Oxidatively induced M-C bond cleavage reactions of Cp*Ir(Me2SO)Me2 and Cp*Rh(Me2SO)Me2 (Cp* = η5-C5Me5)

Article abstract of DOI:10.1039/b107451m

Electrochemical and chemical oxidation of Cp*Ir(Me₂SO)Me₂ (1; Cp* = η⁵-C₅Me₅) in Me₂SO lead to the formation of methane and traces of ethene (ratio ca. 20:1) as well as Cp*Ir(Me₂SO)₂Me⁺. Labeling studies indicate that three of the methane hydrogen atoms arise from the iridium-bonded methyl, while the fourth appears to arise from adventitious water. By contrast, oxidation of the rhodium analog Cp*Rh(Me₂SO)Me₂ gave only ethane and Cp*Rh(Me₂SO)₂Me⁺. Quantitative derivative cyclic voltammetry (DCV) showed that the disappearance of electrode-generated 1^·+ follows a rate law that is second order with respect to Ir. The Rh analog 6^·+ reacted at rates too great to be measured by DCV under these conditions. Possible mechanisms for these reactions are discussed, and particular attention is paid to the reactivity differences between Rh and Ir. Convenient new syntheses of Cp*Ir(Me₂SO)Me₂ and Cp*Rh(Me₂SO)Me₂, as well as X-ray structure determinations of these compounds and of Cp*Ir(Me₂SO)₂Me⁺, are reported.

Full text of DOI:10.1039/b107451m

View Article Online / Journal Homepage / Table of Contents for this issue
Oxidatively induced M–C bond cleavage reactions of
Cp*Ir(Me2SO)Me2 and Cp*Rh(Me2SO)Me2 (Cp* ؍ ꢀ5-C5Me5)†
Erik Fooladi,a Todd Graham,b Michael L. Turner,b Bjørn Dalhus,a Peter M. Maitlis*b and
Mats Tilset*a
a Department of Chemistry, University of Oslo, P.O. Box 1033 Blindern, N-0315 Oslo, Norway.
E-mail: mats.tilset@kjemi.uio.no
b Department of Chemistry, University of Sheffield, Sheffield, UK S3 7HF.
E-mail: P. Maitlis@sheffield.ac.uk
Received 16th August 2001, Accepted 14th November 2001
First published as an Advance Article on the web 13th February 2002
Electrochemical and chemical oxidation of Cp*Ir(Me2SO)Me2 (1; Cp* = η5-C5Me5) in Me2SO lead to the formation
of methane and traces of ethene (ratio ca. 20 : 1) as well as Cp*Ir(Me2SO)2Meϩ. Labeling studies indicate that three
of the methane hydrogen atoms arise from the iridium-bonded methyl, while the fourth appears to arise from
adventitious water. By contrast, oxidation of the rhodium analog Cp*Rh(Me2SO)Me2 gave only ethane and
Cp*Rh(Me2SO)2Meϩ. Quantitative derivative cyclic voltammetry (DCV) showed that the disappearance of electrode-
ϩ
generated 1 follows a rate law that is second order with respect to Ir. The Rh analog 6 ϩ reacted at rates too great to
be measured by DCV under these conditions. Possible mechanisms for these reactions are discussed, and particular
attention is paid to the reactivity differences between Rh and Ir. Convenient new syntheses of Cp*Ir(Me2SO)Me2 and
Cp*Rh(Me2SO)Me2, as well as X-ray structure determinations of these compounds and of Cp*Ir(Me2SO)2Meϩ, are
reported.
ؒ
ؒ
kinetics and mechanism for the reductive elimination of ethane
Introduction
from the cation radical of Cp*Rh(PPh3)Me2 in acetonitrile.7 A
Reductive elimination is a fundamentally important process
of organometallic reactivity and is often the mechanism by
which organic products are liberated from the metal in a
catalytic process. Simple intuitive rules of thumb imply that
when comparisons are made between related compounds,
reductive elimination becomes more facile when the metal
center is more electron deficient, for example through increased
first-order process resulted in intramolecular ethane formation
according to eqn. (2). Solvent coordination and further oxid-
2ϩ
ation yielded the solvento complex Cp*Rh(PPh3)(NCMe)2
as the first observed Rh-containing product after an overall
two-electron process. In a separate experiment, it was shown
2ϩ
that Cp*Rh(PPh3)(NCMe)2 reacts with the substrate to give
the comproportionation product Cp*Rh(PPh3)(NCMe)Meϩ,
charge and/or formal oxidation state.1 Accordingly, it has been
but this reaction was orders of magnitude slower than the
frequently demonstrated that the electron-transfer oxidation
of organotransition-metal complexes containing σ-bonded
initial reductive elimination of ethane from Cp*Rh(PPh3)-
ϩ 3f
7
ؒ
Me2
.
The oxidation of Cn*RhMe3 also resulted in uni-
ligands may induce reductive elimination reactions. Several
distinct mechanisms have been observed for these eliminations,
defined in a broad sense as shown in eqns. (1)–(3). Homolytic2
[eqn. (1)] and concerted reductive elimination [eqn. (2)]
processes2–4 are both found for compounds containing σ-
bonded hydrocarbyl ligands. The heterolytic process shown in
eqn. (3) is commonly seen for metal hydrides.5 The facile
elimination reactions may be viewed as a consequence of the
metal–X bond weakening caused by the one-electron oxidation
process.6
molecular ethane elimination.3j Similarly, oxidation of the
stable hydridomethyl complex Cp*Ir(PPh3)(H)Me initiated the
elimination of methane. In contrast, oxidation of the dimethyl
complex Cp*Ir(PPh3)Me2 resulted in a relatively stable cation
radical3g,8 which eventually reacted to give Cp*Ir(PPh3)(NC-
Me)Meϩ and CH4 in acetonitrile. The fourth hydrogen atom
in CH4 was derived from the Cp* ligand.3g Others have also
demonstrated that Ir-mediated C–H bond activation occurs at
the Cp* ligand of Cp*Ir(PPh3)Me2 ϩ.8 Reductive elimination of
ؒ
ethane occurred only to a minor extent, and these results point
at potentially fundamental differences in the reactivities of Rh
and Ir complexes.
We have previously reported3f a detailed investigation of the
In this contribution we report the results of an investigation
of the oxidation of Cp*Ir(Me2SO)Me2 (1) and its Rh analog
Cp*Rh(Me2SO)Me2 (6). The Ir compound 1 was first reported
by Maitlis and co-workers in 1983 and has since then been
shown to exhibit a rich and varied reaction chemistry, including
exchange of the Me2SO ligand with other two-electron donors
and C–H activation of certain organic compounds.9a–f The
Rh complex 6 has also been briefly examined.9g In order
to shed some further light on reactivity differences between
second- and third-row transition metals, we compare the
behavior of Cp*Ir(Me2SO)Me2 and the Rh analog, as well as
some previously studied related complexes.
† Dedicated to the memory of John Osborn (1939–2000), outstanding
inorganic chemist, superb companion, and generous friend.
Electronic supplementary information (ESI) available: rotatable 3-D
crystal structure diagrams in CHIME format. See http://www.rsc.org/
suppdata/dt/b1/b107451m/
DOI: 10.1039/b107451m
J. Chem. Soc., Dalton Trans., 2002, 975–982
This journal is © The Royal Society of Chemistry 2002
975

View Article Online
mechanism of the reaction of 1 ϩ in Me2SO by derivative cyclic
voltammetry (DCV). The technique has been described in
detail in reviews, and applications of DCV for the investigation
of the reactivity of electrode-generated transient organometal-
lic complexes have been reported.3f,h,j,5e,14 A DCV reaction-order
ؒ
Results
Electrochemical oxidation of Cp*Ir(Me2SO)Me2 (1) in Me2SO
The oxidation of 1 was investigated by cyclic voltammetry
(CV). Fig. 1 shows a cyclic voltammogram recorded in Me2SO/
Ϫ
analysis of the oxidation of 1 in Me2SO/0.1 M Bu4NϩPF6 at
25 ЊC established that the homogeneous reaction that consumes
ϩ
ؒ
1
was cleanly second order over the substrate concentration
ϩ
ؒ
range 0.5–2.0 mM. The rate law for consumption of 1 can
then be second-order with respect to 1 ϩ, or first-order with
ؒ
ϩ
ؒ
respect to each of 1 and 1. When the solvent/electrolyte was
changed to dichloromethane/0.2 M Bu4NϩPF6Ϫ, the reaction
rate did not change significantly when compared with the rate
using Me2SO. CV in acetonitrile led to similar results, but the
analysis of the results was complicated due to partial exchange
of MeCN for coordinated Me2SO (vide infra). We conclude
that there was essentially no solvent effect on the rate of
the reaction of 1 ϩ under these reaction conditions on the time
ؒ
scale (seconds or less) that applies to the CV measurements.
Variable-temperature DCV measurements in the temperature
range 18–50 ЊC revealed an Arrhenius activation energy near
zero (Ea = Ϫ7
4 kJ molϪ1).15 On the assumption that the
Fig. 1 Cyclic voltammogram of Cp*Ir(Me2SO)Me2 (1.0 mM) in
Me2SO/0.1 M Bu4NϩPF6Ϫ at a Pt disk electrode (d = 0.6 mm) at 20 ЊC at
one-electron oxidation was followed by a second-order process
with respect to 1 (EC dimerization mechanism, Scheme 1),
ϩ
ؒ
voltage scan rate ν = 1.0 V sϪ1
.
0.1 M Bu4NϩPF6 (Pt disk electrode, d = 0.6 mm, 20 ЊC) at a
Ϫ
voltage scan rate ν of 1.0 V sϪ1
.
As can be seen in the voltammogram, the electrode reaction
is partially chemically reversible on the measurement time
scale. Voltammograms at variable scan rates revealed that
the electrode process approaches chemical reversibility at high
scan rates (> 100 V sϪ1) and is essentially irreversible at low scan
rates (< 0.1 V sϪ1), indicating that subsequent homogeneous
Scheme 1
the DCV data were converted to rate constants by comparison
with theoretical working curves. The Eyring activation param-
ϩ
ؒ
eters for the bimolecular reaction of 1 were then obtained:
ϩ
ؒ
∆H‡ = Ϫ10 4 kJ molϪ1, ∆S‡ = Ϫ191 10 J KϪ1 molϪ1, and
reactions consume electrode-generated 1 on the CV time
ϩ
ؒ
k(25 ЊC) = 3.2 × 104 MϪ1 Ϫ1. Analogous variable-temperature
s
scale. The reversible electrode potential for the 1/1 couple
[EЊox(1)], taken as the midpoint between the anodic and
cathodic peaks at high scan rates, was measured to be ϩ0.18 V
vs. the Cp2Fe/Cp2Feϩ (Fc) couple. The peak-to-peak separation
(∆Ep) of about 80 mV at ν = 1.0 V sϪ1 is somewhat greater than
for the Fc couple (∆Ep = 63 mV under identical conditions
and with equal concentrations), indicative of a quasi-reversible
electron transfer step. At low scan rates, adsorption at the elec-
trode surface was evident through clear deterioration of the
electrode response after repeated cycling. However, with the
relatively fast scan rates used for the kinetic measurements
described below, no signs of adsorption were discernible. The
observed ∆Ep, coulometry data (see below), and the linear
dependence of ip vs. ν1/2 at ν = 0.4–20 V sϪ1 10 are indicative of an
overall one-electron transfer process on the CV time scale.11
Constant-potential coulometry measurements for the oxida-
tion of 1 were carried out at working electrode potentials 0.10–
DCV measurements were also performed for the bis(deuterio-
methyl) analog 1-d6. Comparison of the kinetic data from these
experiments resulted in a kinetic isotope effect kH/kD = 1.6–2.2
(the magnitude was randomly distributed over the temperature
range 18–50 ЊC). This finding is indicative of a C–H bond
cleavage in a coordinated methyl group before or during the
rate-limiting step.
Chemical oxidation of Cp*Ir(Me2SO)Me2 in Me2SO
When 1 was oxidized in Me2SO using Cp2FeϩBF4Ϫ, the only
observable Ir-containing product was again 2(BF4Ϫ). A GC
analysis of the vapor phase revealed the formation of methane
and ethene in an approximate ratio of at least 20 : 1 at substrate/
oxidant concentrations of 10 mM/10 mM as well as 70 mM/
43 mM in Me2SO, along with traces of ethane (< 0.5%). Since
the oxidation potential of 1 is 0.18 V positive of the Fc couple,
the Nernst equation implies a very low equilibrium concentra-
0.14 V positive of EЊox(1) on 1.3–1.4 mM substrate solutions
Ϫ 12
in Me2SO/0.05 M Me4NϩBF4
. The measurements indicated
tion of 1 ϩ, which could serve effectively to suppress second-
order processes if they were competing with first-order ones.
ؒ
the consumption of approximately 1 Faraday molϪ1 for the
complete consumption of 1 (nobs = 0.9–1.4 F molϪ1).13 Addition
of water (0.2% v/v) to the solution caused no significant dif-
ference in the electron transfer stoichiometry (see below). A GC
analysis of the vapor phase above the solution after exhaustive
electrolysis showed the formation of methane and ethene in
an approximate molar ratio of 9 : 1. The only organometallic
product that could be detected after work-up was Cp*Ir(Me2-
ϩ
ؒ
An effective increase in the concentration of 1 was therefore
attempted by using the considerably stronger oxidant (C5H5)-
(C5H4COMe)FeϩBF4Ϫ (EЊ = ϩ0.25 vs. Fc16). This resulted in a
much less selective process, with a mixture of largely unidenti-
fied Ir-containing products. This may indicate that the stronger
oxidant is capable of intercepting and oxidizing short-lived
intermediates that are not detected by CV, and that are not
intercepted by Cp2Feϩ. These results will not be elaborated
further.
Ϫ
SO)2MeϩBF4 [2(BF4Ϫ)], identified by comparison of its 1H
NMR spectrum with that of an independently prepared sample
(vide infra). The product was isolated as a light yellow, air stable
solid in greater than 80% yield.
Isotope labeling studies
Qualitatively, it was seen that the extent of chemical reversi-
bility of the CV response was dependent on the concentration
of 1, which indicated that the composite reaction-order with
An exhaustive electrolysis was conducted using a 1 : 1 mixture
of 1 and Cp*Ir(Me2SO)(CD3)2 (1-d6) followed by GC–MS
analysis of the vapor phase. This revealed mainly the formation
of CH4 and CHD3 in a ca. 1 : 1 ratio. Determination of the
ϩ
ؒ
respect to 1 and 1 was greater than unity. This was quanti-
fied through an extensive investigation of the kinetics and
976
J. Chem. Soc., Dalton Trans., 2002, 975–982

View Article Online
ethene isotopomer distribution is subject to uncertainties due
to the small quantity of the analyte, but is dominated by C2H4
(53–98%) and some C2D4 (1–40%).17
the relative amounts of 3 and 4 changed and signals possibly
corresponding to 2 appeared at δ 0.83 (Ir–Me) and 3.21 (co-
ordinated Me2SO). When the solvent was removed by evapora-
tion (leading to a gradual increase of the Me2SO concentration)
and the residue was dissolved in CD2Cl2, the 1H NMR analysis
revealed 2 as the major species present. These observations
suggest that 2, 3, 4, Me2SO, and MeCN are involved in a
solvent-dependent equilibrium. Of the three products, we have
been able to produce only 2 as a pure complex, since 3 and 4 will
always exist as an equilibrium mixture. Removal of the solvent
by evaporation ultimately leads to a gradual enrichment of
2 and 3 at the expense of 4 since the concentration of Me2SO
increases as acetonitrile is first removed. It is interesting to
note that acetonitrile-d3 solutions of 1 show no evidence for
exchange of acetonitrile-d3 for Me2SO by 1H NMR analysis.
A crossover experiment using a 1 : 1 mixture of 1 and 1-d6,
Ϫ
and the Cp2FeϩBF4 oxidant resulted in essentially the same
gaseous products as in the electrode oxidation. Thus, CH4 and
CHD3 were produced in an approximately 1 : 1 ratio, and C2H4
was obtained as the major ethene isotopomer (64–91%; C2D4
7–34%; C2H3D 0–13%; C2H2D2 0–1%; C2HD3 0–4%) on the
basis of a GC–MS analysis. In an attempt at tracing the origin
of the hydrogen atoms in the gaseous products, oxidation of
1 was conducted in (CD3)2SO. This resulted in the exclusive
formation of CH4 and C2H4 by GC–MS, establishing that the
fourth hydrogen atom in the methane product is not derived
from the solvent. A 2H{1H} NMR analysis of the organo-
metallic products demonstrated the anticipated exchange of
(CD3)2SO for coordinated (CH3)2SO when (CD3)2SO is used
as solvent. Importantly, the analyses showed no evidence of
deuterium incorporation into coordinated Cp* or methyl
groups, contrasting with the results of the oxidation of
Cp*Ir(PPh3)Me2 in which activation of C–H bonds in the
Cp* ligand occurred. We also considered the possibility that
adventitious water in the Me2SO solvent might be involved.
Ϫ
Exhaustive electrolysis of 1 in acetonitrile/Bu4NϩPF6 by
constant potential coulometry at E = 0.11–0.12 V vs. EЊ(1) indi-
cated a one-electron oxidation of 1 (nobs = 0.92 0.06 F molϪ1).
An analysis of the organometallic products by 1H NMR
spectroscopy showed a product mixture quite similar to that
obtained from the chemical oxidation. Furthermore, as
previously mentioned, the DCV reaction-order analysis in
Ϫ
acetonitrile/0.1 M Bu4NϩPF6 established a second-order pro-
Ϫ
ϩ
This was probed by the oxidation of 1 with Cp2FeϩBF4 in
cess with respect to 1 (or 1 ϩ and 1 combined). The DCV data
gave an Arrhenius activation energy close to the value measured
in Me2SO, and a decrease in the reaction rate to approximately
1/3–1/4 of the rate observed in Me2SO (Ea = Ϫ4 4 and Ϫ8
2 kJ molϪ1 in two separate experiments15). The similarities in
behavior and kinetic parameters in the two solvents Me2SO
ؒ
ؒ
(CD3)2SO to which was added 1% (v/v) of D2O. The GC–MS
analysis of the vapor phase in this case showed ca. 67% CH4
and 33% CH3D. A 1H NMR analysis of the solvent/water
mixture showed significant amounts of non-deuterated water;
if all water had been present as D2O, the CH3D content would
presumably have been even greater than 33%.18 Consistently,
the complementary experiment, oxidation of Cp*Ir[(CD3)2SO]-
ϩ
ؒ
and MeCN suggest that 1 reacts by the same, or by very
similar, mechanisms in the two solvents. The complex mixture
obtained in acetonitrile would then be a result of subsequent
reactions. Alternatively, the complexity could arise from facile
(CD3)2 with Cp2FeϩBF4 in Me2SO to which was added 0.5%
Ϫ
2
H2O, showed predominantly the formation of CHD3. H{1H}
1
ϩ
ؒ
and H NMR analysis of the organometallic products in these
substitution of MeCN for Me2SO in 1 (it has been firmly
established19 that very rapid ligand substitution reactions can
occur by associative mechanisms at 17-electron radicals).
experiments showed, as in previous experiments, no sign of
H/D scrambling into coordinated Cp* or methyl groups. This
series of experiments strongly imply water—intentionally
added or as a residual impurity in the Me2SO solvent—as the
origin of the fourth hydrogen atom in much, if not all of,
the methane product.
1
Ϫ
A H NMR analysis of the Cp2FeϩBF4 oxidation of 1 in
dichloromethane-d2 at temperatures below Ϫ5 ЊC revealed a
complicated spectrum exhibiting at least three major reson-
ances in the Cp* region (δ 1.55–1.80) in addition to those of 1,
and a singlet at δ 12.3. There were no obvious signs of para-
magnetic species. Inverse 1H–13C correlated NMR spectroscopy
(HSQC) analysis showed a weak interaction between the pro-
ton signal at δ 12.3 and a carbon atom resonating at δ 212 (this
signal could only be seen because of the enhanced sensitivity
Chemical and electrochemical oxidation of Cp*Ir(Me2SO)Me2
in acetonitrile and dichloromethane
The oxidation of 1 with Cp2FeϩBF4Ϫ in acetonitrile gave mostly
methane, less than 30% ethane, small quantities (5%) of ethene,
free Me2SO, and a mixture of three Ir-containing compounds.
The 1H NMR spectroscopic data of the products are attributed
to the compounds Cp*Ir(Me2SO)2Meϩ (2), Cp*Ir(Me2SO)-
(NCMe)Meϩ (3), and Cp*Ir(NCMe)2Meϩ (4). Compound
2(BF4Ϫ) was isolated as an air-stable, crystalline material after
protonolysis of 1 with HBF4ؒEt2O in Me2SO (see the Experi-
mental section). The X-ray crystal structure of 2(BArfϪ)7 has
1
caused by the observation of the H signal), thus pointing to
a possible carbenic intermediate (see discussion below). The
signal at δ 12.3 was not observed when trace amounts of Me2SO
(less than 1% v/v) were added prior to the reaction.
Oxidation of Cp*Rh(Me2SO)Me2 (6) in Me2SO
Oxidation of the Rh compound Cp*Rh(Me2SO)Me2 (6) with
1
Ϫ
also been determined (vide infra). The cation 2 displayed H
one equivalent of Cp2FeϩBF4 led to the exclusive formation
NMR signals (CD2Cl2) at δ 3.35 and 3.36 (diastereotopic
methyl groups in each S-coordinated Me2SO), 1.83 (Cp*), and
0.96 (Ir–Me). Compounds 3 and 4 were observed in situ, but not
isolated. The 1H NMR spectrum of 3 (acetonitrile-d3) displayed
signals at δ 0.57 (Ir–Me), 1.75 (Cp*), 3.05 and 3.21 (diastereo-
topic methyls of S-coordinated Me2SO); coordinated MeCN
was not observed due to exchange with the solvent. For 4, the
1H NMR spectrum exhibited resonances at δ 0.68 (Ir–Me) and
1.75 (Cp*); again, coordinated MeCN was not seen due to
exchange with the solvent. Coordinated MeCN was however,
of ethane and Cp*Rh(Me2SO)2Meϩ (7) as judged from GC
and 1H NMR analyses. This result appears to be in accord with
the behavior of the previously investigated Rh complexes
Cp*Rh(PPh3)Me2 3f and Cn*RhMe3.3j
In both these cases it was shown that the reductive elimin-
ation of ethane was intramolecular, presumably yielding the
ϩ
intermediates Cp*Rh(PPh3) and Cn*RhMe ϩ, respectively.
ؒ
ؒ
Further oxidation and either solvent coordination or compro-
2ϩ
portionation would ultimately yield Cp*Rh(PPh3)(NCMe)2
ϩ
and a mixture of Cn*Rh(NCMe)Me2 and Cn*Rh(NCMe)2-
1
seen in the H NMR spectrum of 3 when the oxidation was
Me2ϩ respectively, which were the first observed organometallic
products. In the presence of acetonitrile, Cp*Rh(PPh3)Me2
and Cp*Rh(PPh3)(NCMe)22ϩ were shown to undergo compro-
portionation to give Cp*Rh(PPh3)(NCMe)Meϩ, and therefore
the ultimate product resembles the formation of 7 from oxida-
tion of 6 in Me2SO. The comproportionation of the phosphine
conducted in acetonitrile and CDCl3 was used as NMR solvent.
Compound 4 is only observed when the NMR analysis is
conducted in acetonitrile-d3, hence coordinated MeCN in 4
could not be observed (see the Experimental section for
details). Upon addition of Me2SO to the mixture in CD3CN,
J. Chem. Soc., Dalton Trans., 2002, 975–982
977

View Article Online
Table
1
Selected interatomic distances (Å) and angles (Њ) in
Cp*Ir(Me2SO)Me2 (1)
analogs is, however, much slower (time scale several hours)
than the reductive elimination [k(20 ЊC) = 96 sϪ1 in acetonitrile],
ϩ
.
ؒ
contrasting a possibly direct formation of 7 from 6
Molecule A
Molecule B
The Rh complex 6 exhibited a chemically irreversible cyclic
voltammogram (Ep = 0.22 V vs. Fc) at ν = 1.0 V sϪ1. The
irreversibility persisted even at ν = 50 V sϪ1, thus the follow-up
Ir–S(1)
Ir–C(1)
Ir–C(2)
S(1)–O(1)
Ir–Cp* centroid
2.203(2)
2.11(1)
2.15(1)
1.495(7)
1.886 ave.
2.206(2)
2.10(1)
2.13(1)
1.477(7)
ϩ
ؒ
chemical reaction of electrode-generated 6 was too fast for a
detailed voltammetric investigation of the kinetics and reaction
mechanism using our experimental setup. However, the qualita-
tive findings are indicative of some fundamental differences in
the reactivities of analogous Rh and Ir complexes, as will be
elaborated later.
S(1)–Ir–C(1)
S(1)–Ir–C(2)
C(1)–Ir–C(2)
Ir–S(1)–O(1)
91.4(3)
90.0(3)
83.3(5)
118.0(3)
92.7(3)
89.9(3)
83.2(5)
118.1(3)
X-Ray crystal structure determinations
More convenient syntheses of Cp*Ir(Me2SO)Me2 (1) and Cp*-
Rh(Me2SO)Me2 (6), using dimethylzinc in place of trimethyl
aluminium reagents, were developed (see the Experimental
section). The two compounds are isomorphous, have a pseudo-
octahedral ligand arrangement about the metal center and, as
Rh() and Ir() are virtually the same size,20 show only minor
differences in bond lengths and angles. Fig. 2 (top) shows a
Table
2
Selected interatomic distances (Å) and angles (Њ) in
Cp*Rh(Me2SO)Me2 (6)
Molecule A
Molecule B
Rh–S(1)
Rh–C(1)
Rh–C(2)
S(1)–O(1)
Rh–Cp* centroid
2.211(1)
2.090(4)
2.092(4)
1.484(3)
1.889 ave.
2.210(1)
2.099(5)
2.103(5)
1.480(3)
S(1)–Rh–C(1)
S(1)–Rh–C(2)
C(1)–Rh–C(2)
Rh–S(1)–O(1)
92.0(1)
90.9(1)
83.2(2)
117.7(1)
92.2(1)
91.2(2)
82.9(2)
117.6(1)
Table 3 Selected distances (Å) and angles (Њ) in the cation part of
2(BArfϪ
)
Ir–S(1)
Ir–S(2)
Ir–C(1)
Ir–Cp* centroid
2.283(2)
2.263(2)
2.119(8)
1.886
S(1)–Ir–S(2)
S(1)–Ir–C(1)
S(2)–Ir–C(1)
93.21(8)
87.5(3)
85.3(3)
cationic part of 2(BArfϪ). The anion was unexceptional and is
included in the ESI.† Table 3 shows selected bond distances and
angles in the cation, and crystal structure data are included
in Table 4.
In each of the three structures analyzed, the complex adopts
the geometry of a three-legged pianostool, with Cp* as the base
and the methyl and S-coordinated Me2SO ligands as the legs.
The Cp*(centroid)–M distances are 1.886 Å and 1.889 Å for
1 and 6 respectively (these are averages as there are two inde-
pendent molecules per asymmetric unit). The Me2SO ligand is
S-bound with metal–S bond lengths of 2.205(2) and 2.211(1) Å
for M = Ir and Rh, respectively. The Ir–S distances in 2 are
slightly different [2.263(2) and 2.283(2) Å] and slightly longer
than those in 1 and 6, possibly for steric reasons. The metal–
methyl carbon bond lengths (average values for crystallographi-
cally independent molecules) of 2.122 (1), 2.096 (6), and
2.119(8) Å for (2) are normal. The Ir–Me and Ir–Cp*(centroid)
bond distances of 2 are close to those of 1 [1: Ir–Me 2.11(1) and
2.14(1) Å, Ir–Cp* 1.886 Å]. The S–Ir–S and S–metal–Me bond
angles reflect the pseudo-octahedral coordination geometry
around the metal center and the greater steric demands of
the Me2SO compared to the methyl group. Thus, the angles
(degrees) at the metal increase in the order Me–Ir–Me 83.3
(av. in 1) ≈ Me–Rh–Me 83.0 (av. in 6) < Me–Ir–S 85.3 and 87.5
(in 2) < Me–Ir–S 90.0 and 92.0 (av. in 1) ≈ Me–Rh–S 91.0 and
92.1 (av. in 6) < S–Ir–S 93.2Њ (in 2). The chief feature is that the
angles reflect the greater steric demands of the S-bonded
Me2SO by comparison with Me, thus, S–metal–S > S–metal–C
> C–metal–C. It also appears that in the cation 2, the S–metal–
C angle is somewhat larger than in the neutral complexes 1 and
6. The Me2SO ligand with the shortest bond to the metal center
also exhibits the smallest S–Ir–Me angle.
Fig. 2 ORTEP27 structures of Cp*Ir(Me2SO)Me2 (1, top) and Cp*-
Ir(Me2SO)2Meϩ (2, bottom). Hydrogen atoms have been omitted for
clarity.
perspective view of Cp*Ir(Me2SO)Me2 (the Rh analog is not
shown because of the close structural resemblance to 1,
vide infra). Key bond lengths and angles for 1 and 6 are given in
Tables 1 and 2, respectively and details of data collection and
refinement are summarised in Table 4 in the Experimental
section.
The salt Cp*Ir(Me2SO)2MeϩBArfϪ 2(BArfϪ) was obtained by
protonolysis of 1 with HBF4ؒEt2O followed by anion exchange
with NaϩBArfϪ in dichloromethane (see the Experimental
section). The X-ray structure (150 K) of this species was also
determined. Fig. 2 (bottom) includes an ORTEP plot of the
978
J. Chem. Soc., Dalton Trans., 2002, 975–982

View Article Online
Table 4 X-Ray crystallographic data for 1, 6, and 2(BArfϪ
)
Compound
1
6
2(BArfϪ
)
Chemical formula
FW
Crystal system
Space group
C14H27OIrS
435.62
C14H27ORhS
346.33
C47H42BF24IrO2S2
1361.97
Monoclinic
P21/n
Triclinic
Triclinic
¯
¯
P1 (no. 2)
P1 (no. 2)
Z
a/Å
b/Å
c/Å
α/Њ
β/Њ
γ/Њ
4
4
4
8.832(1)
12.913(2)
14.394(2)
99.865(2)
96.010(2)
93.146(2)
1604.0(3)
1.804
0.23 × 0.16 × 0.08
150
Bruker P4/RA/-
SMART 1000 CCD
MoKα
8.813(1)
12.864(1)
14.435(2)
100.266(2)
96.114(2)
93.100(2)
1596.6(3)
1.441
0.25 × 0.15 × 0.09
150
Bruker P4/RA/-
SMART 1000 CCD
MoKα
13.547(1)
25.816(4)
16.645(3)
90
112.06(1)
90
5395(1)
1.677
0.2 × 0.3 × 0.4
150
Volume/Å3
ρcalc/g cmϪ3
Crystal dimensions/mm
Temperature/K
Diffractometer
Siemens SMART CCD
Radiation
MoKα
2θ limit/Њ
56.6
8389
7262
5991
0.081
322
8.440
56.6
8389
7226
5407
0.047
322
1.186
57.5
28879
11609
8490
0.062
714
2.67
No. of data collected
No. of unique data
No. of observed data (I > 2σ(I )
Agreement between equivalent data (Rint
No. of parameters varied
µ/mmϪ1
)
Absorption correction
Gaussian integration
(face-indexed)
0.0576
Gaussian integration
(face-indexed)
0.0394
Analytical (xprep, Siemens 1997)
R1 (F0)
0.0655
0.1344
2
wR2 (F0 ) (I > 2σ)
0.1593
0.0928
Scheme 2
products in acetonitrile were 1 : 1 : 1 molar ratios of CpRe-
Discussion
(NO)(PR )(᎐CH )ϩ, CH , and CpRe(NO)(PR )(NCMe)ϩ. The
᎐
3
2
4
3
Mechanistic aspects of the oxidation of Cp*Ir(Me2SO)Me2
methylene complex underwent quantitative intermolecular
ligand coupling in a slower step to give equimolar amounts
The ratio of methane to ethene formed from the chemical
and electrochemical one-electron oxidations of 1 showed some
differences depending on the reaction conditions. However,
the similarities suggest that the fundamental chemistry that is
initiated by the one-electron process is similar in both cases. The
electrochemical results, in particular the quantitative DCV
data, shed some light on features of the oxidation mechanism
under the particular reaction conditions and on the relatively
short (less than a second) CV timescale. The following features
are particularly interesting: (1) the reaction of 1 is second
order in Ir, (2) the enthalpy of activation is near zero and the
entropy of activation is strongly negative, the latter being con-
sistent with the second-order nature of the reaction, and (3)
there is an appreciable kH/kD kinetic isotope effect for the
reaction of Ir–CH3 vs. Ir–CD3.
The formation of ethene together with the reactivity implied
by the quantitative DCV data is reminiscent of the behavior
that was observed for the oxidation of CpRe(NO)(PR3)Me (R =
Ph,21 Me14a), interpreted according to Scheme 2. The reactions
of the Re cation radicals were also second-order processes with
low (even negative) values for ∆H‡, strongly negative ∆S‡, and
kinetic isotope effects kH/kD in the range 2–3. The first observed
of the ethene complex CpRe(NO)(PR )(CH ᎐CH )ϩ and CpRe-
᎐
3
2
2
(NO)(PR3)(NCMe)ϩ. The low values observed there for ∆H‡
result from the negative ∆HЊ1 for the exothermic preequilibrium
dimerization canceling, in part, the positive ∆H‡ for the rate-
2
limiting step.
In the present case, however, the amounts of ethene formed
are far less than would be expected from a mechanism
analogous to that outlined in Scheme 2. A possible reason may
be that two mechanisms operate in parallel, one forming
methane and the other giving ethene (or both) (Scheme 3). One
might expect at least one additional organometallic product to
be formed in such a case (although the diverging pathways
might conceivably converge on the same organometallic
product). Taking into account that the chemical oxidation
generates only 5–6% of ethene, the amounts of the putative
alternative organometallic product(s) may well be close to the
ϩ
ؒ
1
detection limit of the H NMR analysis and these species may
remain unobserved. In the case of electrochemical oxidation,
somewhat larger amounts of ethene are produced (12–13%).
However, in this case the NMR analysis of the product
mixtures are preceded by dichloromethane extraction and
J. Chem. Soc., Dalton Trans., 2002, 975–982
979

View Article Online
reduced steric bulk of the Me2SO vs. the PPh3 ligand may also
be a contributing factor.
Concluding remarks
The differences in reactivities of Cp*Ir(Me2SO)Me2 and its Rh
analog upon oxidation are striking. As elaborated in the Intro-
duction, marked differences exist for other systems of these
metals as well. It is a generally accepted trend that metal–
carbon, metal–hydrogen, and metal–metal bond dissociation
energies increase on descending from second- to third-row
metals in the Periodic table. The Rh di- and tri-alkyl
compounds Cp*Rh(PPh3)Me2 and Cn*RhMe3 both yield
carbon–carbon coupling products upon oxidation, involving
simultaneous bond breaking of two metal–carbon bonds in
the same molecule. The mechanism for the oxidation of
Cp*Rh(Me2SO)Me2 is not known but the fact that ethane is
formed as the only organic product suggests that the same
applies here. In contrast, the Ir dialkyls Cp*Ir(PPh3)Me2
and Cp*Ir(Me2SO)Me2 appear to react after oxidation by
mechanisms that involve bond formation as a key step (Ir
insertion into a Cp* C–H bond for L = PPh3; Ir–Ir bond form-
ation, possibly via an Ir–Me–Ir bridge, for L = Me2SO under
conditions that favor second-order reactivity). The organic
products are eliminated in a subsequent step. These divergent
reaction paths are fully consistent with the relative strengths of
bonds to second- vs. third-row metals. Analogous differences
may be anticipated for other pairs of metals as well, and will
have important implications for the use of second- vs. third-row
metals in catalysis.
Scheme 3
work-up (see the Experimental section for details), and it is
conceivable that the products from the proposed second-order
ethene forming reaction path in Scheme 3 may be lost during
work-up.
It is well known that Cp*Ir complexes readily form dinuclear
species, for example µ-hydride and µ-methylene complexes.22
Furthermore, the dimerization of the Cp*Ir(CO)2 moiety gives
a variety of di-iridium species23 including [{Cp*Ir(CO)2}2]2ϩ. It
is important to appreciate that the CV experiments, carried out
ϩ
ؒ
on a stationary solution, measure what happens to 1 under
conditions where the local concentration of 1 ϩ is rather high in
ؒ
the undisturbed diffusion layer at the electrode surface. On the
other hand, the bulk electrolysis is performed on a vigorously
stirred solution of 1 with an experimental timescale of a few
ϩ
ؒ
minutes. Under these conditions, 1 is effectively transported
away from the electrode diffusion layer and diluted into the
bulk solution before significant reaction can occur. Con-
sequently, second-order reactions of 1 ϩ, which should give
ؒ
equal amounts of methane and ethene on a carbon basis, will
be less favored and may be suppressed by competing first-order
pathways, giving only methane, that are not kinetically sig-
nificant and therefore not detected in the CV experiments.
Chemical oxidation with Cp2FeϩBF4Ϫ occurs under conditions
where the equilibrium concentration of 1 ϩ is relatively low; the
ؒ
Experimental
second-order reaction that is implied by the CV data is again
attenuated due to significant contributions from the competing
first-order pathway.
General procedures
All operations were carried out under an atmosphere of Ar (for
electrochemical experiments) or N2 with the use of vacuum line,
Schlenk, syringe, or drybox techniques. THF and ether were
distilled from blue/purple solutions of sodium benzophenone
ketyl. Acetonitrile, Me2SO, and dichloromethane were distilled
from CaH2. Me2SO was further stored over activated 4 Å
The most obvious reaction that would produce 2 from 1
with deuterium incorporation from added water would be a
protonolysis of 1. However, solutions of 1 in Me2SO-d6 are
stable on the timescale of the oxidation reactions, rendering
this explanation unattractive. Other possibilities include a
1
protonolysis of 17-electron 1 ϩ or—perhaps more likely—of an
molecular sieves. H NMR and 13C{1H} NMR spectra were
ؒ
ϩ
.
ؒ
electron-rich, 19-electron Me2SO adduct Cp*Ir(Me2SO)2Me2
recorded on Bruker Avance DXP 300, Bruker ACS 250, and
Bruker AMX 400 instruments. All NMR spectra are reported
in ppm (δ) relative to tetramethylsilane, with the residual
solvent proton resonance and carbon resonance as internal
standards. IR spectra were obtained on a Nicolet 560 E.S.P
FTIR spectrometer. GC analyses were conducted on Hewlett-
Packard 5710A and HP 6890ϩ gas chromatographs using a
30 m × 0.54 mm Megabore GS-Q capillary column. The
elemental analyses were carried out at the elemental analysis
laboratory, Department of Chemistry, University of Sheffield.
The GC–MS analyses were carried out with a Fisons GC 8000
gas chromatograph (5 m × 0.25 mm Chrompack CP-SIL fused
silica column) interfaced to a Fisons VG ProSpec-Q mass
spectrometer, and with a Fisons 8060 gas chromatograph
(30 m × 0.22 mm SGE BPX5 fused silica column) interfaced
to a Micromass ProSpec magnetic sector mass spectrometer.
Electrochemical measurements were performed on an
EG&G-PAR Model 273 potentiostat/galvanostat driven by
an external HP 3314A function generator. The signals were fed
to a Nicolet 310 digital oscilloscope and processed by an on-
line personal computer. The working electrodes were Pt disk
electrodes (d = 0.4–1.0 mm). The counter electrode was a Pt
wire, and the Ag wire reference electrode assembly was filled
with acetonitrile/0.01 M AgNO3/0.1 M Bu4NϩPF6Ϫ. The
reference electrode was calibrated against the Cp2Fe/Cp2Feϩ
(Fc) couple. The positive feedback iR compensation circuitry of
the potentiostat was employed. A Pt gauze working electrode
was used for the constant potential coulometry and preparative
The oxidizing reaction conditions may generate a more acidic
medium than would have been the case without the oxidant.
Comparison of reactivities of Cp*M(Me2SO)Me2 and
Cp*M(PPh3)Me2 for M ؍ Rh, Ir
Due to the stronger donor power of PPh3 compared to Me2SO,
the oxidation potentials of Cp*Rh(PPh3)Me2 (0.04 V vs. Fc
in MeCN–CH2Cl2 9 : 1)3f and Cp*Ir(PPh3)Me2 (Ϫ0.01 V in
MeCN)3g are markedly lower than those of the corresponding
Cp*M(Me2SO)Me2 compounds 1 (0.18 V in MeCN) and 6 (Ep =
0.22 V in MeCN; this represents a minimum value due to the
unknown kinetic potential shift caused by the rapid homo-
geneous reaction of 6 ϩ). This may also, in part, help explain
ؒ
the significant differences in the rates of the oxidatively induced
reactions of the PPh3 vs. Me2SO pairs: whereas the rate con-
stant for the oxidatively induced reaction of Cp*Rh(PPh3)Me2
is close to 100 sϪ1 at 20 ЊC, 6 reacts too fast (> 1000 sϪ1) to be
measured in our labs. The pair 1 and Cp*Ir(PPh3)Me2 exhibit a
similar trend, as the Me2SO compound displays a considerably
higher reaction rate [k(25 ЊC) = 3.2 × 104 MϪ1 Ϫ1] than the
s
PPh3 analog, which reacts too slowly to be measured by cyclic
voltammetry. The enhanced electron deficiency of 1 and 6 com-
pared to the PPh3 complexes might be expected to provide
kinetically more reactive radical cations, which in all cases
undergo reactions that are, at least formally, reductive elimin-
ation reactions as defined in a broad sense in eqns. (1)–(3). The
980
J. Chem. Soc., Dalton Trans., 2002, 975–982

View Article Online
electrolysis experiments. The electrochemical experiments in
Me2SO and acetonitrile were carried out at 20–25 ЊC unless
otherwise stated, and at 0 ЊC in dichloromethane.
Cp*Ir(Me2SO)2Me؉BArf؊ [2(BArf؊)]. To
a
solution of
2(BF4Ϫ) (30 mg, 51 µmol) in dichloromethane (10 mL) was
added NaϩBArfϪ (45 mg, 51 µmol) and the mixture was stirred
at ambient temperature for 1 h. The volatiles were removed
in vacuo and the product was extracted from the mixture by
dissolving in toluene. Filtration through Celite and evaporation
of the solvent gave the product in quantitative yield. X-Ray
quality crystals were obtained by slow evaporation of diethyl
ether into a solution of 2(BArfϪ) in CH2Cl2. 1H NMR (CDCl3,
200 MHz) δ 0.92 (s, 3 H, IrMe), 1.66 (s, 15 H, C5Me5), 3.17 (s, 12
H, 2 Me2SO), 7.52 (s, 4 H, p-H at BArfϪ), 7.69 (br, 8 H, o-H at
BArfϪ).
Syntheses
Cp2FeϩBF4Ϫ and Cp(C5H4COMe)FeϩBF4Ϫ were prepared from
the appropriately substituted ferrocenes with AgBF4,16 and
(CD3)2Mg,24 [Cp*IrCl2]2,25 and NaϩBArfϪ 26 were prepared
according to published procedures.
Cp*Ir(Me2SO)Me2. A more convenient synthesis than the
published one using Al2Me6 9a was as follows: ZnMe2 (130 µL of
a 2.0 M solution in toluene, Aldrich; 0.260 mmol) was added
to Cp*IrBr2(Me2SO) (102 mg, 0.18 mmol) in THF (10 mL) at
Ϫ80 ЊC. The solution was warmed to room temperature and
stirred (30 min) and was then hydrolyzed by adding water
(100 µL). The solvent was removed in vacuo and the residue
extracted in air with CH2Cl2 (3 × 5 mL). The extracts were
filtered through Celite and the solvent was removed under
reduced pressure. The residue was then extracted with n-
pentane (5 × 5 mL doped with Me2SO) and filtered, and the
solvent volume was reduced to ca. 5 mL. The solution was
cooled at Ϫ20 ЊC for 16 h to give the product as pale yellow
crystals (51 mg, 64%).
Constant-potential coulometry experiments
The experiments were conducted using 3–4 mM solutions of
the substrate in an electrolyte solution of 0.05–0.06 M Me4-
Ϫ
NϩBF4 in Me2SO unless otherwise described. The electrode
potential was maintained at ϩ0.1 V relative to EЊox(1). An
H-shaped electrolysis cell was used, in which the two cell com-
partments were separated by a grade 4 glass frit junction. Each
compartment was filled with 15–20 mL of electrolyte solution.
For gas analysis experiments, a modified electrolysis cell
was used. This cell was equipped with a gas-tight reaction com-
partment with a rubber septum for withdrawal of gas samples.
Chemical oxidation of Cp*M(Me2SO)Me2 (M ؍ Ir, Rh) with
Cp*Rh(Me2SO)Me2. Was obtained as large yellow crystals
(1.9 g, 85%) by a similar reaction sequence, starting from
[Cp*RhCl2]2 (2.0 g, 3.2 mmol) and ZnMe2 (5 mL of a 2.0 M
solution). Both compounds were spectroscopically identical to
authentic samples and their identities were further confirmed
by an X-ray crystal structure determination of each.
؊
Cp2Fe؉PF6 —analysis of the organometallic and gaseous
products
A 5 mL Schlenk flask was loaded with a magnetic stirrer and
the solid reagents (typically ca. 10 mg of the substrate, 23 µmol
for M = Ir). A slight excess of the substrate was used to ensure
complete consumption of the oxidant, thereby avoiding the
presence of paramagnetic species in the product mixture. The
flask was fitted with a septum and the solvent (5 mL) was added
by syringe while stirring. For the Ir complex, the complete con-
sumption of the oxidant took place during a 2–4 h period as
judged by the complete color change from blue, via green,
to yellow-orange, whereas oxidation of the Rh complex was
complete in a few seconds. Gas samples for GC or GC–MS
measurements were withdrawn by syringe from the headspace
above the solution. The volatiles were removed in vacuo and the
residue was subjected to analysis by 1H NMR spectroscopy.
Cp*Ir(Me2SO)(CD3)2 (1-d6). [Cp*IrCl2]2 (50 mg, 0.063 mmol)
was dissolved in THF (10 mL) and Me2SO (30 µL, 0.42 mmol)
was added while stirring. The solution was cooled to Ϫ78 ЊC
and a 0.05 M solution of (CD3)2Mg in THF (2 mL, 0.1 mmol)
was added dropwise. The mixture was slowly warmed to room
temperature, and stirring was continued for 2 days after which
wet THF (THF–water 3 : 1) was added dropwise to ensure that
no (CD3)2Mg was left as evidenced by the cessation of gas evo-
lution. The volatiles were removed in vacuo, and the product
was extracted with pentane (10 mL). The extract was filtered
and washed with water (2 × 5 mL) to remove most of the mono-
methylated contaminant. The solution was dried over MgSO4
for 30 min and filtered. The solvent was removed under
vacuum, and the product was purified by sublimation at 40 ЊC/
1.5 mm Hg to give a yellow solid (11 mg, 20%). Use of excess
(CD3)2Mg should be avoided since this leads to the formation
of Cp*Ir(CD3)4 which is difficult to separate from the desired
product (separation is possible by flash chromatography using
silica–toluene, but this leads to lower yields). The isotopic
Chemical oxidation of Cp*Ir(Me2SO)Me2 in acetonitrile. A 5
mL round-bottomed flask was loaded with Cp*Ir(Me2SO)Me2
Ϫ
(6.9 mg, 0.016 mmol) and Cp2FeϩBF4 (4.2 mg, 0.015 mmol).
Acetonitrile (2 mL) was added while stirring, and stirring was
continued for 30 min. The volatiles were removed in vacuo
and the residue was dissolved in CDCl3. 1H NMR (CDCl3, 300
MHz) 2: δ 0.92 (s, 3 H, IrMe), 1.82 (s, 15 H, C5Me5), 3.38 (s,
2 × 6 H, Me2SO); 3: δ 0.74 (s, 3 H, IrMe), 1.75 (s, 15 H, C5Me5),
2.56 (s, 3 H, MeCN), 3.09 (s, 3 H, Me2SO), 3.48 (s, 3 H,
Me2SO). Electrospray MS (MeCN reaction mixture): m/z
calc. for [C15H24IrN2]ϩ (Mϩ of 4) 423, 424, 425, 426, 427; found
m/z = 423, 424, 425, 426, 427; also m/z = 382, 383, 384, 385, 386
[Mϩ Ϫ MeCN].
1
purity of 1-d6 was checked by H NMR spectroscopy, and we
estimate the complex to be more than 99% deuterated at the
Ir-methyl groups. 1H NMR (CDCl3, 300 MHz) δ 1.69 (s, 15 H,
C5Me5), 2.89 (s, 6 H, Me2SO).
Cp*Ir(Me2SO)2Me؉BF4؊ [2(BF4؊)]. HBF4ؒEt2O (33 µL, 0.22
mmol) was added dropwise to a stirred solution of Cp*Ir(Me2-
SO)Me2 (100 mg, 0.23 mmol) and Me2SO (160 µL, 2.25 mmol)
in ether (5 mL), leading to the instantaneous precipitation of
a yellow-white solid. The volatiles were removed in vacuo and
the residue was washed with ether (6 × 2 mL). The residue
was dissolved in dichloromethane, the solution was filtered
through Celite, and the solvent was removed in vacuo yielding
Chemical oxidation of Cp*Ir(Me2SO)Me2 in acetonitrile-d3.
An NMR tube fitted with a ground-glass joint was loaded with
Cp*Ir(Me2SO)Me2 (7.7 mg, 0.018 mmol) and Cp2FeϩBF4
Ϫ
(3.9 mg, 0.014 mmol). The solvent (0.5 mL) was added by vac-
uum transfer and the tube was flame sealed under vacuum.
1H NMR (CD3CN, 200 MHz) 2: not observed; 3: δ 0.57 (s, 3 H,
IrMe), 1.75 (s, 15 H, C5Me5), 3.05 (s, 6 H, Me2SO), 3.21 (s, 6 H,
Me2SO); 4: δ 0.68 (s, 3 H, IrMe), 1.75 (s, 15 H, C5Me5). The tube
was opened and equipped with a rubber septum. (CH3)2SO
1
the product as a pale yellow solid (131 mg, 99%). H NMR
(CD2Cl2, 300 MHz) δ 0.96 (s, 3 H, IrMe), 1.83 (s, 15 H, C5Me5),
3.35 (s, 6 H, Me2SO), 3.36 (s, 6 H, Me2SO). 13C{1H} NMR
(CDCl3, 62.9 MHz) δ Ϫ16.0 (IrMe), 8.6 (C5Me5), 45.5, 47.2
(Me2SO), 100.4 (C5Me5). Anal. calc. for C15H30BF4IrO2S2: C,
30.76; H, 5.17; S, 10.95. Found: C, 30.45; H, 5.06; S, 10.96%.
1
(5–10 µL) was added by syringe and a H NMR analysis was
conducted. The solvent was then removed in vacuo, CD2Cl2
added, and another 1H NMR analysis was performed. The
findings are included in full in the Results section.
J. Chem. Soc., Dalton Trans., 2002, 975–982
981

View Article Online
Cp*Rh(Me2SO)2Me؉ spectroscopic data
1H NMR (CD2Cl2, 200 MHz) δ 0.96 (d, J = 2.3 Hz, 3 H, RhMe),
1.77 (s, 15 H, C5Me5), 3.15 (s, 6 H, Me2SO), 3.20 (s, 6 H,
Me2SO).
6 (a) M. Tilset, J. Am. Chem. Soc., 1992, 114, 2740; (b) V. Skagestad
and M. Tilset, J. Am. Chem. Soc., 1993, 115, 5077; (c) M. Tilset,
J.-R. Hamon and P. Hamon, Chem. Commun., 1998, 765; (d ) M.
Tilset, I. Fjeldahl, J.-R. Hamon, P. Hamon, L. Toupet, J.-Y. Saillard,
K. Costuas and A. Haynes, J. Am. Chem. Soc., 2001, 123, 9984.
7 Abbreviations: Cp*
=
η5-C5H5; Cn*
=
1,4,7-trimethyl-1,4,7-
Ϫ
.
triazacyclononane; BArfϪ = B[3,5-C6H3(CF3)2]4
X-Ray crystal structure determinations
8 (a) P. Diversi, S. Iacoponi, G. Ingrosso, F. Laschi, A. Lucherini and
P. Zanello, J. Chem. Soc., Dalton Trans., 1993, 351; (b) P. Diversi,
S. Iacoponi, G. Ingrosso, F. Laschi, A. Lucherini, C. Pinzino,
G. Uccellobarretta and P. Zanello, Organometallics, 1995, 14, 3275;
(c) P. Diversi, V. Ermini, G. Ingrosso, A. Lucherini, C. Pinzino and
F. Simoncini, J. Organomet. Chem., 1998, 555, 135; (d ) P. Diversi,
F. F. deBiani, G. Ingrosso, F. Laschi, A. Lucherini, C. Pinzino and
P. Zanello, J. Organomet. Chem., 1999, 584, 73.
X-Ray diffraction quality crystals of 1 and 6 were obtained by
slow cooling of concentrated n-pentane solutions of the com-
pounds to Ϫ20 ЊC. Cp*Ir(Me2SO)Me2 and Cp*Rh(Me2SO)-
¯
Me2 crystallized in the P1 (no. 2) space group. Details of data
collection and refinement are summarised in Table 4.
CCDC reference numbers 169020–169022.
9 (a) A. Vazquez de Miguel, M. Gomez, K. Isobe, B. F. Taylor,
B. E. Mann and P. M. Maitlis, Organometallics, 1983, 2, 1724;
(b) J. M. Kisenyi, G. J. Sunley, J. A. Cabeza, A. J. Smith, H. Adams,
N. J. Salt and P. M. Maitlis, J. Chem. Soc., Dalton Trans., 1987, 2459;
(c) M. Gomez, P. I. W. Yarrow, D. J. Robinson and P. M. Maitlis,
J. Organomet. Chem., 1985, 279, 115; (d ) G. J. Sunley, P.del
C. Menanteau, H. Adams, N. A. Bailey and P. M. Maitlis, J. Chem.
Soc., Dalton Trans., 1989, 2415; (e) N. Dudeney, O. N. Kirchner,
J. C. Green and P. M. Maitlis, J. Chem. Soc., Dalton Trans., 1984,
1877; ( f ) M. Gomez, J. M. Kisenyi, G. J. Sunley and P. M. Maitlis,
J. Organomet. Chem., 1985, 296, 197; (g) F. P. Fanizzi, G. J. Sunley,
J. A. Wheeler, H. Adams, N. A. Bailey and P. M. Maitlis, Organo-
metallics, 1990, 9, 131.
See http://www.rsc.org/suppdata/dt/b1/b107451m/ for crys-
tallographic data in CIF or other electronic format.
Acknowledgements
We thank Harry Adams (University of Sheffield) and Dr
Robert MacDonald (University of Alberta, Edmonton) for
collecting the X-ray data for 1 and 6 and solving their struc-
tures, Simon Thorpe for kind assistance with labeling studies,
Dr Tony Haynes for helpful discussions, the EPSRC for support
(grant no. GR/L94192Z/01), and the British Council/Research
Council of Norway Collaborative Scheme for travel grants.
10 Experimental conditions: 2.0 mM substrate in Me2SO/0.1 M
Bu4NϩPF6Ϫ, 25 ЊC d = 0.4 mm Pt disk electrode.
11 A. J. Bard and L. R. Faulkner, Electrochemical Methods.
Fundamentals and Applications, Wiley, New York, 1980.
12 The change of supporting electrolyte compared to that used for the
CV investigation was done in order to facilitate the separation of
product and electrolyte during work-up, see the Experimental
section for details.
13 In these measurements a faint, but fairly constant current was
observed to flow through the cell well after the consumption of
1 Faraday molϪ1 and after unusually long electrolysis times. Similar
behavior has been reported by others and has been attributed to the
generation of species that are capable of mediating electrocatalytic
solvent oxidation, see G. S. Bodner, J. A. Gladysz, M. F. Nielsen and
V. D. Parker, Organometallics, 1987, 6, 1628.
14 (a) M. Tilset, in Energetics of Organometallic Species, ed. J. A. M.
Simões, Kluwer Academic, Dordrecht, 1992, p. 109; (b) M. Tilset,
V. Skagestad and V. D. Parker, in Molecular Electrochemistry of
Inorganic, Bioinorganic and Organometallic Compounds, eds. A. J. L.
Pombeiro and J. A. McCleverty, Kluwer Academic, Dordrecht,
1992, p. 267.
References and notes
1 (a) J. P. Collman, L. S. Hegedus, J. R. Norton and R. G. Finke,
Principles and Applications of Organotransition Metal Chemistry,
University Science Books, Mill Valley, 1989.
2 (a) J. Y. Chen and J. K. Kochi, J. Am. Chem. Soc., 1977, 99, 1450;
(b) W. H. Tamblyn, R. J. Klingler, W. S. Hwang and J. K. Kochi,
J. Am. Chem. Soc., 1981, 103, 3161; (c) R. J. Klingler, J. C. Huffman
and J. K. Kochi, J. Am. Chem. Soc., 1982, 104, 2147; (d ) J. C. Kotz,
W. Vining, W. Coco, R. Rosen, A. R. Dias and M. H. Garcia,
Organometallics, 1983, 2, 68; (e) R. F. Jordan, R. E. LaPointe,
C. S. Bajgur, S. F. Echols and R. Willett, J. Am. Chem. Soc., 1987,
109, 4111; ( f ) T. Aase, M. Tilset and V. D. Parker, J. Am. Chem.
Soc., 1990, 112, 4974; (g) M. R. Burk, W. Tumas, M. D. Ward and
D. R. Wheeler, J. Am. Chem. Soc., 1990, 112, 6133; (h) A. L. Seligson
and W. C. Trogler, J. Am. Chem. Soc., 1992, 114, 7085; (i) S. L.
Borowsky, N. C. Baenziger and R. F. Jordan, Organometallics,
1993, 12, 486; (j) D. Zhu, S. V. Lindeman and J. K. Kochi,
Organometallics, 1999, 18, 2241.
3 (a) D. G. Morrell and J. K. Kochi, J. Am. Chem. Soc., 1975, 97, 7262;
(b) T. T. Tsou and J. K. Kochi, J. Am. Chem. Soc., 1978, 100, 1634;
(c) M. Almemark and B. Åkermark, J. Chem. Soc., Chem. Commun.,
1978, 66; (d ) W. Lau, J. C. Huffman and J. K. Kochi,
Organometallics, 1982, 1, 155; (e) K. Ishikawa, S. Fukusumi and
T. Tanaka, Inorg. Chem., 1989, 28, 1661; ( f ) A. Pedersen and M.
Tilset, Organometallics, 1993, 12, 56; (g) A. Pedersen and M. Tilset,
Organometallics, 1994, 13, 4887; (h) A. Pedersen, M. Tilset,
K. Folting and K. G. Caulton, Organometallics, 1995, 14, 875;
(i) K. Koo and G. L. Hillhouse, Organometallics, 1995, 14, 4421;
(j) E. Fooladi and M. Tilset, Inorg. Chem., 1997, 36, 6021;
(k) R. Han and G. L. Hillhouse, J. Am. Chem. Soc., 1997, 119,
8135; (l ) M. Scmittel, A. Burghart, H. Werner, M. Laubender and
R. Söllner, J. Org. Chem., 1999, 64, 3077.
15 Uncertainties in activation parameters are given as two standard
deviations from the regression analysis.
16 N. G. Connelly and W. E. Geiger, Chem. Rev., 1996, 96, 877.
17 Isotopomer distribution was determined by comparison of the mass
spectrum of the experimental isotopomer mixture to peak intensities
in reference spectra of the individual isotopomers: C2H4 (53–98%),
C2D4 (1–40%), C2H3D (0–29%), C2H2D2 (0–2%), C2HD3 (0–15%).
18 Approximate H2O content of 0.6 mol% was estimated by
comparison of NMR integrals of water and residual protio-
dimethylsulfoxide in the deuterated solvent.
19 For some pertinent reviews, see: (a) D. Astruc, Electron Transfer and
Radical Processes in Transition-Metal Chemistry, VCH, Weinheim,
1995; (b) D. Astruc, Organometallic Radical Processes, ed. W. C.
Trogler, Elsevier, Amsterdam, 1990.
20 P. M. Maitlis, Chem. Soc. Rev., 1981, 10, 1.
4 It is frequently difficult to distinguish truly concerted reactions from
two-step ones for the two-ligand eliminations in eqn. (2). See
discussion in ref. 5g.
21 M. Tilset, G. S. Bodner, D. R. Senn, J. A. Gladysz and V. D. Parker,
J. Am. Chem. Soc., 1987, 109, 7551.
22 C. White, A. J. Oliver and P. M. Maitlis, J. Chem. Soc., Dalton
Trans., 1973, 1901; A. Vázquez de Miguel, K. Isobe, B. F. Taylor,
A. Nutton and P. M. Maitlis, J. Chem. Soc., Chem. Commun., 1982,
758.
23 F. W. B. Einstein, R. H. Jones, X. Zhang, X. Yan, R. Nagelkerke and
D. Sutton, J. Chem. Soc., Chem. Commun., 1989, 1424.
24 A. C. Cope, J. Am. Chem. Soc., 1935, 57, 2238.
25 R. G. Ball, W. A. G. Graham, D. M. Heinekey, J. K. Hoyano,
A. D. McMaster, B. M. Mattson and S. T. Michel, Inorg. Chem.,
1990, 29, 2023.
26 M. Brookhart, B. Grant and A. F. Volpe, Organometallics, 1992, 11,
3920.
27 M. N. Burnett and C. K. Johnson, ORTEP3, Report ORNL-6895,
Oak Ridge National Laboratory, Oak Ridge, TN, 1996.
5 (a) O. B. Ryan, M. Tilset and V. D. Parker, J. Am. Chem. Soc., 1990,
112, 2618; (b) O. B. Ryan, M. Tilset and V. D. Parker,
Organometallics, 1991, 10, 298; (c) O. B. Ryan and M. Tilset,
J. Am. Chem. Soc., 1991, 113, 9554; (d ) O. B. Ryan, K.-T. Smith and
M. Tilset, J. Organomet. Chem., 1991, 421, 315; (e) K.-T. Smith,
C. Rømming and M. Tilset, J. Am. Chem. Soc., 1993, 115, 8681;
( f ) A. A. Zlota, M. Tilset and K. G. Caulton, Inorg. Chem., 1993,
32, 3816; (g) K.-T. Smith, M. Tilset, R. Kuhlman and K. G.
Caulton, J. Am. Chem. Soc., 1995, 117, 9473; (h) R. J. Klingler,
J. C. Huffman and J. K. Kochi, J. Am. Chem. Soc., 1980, 102, 208;
(i) D. E. Westerberg, L. F. Rhodes, J. Edwin, W. E. Geiger and
K. G. Caulton, Inorg. Chem., 1991, 30, 1107; (j) M. T. Costello and
R. A. Walton, Inorg. Chem., 1988, 27, 2563.
982
J. Chem. Soc., Dalton Trans., 2002, 975–982