Does the rate of competing isomerisation during alkene metathesis depend on pre-catalyst initiation rate?

Article abstract of DOI:10.1039/c4dt00007b

Experimental studies of the ring-closing metathesis reaction of 1,8-nonadiene and the ROMP reaction of cycloheptene show that the rate of isomerisation is not correlated to the initiation rate of the pre-catalyst, and that the absence of phosphine leads to a greatly increased rate of isomerisation. A range of pre-catalysts and solvents were probed and it is proposed that the isomerisation is mediated by a ruthenium hydride complex; our results are consistent with the rate-determining formation of such a species, which might be trapped in situ by tricyclohexylphosphane.

Full text of DOI:10.1039/c4dt00007b

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Does the rate of competing isomerisation
during alkene metathesis depend on pre-catalyst
initiation rate?†
Cite this: Dalton Trans., 2014, 43,
4674
David J. Nelson* and Jonathan M. Percy*
Experimental studies of the ring-closing metathesis reaction of 1,8-nonadiene and the ROMP reaction of
cycloheptene show that the rate of isomerisation is not correlated to the initiation rate of the pre-catalyst,
and that the absence of phosphine leads to a greatly increased rate of isomerisation. A range of pre-
catalysts and solvents were probed and it is proposed that the isomerisation is mediated by a ruthenium
hydride complex; our results are consistent with the rate-determining formation of such a species, which
might be trapped in situ by tricyclohexylphosphane.
Received 2nd January 2014,
Accepted 27th January 2014
DOI: 10.1039/c4dt00007b
www.rsc.org/dalton
target. In most cases, these occur by isomerising terminal
alkenes to internal alkenes,6 or by bringing alkene functional-
Introduction
ity into conjugation with other functional groups.7 Various
researchers have approached this problem, seeking to under-
stand its origin and provide means by which it can be sup-
pressed. Others have exploited this process in synthetic
applications.8 Wagener has explored these competing reac-
tions,9 and implicated ruthenium hydride species in these pro-
cesses via deuterium labelling experiments.10 Nolan and
Prunet proposed a mechanism based on C–H abstraction in
the propagating carbene,11 while researchers at Sasol have pro-
posed and detected thermal decomposition of ruthenacyclo-
butanes via allylruthenium species.12,13 Other researchers have
identified and characterised various species, such as dinuclear
1 and [RuCl(H)(CO)(PCy3)(SIMes)] (2), from the decomposition
of ruthenium-based pre-catalysts under various conditions;14–19
these have been implicated in isomerisation processes, and
reagents such as benzoquinone (inter alia) have been proposed
to remove such species efficiently from reaction mixtures.20
Nolan and co-workers have characterised the product of an
interesting indenylidene to η5-indenyl rearrangement,21 and
shown that (with heating) the resulting complex can isomerise
alkenes.22 We have studied and compared three key mecha-
nisms using in silico methods.23 It was discovered that the
ruthenium hydride pathway, based on a series of hydro-
ruthenation/β-hydride elimination steps from [RuCl(H)(CO)-
(SIMes)] (3), constituted the lowest energy pathway (Scheme 1).
The presence of [RuCl(H)(CO)(PCy3)(SIMes)] (2), formed in situ,
was observed during kinetic studies of the RCM of 1,8-
nonadiene, indicating strongly that this complex is a potential
isomerisation pre-catalyst. Notably, the highest barrier en-
countered was the barrier to phosphine dissociation from the
16e complex to yield an active 14e species.
Alkene metathesis has become a staple technique in organic
chemistry,1 allowing in particular the elegant syntheses of a
number of cycloalkenes in natural product chemistry.2,3 These
advances have largely been stimulated by the development of
user-friendly ruthenium-based pre-catalysts, initially com-
plexes such as G1, and subsequently complexes including G2
and GH2 which bear N-heterocyclic carbene (NHC) ligands.4,5
However, competitive alkene isomerisation can compromise
the effectiveness of this technique, via the non-productive con-
sumption of valuable starting materials and the production of
side products that can be very difficult to separate from the
WestCHEM/Department of Pure & Applied Chemistry, University of Strathclyde,
295 Cathedral Street, Glasgow, G1 1XL, UK. E-mail: david.nelson@strath.ac.uk,
jonathan.percy@strath.ac.uk
†Electronic supplementary information (ESI) available: Concentration/time plots
for experiments presented in this manuscript, initiation rate data for Zhan1B.
See DOI: 10.1039/c4dt00007b
4674 | Dalton Trans., 2014, 43, 4674–4679
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
at ca. 10−1 mol L−1, a mixture of cycloheptene, 1,8-cyclotetra-
decadiene, and higher oligomers are produced.23,26,27 Notably,
under the mild conditions used for these experiments, isomeri-
sation processes are responsible for the consumption of a
considerable proportion of the initial substrate charge (typi-
cally 10–50% in a matter of hours), and therefore comprise a
major pathway in these reactions.
Metathesis reactions under the conditions selected here
(0.5 mol L−1 1,8-nonadiene, 3 mol% pre-catalyst, 298 K) occur
on a timescale that is convenient for the collection of detailed
concentration/time profiles by 1H NMR spectroscopy. Ruthenium
carbene species can be tracked by inspection and integration of
Scheme 1
1
the low field region of the H NMR spectrum (ca. 16–20 ppm).
Very recently, Fogg and co-workers have conducted interest-
ing benchmarking studies, evaluating the isomerisation
activity of complexes 1 and 2 (inter alia).24 They concluded that
neither of these species were sufficiently active to account for
the observed levels of isomerisation in a model reaction;
however, [RuCl(H)(CO)(PPh3)3], which bears more labile phos-
phine ligands, was found to be a highly active isomerisation
catalyst. In addition, added PCy3 was found to inhibit the iso-
merisation activity.
Our working hypothesis at this stage is that 14e [RuCl(H)-
(CO)(SIMes)] 3 is formed at some stage during the reaction, by
a mechanism that is yet to be fully established; i.e. the 14e
species is generated directly, and only manifests as 16e 2 when
trapped by tricyclohexylphosphane in a subsequent step. A ten-
tative mechanism for the formation of species such as 2 has
been proposed by Dinger and Mol,18 but this has not yet been
supported with computational or experimental work. Our
hypothesis might then explain both the detection of 2 by 1H
NMR spectroscopy and the measured sluggish reactivity of 2 as
an isomerisation catalyst. In Fogg’s experiments the entire
stock of 2 that is charged to the reaction must first undergo an
initiation process to which there is a significant energetic
barrier.23 Once formed, 3 might be trapped by tricyclohexyl-
phosphane, which dissociates from 2 slowly to reform the
active species 3, or can directly react with alkene present in the
reaction mixture. Under typical reaction conditions (i.e. ≤ ca.
5 mol% levels of pre-catalyst), the concentration of alkene will
exceed that of phosphine significantly.
The relatively low temperature avoids thermally-triggered
decomposition of the ruthenium species. 1,3,5-Trimethoxyben-
zene was added as an internal standard.27 Oligomer concen-
trations cannot be accurately quantified, as these species
comprise a mixture of linear and cyclic species of various
lengths.26 Their signals cannot be deconvoluted from the
signals corresponding to 1,8-nonadiene using 1D 1H spectra
1
(the 1D [1H, H] geTOCSY technique has been used effectively
in this role). All solvents were dried over activated 4 Å mole-
cular sieves, and degassed by sparging with inert gas (nitrogen
or argon). The water content was determined by Karl-Fischer
titrometry to be ≤10 ppm.
The diene concentration for all reactions was ca. 500 mmol
L−1. Plots are typically presented for cycloheptene and cyclo-
hexene (at minimum), for reasons of clarity. For full concen-
tration/time profiles for all components (cycloheptene,
cyclohexene, cyclic dimer, ethene), please see the ESI.† The
remaining mass balance (i.e. the organic material unaccounted
for as cycloalkene, ethene, or cyclic dimer) comprises a
mixture of diene and oligomer.
The concentration of cyclohexene is used to report on the
degree of isomerisation in the reaction. The RCM of 1,7-octa-
diene is known to be fast, complete and irreversible, even at
4 mol L−1 26,27
the assumption is made here that 1,7-nona-
;
diene, the immediate product of 1,8-nonadiene isomerisation,
undergoes RCM much faster than oligomerisation (i.e. this too
has a high EM). Therefore, the relative rates of isomerisation
can be inferred from the relative rates of cyclohexene
production.
We therefore sought to explore various aspects of reaction
conditions in order to begin to elucidate the factors that
control isomerisation processes that occur during metathesis
chemistry.
Pre-catalyst structure
Initially, a series of pre-catalysts was systematically explored,
covering a range of common motifs. Bis(phosphine) complex
G1 is known to initiate rapidly28,29 and its use tends to result
in very little isomerisation.30 The addition of pre-catalyst to a
benzene-d6 solution of 1,8-nonadiene produced a high concen-
Results and discussion
Experimental approach
A series of kinetic reactions were used to elucidate the effects tration of ethene rapidly, along with an equilibrium concen-
of pre-catalyst and reaction conditions on isomerisation pro- tration of cycloheptene (ca. 60 mmol L−1) (Fig. 1). As the
cesses, as kinetic profiles yield considerably more information ethene concentration slowly decreased, the cycloheptene con-
than single-point yield or conversion determinations. The centration increased slightly, suggesting that ethene build-up
RCM reaction of 1,8-nonadiene was selected for this study. and retention was suppressing further conversion. Negligible
Cycloheptene has a thermodynamic effective molarity (EMT)25 quantities of cyclohexene and cyclic dimer were produced.
of ca. 50 mmol L−1; when metathesis reactions are conducted Inspection of the low field region allowed identification and
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 4674–4679 | 4675

View Article Online
Paper
Dalton Transactions
Fig. 1 Concentration/time profile for the RCM of 1,8-nonadiene
(0.5 mol L−1 in benzene-d6) by G1 (3 mol%) at 298 K, showing ethene
(black triangles), cycloheptene (red diamonds), G1 (black open circles),
alkylidene (red open circles) and methylidene (blue open circles).
Fig. 2 Concentration/time profile for the RCM of 1,8-nonadiene
(0.5 mol L−1 in benzene-d6) by G2 (3 mol%) at 298 K, showing (a) ethene
(black triangles), cycloheptene (red diamonds), cyclohexene (green
diamonds), and cyclic dimer (blue diamonds); and (b) G2 (black circles),
alkylidene (red circles) and methylidene (blue circles).
quantification of the various ruthenium species, showing
rapid initiation of the pre-catalyst,29 an initial high concen-
tration of alkylidenes 4a, and a growing population of methyli-
dene 5a which then decreased gradually.
predominantly occurs after the productive metathesis has fin-
ished. Notably, the acceleration in isomerisation rate appeared
to occur as a population of methylidene increased.
Both G1 and G2 bear ‘throw-away’ phosphine ligands,
which have been implicated in the decomposition mechanism
of some metathesis pre-catalysts, via nucleophilic attack of the
14e methylidene intermediate.14 Phosphine-free GH2 and
more rapidly initiating analogues Zhan1B and Grela2 were
therefore tested, to determine if they led to reduced or
increased rates of isomerisation, and whether the pre-catalyst
initiation rate affected the isomerisation rate.
The chemical shift of the alkylidene (RuvCHR) proton was
consistent with that reported for alkylidenes derived from
linear 1-alkenes.31 The alkylidene decay followed excellent first
order kinetics (kobs = 2.78 × 10−4 −1) while no hydride species
s
were detected in the high field region. The mass balance of
ruthenium species is not maintained in these reactions (ca.
20% is lost after 1 h, 33% after 2 h, and 50% after 5 h), but the
final fate of the ruthenium complex is unknown. These results
suggest that either (i) no hydride species form under these
conditions or that (ii) those hydride species that do form are
not active enough to isomerise appreciable quantities of sub-
strate. Fogg has reported the sluggish isomerisation ability of
[RuCl(H)(CO)(PCy3)2],24 although it is unclear whether this is a
consequence of slow initiation or slow isomerisation.
In all three cases, isomerisation was far more rapid than
exhibited by G2 (Fig. 3); within ca. 4 hours, one third of the
starting material had been consumed by isomerisation pro-
cesses. In each case, the equilibrium concentration of cyclo-
heptene was reached rapidly.
Second generation complex G2 was examined next. This
commercially-available complex represents one of the most
commonly-deployed metathesis complexes in organic syn-
thesis. Significant isomerisation was observed in the 1,8-nona-
diene RCM experiment, to the point that cyclohexene
concentration outstripped cycloheptene concentration (Fig. 2).
Quantities of cyclic dimer were also produced. The pre-catalyst
initiation was much slower than for G1 and incomplete even
after 5 h, consistent with what is known about the initiation of
this species.28,29 Alkylidene and methylidene populations
1
could be tracked via the low field region of the H NMR spec-
Fig. 3 Concentration/time profile for the RCM of 1,8-nonadiene
(0.5 mol L−1 in benzene-d6) by GH2 (circles), Zhan1B (triangles) and
Grela2 (diamonds) (3 mol%) at 298 K, showing ethene (black),
cycloheptene (red), and cyclohexene (green); the lines represent
concentration/time profiles for cycloheptene (red) and cyclohexene
trum. Again, significant proportions of the ruthenium popu-
lation are unaccounted for, part of which must comprise
the isomerisation agent. Interestingly, the isomerisation
process accelerated as the reaction reached the equilibrium
concentration of cycloheptene, suggesting that isomerisation (green) from Fig. 2 (using G2).
4676 | Dalton Trans., 2014, 43, 4674–4679
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Clearly, the absence of a phosphine ligand does not pre-
clude isomerisation. Indeed, the lack of phosphine in the reac-
tion mixture may actually serve to prevent capture of the
postulated active 14e hydride species 3 and thus promote the
isomerisation. Computational studies have suggested that,
during the initiation of 2, the phosphine dissociation event (to
yield 3) is highly endergonic (22.9 kcal mol−1).23 Phosphine
dissociation from G2 has been shown to present similar yet
slightly higher barriers; Jensen and co-workers have estimated
this barrier at 23.7 kcal mol−1 using detailed DFT calculations,
while Grubbs and co-workers reported an experimental value
Fig. 4 Concentration/time profile for the RCM of 1,8-nonadiene
(0.5 mol L−1 in benzene-d6) by M853-SIPr (3 mol%) at 298 K, showing
cycloheptene (red diamonds) and cyclohexene (green diamonds); for
comparison, the corresponding profiles with G2 (solid lines, from Fig. 2)
and GH2 (dashed lines, from Fig. 3) are presented on the same scale.
of 23.0 0.4 kcal mol−1
.
The Hoveyda-type catalysts considered here have very
different initiation rate constants, covering a 10-fold range (at
298 K in DCM solution: 0.026 L mol−1 s−1 for GH2, 0.132 L
mol−1 s−1 for Zhan1B, 0.317 L mol−1 s−1 for Grela2); initiation
rates for GH2 and Grela2 have been reported previously,32,33
while the initiation rate of Zhan1B was measured for this work
(see the ESI† for full details). This difference in initiation rate
is reflected in the concentration/time profiles of the pre-cata-
lyst concentrations in these experiments (see the ESI†); the
concentrations of methylidene and propagating carbene
cannot be monitored under these conditions, as these 14e
alkylidene species are not detectable by NMR spectroscopy.34
Unfortunately, this also precluded detection of ruthenium
hydride species and insights into the mass balance of ruthe-
nium species. The fact that isomerisation rate is unrelated to
initiation rate (and therefore the concentration of 14e ruthe-
nium species) is somewhat surprising, and suggests that the
rate is limited by some other species or process.
formation of the SIPr-containing hydride complex, its higher
activity, or a combination of these two factors.
These results suggest strongly that the structure of the iso-
merisation catalyst depends on that of the pre-catalyst; the iso-
merisation agent is not common in all of the experiments
studied herein. The ancillary ligands present on the meta-
thesis pre-catalyst therefore influence those present on the iso-
merisation agent, and in so doing (a) determine the activity of
this agent, and/or (b) the rate of its formation (both in the
order SIPr > SIMes > PCy3). Further work is necessary in order
to separate these two different effects. Banti et al. have shown
that the isomerisation activity of [RuCl(H)(CO)(PCy3)(L)] com-
plexes (in neat 1-octene at 100 °C) decreases in the order
SIMes > PCy3 > SIPr, but that the NHC-bearing complexes are
only active at higher temperatures.19 Fogg has shown that the
SIMes-bearing hydride 2 is more active for allylbenzene iso-
merisation than its phosphine-bearing congener at lower
temperatures (40–80 °C in DCM or toluene).24 For catalysts
with the same non-dissociating ligands (i.e. NHC, two chloride
ligands, and an alkylidene) we have shown here that the
initiation rate has no effect on the rate of the isomerisation
process, despite the demonstrably different concentrations of
ruthenium that have been delivered into the reaction.
Finally, the reactivity of M853-SIPr was explored (initiation
rate constant of 0.571 L mol−1 −1).33 This complex bears the
s
SIPr ligand, rather than SIMes; complexes bearing the SIPr
ligand are frequently more active in the metathesis of less hin-
dered alkenes, but are less efficacious for the synthesis of
tri- and tetra-substituted alkenes.35–37
Solvent effects
Further experiments were conducted to explore the effect of
reaction conditions on isomerisation activity. In particular the
effect of solvent was probed, as this has been shown to influ-
ence the rate of metathesis,27,38,39 although this typically
occurs predominantly via changes to the pre-catalyst initiation
When this pre-catalyst was tested in the RCM of 1,8-nona- rate.40,41
diene, significantly more isomerisation occurred, consuming
Reactions were conducted with G2 in chloroform-d, dichloro-
half of the starting material in less than 2 hours, leading to methane-d2 and toluene-d8. There were clear solvent effects
consumption of some of the cycloheptene formed initially on the rate of isomerisation, with toluene leading to far less
(Fig. 4). While this complex is known to initiate very rapidly, isomerisation than dichloromethane (Fig. 5).
this alone cannot explain the increased rate of isomerisation,
Water is unlikely to be the cause of this difference in reac-
as GH2, Zhan1B and Grela2 all lead to the same rate of isomeri- tivity; not only is metathesis under aqueous conditions well
sation (vide supra). The difference must therefore be due to documented,42 but the use of undried chloroform-d (contain-
the different NHC ligand present on the pre-catalyst (i.e. SIPr ing ca. 30 ppm water, as determined by Karl Fischer titro-
versus SIMes); this difference must be a result of either faster metry) led to only a slight reduction in metathesis rate but a
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 4674–4679 | 4677

View Article Online
Paper
Dalton Transactions
catalyst and alkylidene was excellent, with only ca. 5% of the
ruthenium population unaccounted for after ca. 4 hours (see
the ESI†).
This reaction was repeated and ca. 50 mmol L−1 1,7-octa-
diene added after 15 minutes; this substrate undergoes almost
instantaneous, complete and irreversible RCM to yield cyclo-
hexene plus ethene.27,43 Methylidene was now generated by
diene turnover, appearing in the low field region of the 1H
NMR spectrum; mass balance in the alkylidene region was not
reached, and isomerisation now began (see the ESI†). In
addition, very little ethene was observed, consistent with its
consumption by reaction with the alkylidene to form diene. It
seems likely that the key precursor to an active isomerisation
catalyst is indeed the methylidene complex produced upon
diene turnover in RCM.
Fig. 5 Concentration/time profiles showing (a) cycloheptene and
(b) cyclohexene from the RCM of 1,8-nonadiene (0.5 mol L−1 at 298 K
with 3 mol% G2) in benzene-d6 (green), chloroform-d (blue), dichloro-
methane-d2 (black) and toluene-d8 (red).
significant reduction in isomerisation rate (see the ESI†). In
addition, the relative rates of isomerisation (DCM > benzene ≈
chloroform > toluene) do not mirror the relative rates of
initiation (toluene > benzene > DCM > chloroform),41 support-
ing the view that this difference is not simply a consequence of
the rate of formation of the 14e metathesis-active species. This
is consistent with the results reported above for the pre-cata-
lysts which have different initiation rates but converge on the
same active species after one turnover.
Conclusions
The progress of a model ring-closing metathesis reaction
under mild conditions has been monitored using 1H NMR
kinetic experiments. The degree of isomerisation has been
found to depend on the nature of the propagating species, i.e.
the ligand retained by the ruthenium throughout the meta-
thesis cycle. G1 leads to no isomerisation, while all NHC-
bearing pre-catalysts produce cyclohexene as a side product.
Hoveyda-type complexes (GH2, Zhan1B and Grela) produce
cyclohexene more rapidly than G2, despite the active species
being the same, yet at a rate that was independent of the pre-
catalyst initiation rate. M853-SIPr generated cyclohexene faster
than SIMes-bearing complexes. These results suggest an iso-
merisation process mediated by a ruthenium complex that
forms in situ and is not part of the metathesis cycle; if isomeri-
sation was a result of side-reactions occurring during the RCM
reaction (such as those proposed by Nolan and Prunet,11 or
researchers at Sasol)12,13 then more rapid initiation would lead
to more rapid isomerisation due to the higher population of
ruthenium species in the cycle when initiation is rapid. Simi-
larly, while solvent effects were evident, these were not the
same as those on the rates of pre-catalyst initiation.41
RCM versus ROMP
Methylidene complexes, generated in the presence of ethene
or when terminal alkenes undergo metathesis, have been
implicated as the most fragile ruthenium alkylidenes, and the
most likely sources of isomerisation-active decomposition
agents.15 The ROMP of cycloheptene was therefore probed, as
this avoids the production of methylidene complexes.
When a sample of cycloheptene (500 mmol L−1 in benzene-
d6) was exposed to G2 (3 mol%) under the same conditions as
the RCM reactions, ROMP was rapid, leading to equilibrium
concentrations of cycloheptene (ca. 90 mmol L−1) and cyclic
dimer (ca. 45 mmol L−1) (Fig. 6). Isomerisation processes
occurred, but at a much lower rate. Inspection of the alkyli-
dene region revealed no methylidene (5b), smooth consump-
tion of the pre-catalyst, and production of the ruthenium
alkylidene intermediate (4b). The mass balance between pre-
Experiments with the corresponding ROMP reaction (of
cycloheptene) under similar conditions resulted in very little
isomerisation until a portion of 1,7-octadiene was added; iso-
merisation processes then occurred much more quickly.
These results suggest strongly that the active isomerisation
agent is generated from the 14e methylidene species. In the
presence of phosphine, which can trap both 14e methylidene
and 14e hydride, the isomerisation might be somewhat sup-
pressed, explaining the higher isomerisation rates in the pres-
ence of the Hoveyda-type complexes. The phosphine
dissociation events from methylidene complex [RuCl2(PCy3)-
(SIMes)(vCH2)] (5)29 and hydride complex [RuCl(H)(CO)(PCy3)-
(SIMes)] (2)23 are known to be slow; a rate could not be
measured for the former. However, the lack of dependence of
isomerisation rate on initiation rate suggests that the
Fig. 6 Concentration/time profile for the ROMP of cycloheptene
(0.5 mol L−1 in benzene-d6) by G2 (3 mol%) at 298 K, showing cyclo-
heptene (red), cyclohexene (green), and cyclic dimer (blue).
4678 | Dalton Trans., 2014, 43, 4674–4679
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
formation of the active hydridocarbonyl species may be rate- 18 M. B. Dinger and J. C. Mol, Organometallics, 2003, 22,
determining for isomerisation. 1089–1095.
Further experimental and theoretical studies into the mech- 19 D. Banti and J. C. Mol, J. Organomet. Chem., 2004, 689,
anism of formation of the active ruthenium hydride are cur-
rently underway in our laboratories.
3113–3116.
20 S. H. Hong, D. P. Sanders, C. W. Lee and R. H. Grubbs,
J. Am. Chem. Soc., 2005, 127, 17160–17161.
21 S. Manzini, C. A. Urbina-Blanco, A. Poater, A. M. Z. Slawin,
L. Cavallo and S. P. Nolan, Angew. Chem., Int. Ed., 2012,
124, 1066–1069.
22 S. Manzini, D. J. Nelson and S. P. Nolan, ChemCatChem,
2013, 5, 2848–2851.
23 I. W. Ashworth, I. H. Hillier, D. J. Nelson, J. M. Percy and
M. A. Vincent, Eur. J. Org. Chem., 2012, 5673–5677.
24 C. S. Higman, L. Plais and D. E. Fogg, ChemCatChem, 2013,
5, 3548–3551.
25 A. J. Kirby, Adv. Phys. Org. Chem., 1980, 17, 183–278.
26 D. J. Nelson, I. W. Ashworth, I. H. Hillier, S. H. Kyne,
S. Pandian, J. A. Parkinson, J. M. Percy, G. Rinaudo and
M. A. Vincent, Chem.–Eur. J., 2011, 17, 13087–13094.
27 I. W. Ashworth, D. Carboni, I. H. Hillier, D. J. Nelson,
J. M. Percy, G. Rinaudo and M. A. Vincent, Chem. Commun.,
2010, 46, 7145–7147.
Acknowledgements
We thank Mr Craig Irving and Dr John Parkinson for assist-
ance with NMR experiments, and Mr Gavin Bain for conduct-
ing Karl Fisher water content measurements. AstraZeneca
(Industrial CASE studentship to D.J.N.) is thanked for funding.
Dr Ian Ashworth (AstraZeneca) is thanked for helpful discus-
sions. Prof. Marc Mauduit and Omega Cat System are thanked
for a sample of complex M853-SIPr.
Notes and references
1 A. H. Hoveyda and A. R. Zhugralin, Nature, 2007, 450,
243–251.
2 K. C. Nicolaou, P. G. Bulger and D. Sarlah, Angew. Chem.,
Int. Ed., 2005, 44, 4490–4527.
28 M. S. Sanford, M. Ulman and R. H. Grubbs, J. Am. Chem.
Soc., 2001, 123, 749–750.
29 M. S. Sanford, J. A. Love and R. H. Grubbs, J. Am. Chem.
Soc., 2001, 123, 6543–6554.
3 J. Prunet, Eur. J. Org. Chem., 2011, 3634–3647.
4 G. Vougioukalakis and R. Grubbs, Chem.–Eur. J., 2008, 14, 30 K. Mori, Tetrahedron, 2009, 65, 3900–3909.
7545–7556.
31 D. R. Lane, C. M. Beavers, M. M. Olmstead and
5 C. Samojlowicz, M. Bieniek and K. Grela, Chem. Rev., 2009,
109, 3708–3742.
6 G. De Bo and I. E. Markó, Eur. J. Org. Chem., 2011, 1859–
1869.
7 A. E. Sutton, B. A. Seigal, D. F. Finnegan and M. L. Snapper,
J. Am. Chem. Soc., 2002, 124, 13390–13391.
N. E. Schore, Organometallics, 2009, 28, 6789–6797.
32 I. W. Ashworth, I. H. Hillier, D. J. Nelson, J. M. Percy and
M. A. Vincent, Chem. Commun., 2011, 47, 5428–5430.
33 D. J. Nelson, P. Queval, M. Rouen, M. Magrez, L. Toupet,
F. Caijo, E. Borré, I. Laurent, C. Crévisy, O. Baslé,
M. Mauduit and J. M. Percy, ACS Catal., 2013, 3, 259–264.
8 T. J. Donohoe, T. J. C. O’Riordan and C. P. Rosa, Angew. 34 E. F. van der Eide and W. E. Piers, Nat. Chem., 2010, 2,
Chem., Int. Ed., 2009, 48, 1014–1017. 571–576.
9 S. E. Lehman, J. E. Schwendeman, P. M. O’Donnell and 35 T. Ritter, A. Hejl, A. G. Wenzel, T. W. Funk and
K. B. Wagener, Inorg. Chim. Acta, 2003, 345, 190–198. R. H. Grubbs, Organometallics, 2006, 25, 5740–5745.
10 F. C. Courchay, J. C. Sworen, I. Ghiviriga, K. A. Abboud and 36 H. Clavier, C. A. Urbina-Blanco and S. P. Nolan, Organo-
K. B. Wagener, Organometallics, 2006, 25, 6074–6086. metallics, 2009, 28, 2848–2854.
11 D. Bourgeois, A. Pancrazi, S. P. Nolan and J. Prunet, 37 C. A. Urbina-Blanco, X. Bantreil, J. Wappel, T. E. Schmid,
J. Organomet. Chem., 2002, 643–644, 247–252.
12 W. Janse van Rensburg, P. J. Steynberg, W. H. Meyer,
A. M. Z. Slawin, C. Slugovc and C. S. J. Cazin, Organo-
metallics, 2013, 32, 6240–6247.
M. M. Kirk and G. S. Forman, J. Am. Chem. Soc., 2004, 126, 38 C. S. Adjiman, A. J. Clarke, G. Cooper and P. C. Taylor,
14332–14333. Chem. Commun., 2008, 2806–2808.
13 W. J. van Rensburg, P. J. Steynberg, M. M. Kirk, 39 M. D. Schulz and K. B. Wagener, ACS Macro Lett., 2012, 1,
W. H. Meyer and G. S. Forman, J. Organomet. Chem., 2006,
691, 5312–5325.
14 S. H. Hong, M. W. Day and R. H. Grubbs, J. Am. Chem. Soc.,
2004, 126, 7414–7415.
15 S. H. Hong, A. G. Wenzel, T. T. Salguero, M. W. Day and
R. H. Grubbs, J. Am. Chem. Soc., 2007, 129, 7961–7968.
16 S. H. Hong, A. Chlenov, M. W. Day and R. H. Grubbs,
Angew. Chem., Int. Ed., 2007, 46, 5148–5151.
17 M. B. Dinger and J. C. Mol, Eur. J. Inorg. Chem., 2003, 2003,
2827–2833.
449–451.
40 D. J. Nelson, D. Carboni, I. W. Ashworth and J. M. Percy,
J. Org. Chem., 2011, 76, 8386–8393.
41 I. W. Ashworth, D. J. Nelson and J. M. Percy, Dalton Trans.,
2013, 42, 4110–4113.
42 D. Burtscher and K. Grela, Angew. Chem., Int. Ed., 2009, 48,
442–454.
43 S. Pandian, I. H. Hillier, M. A. Vincent, N. A. Burton,
I. W. Ashworth, D. J. Nelson, J. M. Percy and G. Rinaudo,
Chem. Phys. Lett., 2009, 476, 37–40.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 4674–4679 | 4679