B. Hursthouse, M. E. Wright, E. L. Munro, J. H. R.
Tucker and J. Westwood, Organometallics, 2000, 19, 3312; J. D. Carr,
S. J. Coles, M. B. Hursthouse and J. H. R. Tucker, J. Organomet.
Chem., 2001, 637–639, 304; B. Wrackmeyer, H. E. Maisel and
M. Herberhold, J. Organomet. Chem., 2001, 637–639, 727;
N. J. Long, J. Martin, G. Opromolla, A. J. P. White, D. J. Williams
and P. Zanello, J. Chem. Soc., Dalton Trans., 1999, 1981; V. C.
Gibson, N. J. Long, A. J. P. White, C. K. Williams, D. J. Williams,
M. Fontani and P. Zanello, J. Chem. Soc., Dalton Trans., 2002, 3280.
4 V. C. Gibson, N. J. Long, A. J. P. White, C. K. Williams and
D. J. Williams, Chem. Commun., 2000, 2359.
1
removed and the sample analysed by H, 31P{1H} NMR and
FAB mass spectrometry.
Acknowledgements
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.
P. Z. gratefully acknowledges the financial support of the
University of Siena (PAR 2003).
5 M. Herberhold, O. Nuyken and T. Pohlmann, J. Organomet. Chem.,
1991, 405, 217.
6 V. C. Gibson, N. J. Long, R. J. Long, A. J. P. White, C. K. Williams
and D. J. Williams, Organometallics., 2003, submitted.
7 E. W. Abel, N. J. Long, K. G. Orrell, A. G. Osborne and V. Sik,
J. Organomet. Chem., 1989, 378, 473; E. W. Abel, N. J. Long,
K. G. Orrell, A. G. Osborne, V. Sik, P. A. Bates and M. B.
Hursthouse, J. Organomet. Chem., 1990, 394, 455; E. W. Abel,
N. J. Long, K. G. Orrell, A. G. Osborne and V. Sik, J. Organomet.
Chem., 1991, 405, 375; B. McCulloch, D. L. Ward, J. D. Woollins
and C. H. Brubaker, Jr., Organometallics, 1985, 4, 1425.
References
1 For a detailed literature review, see: Ferrocenes: Homogeneous
Catalysis – Organic Synthesis – Materials Science, ed. A. Togni and
T. Hayashi, VCH, Weinheim, Germany, 1995; Metallocenes, ed.
A. Togni and R. L. Haltermann, Wiley-VCH, Weinheim, Germany,
1998.
2 For a comprehensive overview of ferrocene and other metallocene
chemistry, see: N. J. Long, Metallocenes: An Introduction to
Sandwich Complexes, Blackwell Science, Oxford, 1998.
3 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; I. R. Butler and S. C. Quayle, J. Organomet. Chem.,
1998, 552, 63; 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; I. R. Butler, S. Mussig and M. Plath, Inorg.
Chem. Commun., 1999, 2, 424; W.-P. Deng, X.-L. Hou, L.-X. Dai,
Y.-H. Yu and W. Xia, Chem. Commun., 2000, 285; W. P. Deng,
X. L. Hou, L. X. Dai and X.-W. Dong, Chem. Commun., 2000, 1483;
T. Y. Dong and C.-K. Chang, J. Chin. Chem. Soc., 1998, 45, 577;
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; R. Broussier, S. Ninoreille, C. Bouron,
O. Blacque, C. Ninoreille, M. M. Kubicki and B. Gautheron,
J. Organomet. Chem., 1998, 561, 85; G. A. Balavoine, J.-C. Daran,
G. Iftiem, E. Manoury and C. Moveau-Bossuet, J. Organomet.
Chem., 1998, 567, 191; J. A. Adeleke, Y.-W. Chen and L.-K. Liu,
Organometallics, 1992, 11, 2543; J. A. Adeleke and L.-K. Liu,
J. Chin. Chem. Soc., 1992, 39, 61; J. Priego, O. Garcia Mancheno,
S. Cabrera, R. Gomez Arraya, T. Llamas and J. C. Carretero, Chem.
Commun., 2002, 2512; H. Seo, H.-J. Park, B. Yun Kim, J. Hoon Lee,
S. Ukk Son and Y. Keun Chung, Organometallics, 2003, 22, 618;
B. Neumann, U. Siemeling, H. G. Stammler, U. Vorfeld, J. G. P.
Delis, P. W. N. M. van Leeuwen, K. Vrieze, J. Fraanje, K. Goubitz,
F. Fabrizi de Biani and P. Zanello, J. Chem. Soc., Dalton Trans.,
8 V. C. Gibson, N. J. Long, A. J. P. White, C. K. Williams and
D. J. Williams, Organometallics, 2000, 19, 4425.
9 N. Brugat, A. Polo, A. Alvarez-Larena, J. F. Piniella and J. Real,
Inorg. Chem., 1999, 38, 4829; E. Hauptman, P. J. Fagan and
W. Marshall, Organometallics, 1999, 18, 2061; R. Cao, M. Hong,
F. Jiang, B. Kang, X. Xie and H. Liu, Polyhedron, 1996, 15, 2661.
10 D. F. Evans, J. Chem. Soc., 1959, 2003.
11 (a) V. Palaniappan, R. M. Singru and U. C. Agarwala, Inorg. Chem.,
1988, 27, 181; (b) D. Astruc, Electron Transfer and Radical Processes
in Transition-Metal Chemistry, Wiley-VCH, New York, 1995; (c)
B. S. Brunschwig and N. Sutin, Coord. Chem. Rev., 1999, 187, 233;
(d ) K. D. Demadis, C. M. Hartshorn and T. J. Meyer, Chem. Rev.,
2001, 101, 2655; (e) F. Barriere, N. Camire, U. T. Mueller-Westerhoff
and R. Sanders, J. Am. Chem. Soc., 2002, 124, 7262.
12 M. B. Robin and P. Day, Adv. Inorg. Chem. Radiochem., 1967, 10,
247.
13 V. C. Gibson, N. J. Long, A. J. P. White, C. K. Williams and
D. J. Williams, Organometallics, 2002, 21, 770.
14 K. Chang, PhD Thesis, 2001, Imperial College, London.
15 R. J. Errington, Advanced Practical Inorganic and Metalorganic
Chemistry, Blackie, London, 1997.
16 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.
D a l t o n T r a n s . , 2 0 0 3 , 3 5 9 9 – 3 6 0 5
3605