Reactions of Grubbs catalysts with excess methoxide: Formation of novel methoxyhydride complexes

Article abstract of DOI:10.1021/om201288p

On exposure to NaOMe (≥3 equiv) in CH₂Cl₂-MeOH at 23 °C, the first-generation Grubbs catalyst RuCl₂(PCy ₃)₂(=CHPh) (1a) is immediately transformed into the six-coordinate methoxyhydride complexes RuH(OMe)(CO)₂(PCy ₃)₂ (4a) and RuH(OMe)(CO)(H₂)(PCy ₃)₂ (5a). Complex 5a can be recycled into 4a under conditions conducive to removal of H₂. The second-generation catalyst RuCl₂(IMes)(PCy₃)(=CHPh) (1b; IMes = N,N′- bis(mesityl)imidazol-2-ylidene) reacts more slowly, requiring several hours even at 20 equiv of NaOMe, and terminates at five-coordinate RuH(OMe)(CO)(IMes) (PCy₃) (3b). Experiments in the presence of added PCy₃ reveal that consumption of 1a, but not 1b, proceeds via the four-coordinate intermediate formed by equilibrium loss of phosphine, a function of the lability of the PCy₃ ligand at ambient temperatures. The poor accessibility of such an intermediate for 1b at 23 °C retards salt metathesis and inhibits further reaction of 3b. For the bis(PCy₃) analogue 3a, fast transformation into 4a is proposed to involve reversible loss of PCy ₃, coordination of methanol, σ-metathesis of methanol at the hydride site to liberate H₂, and β-elimination/decarbonylation of bound methoxide. Competitive uptake of H₂ by 3a yields six-coordinate 5a (the dihydrogen adduct of 3a). Independent routes to RuH(OMe)(CO)₂(L)(PCy₃) (4a/b; a, L = PCy₃; b, L = IMes) were developed: these involved sequential transformation of RuHCl(CO)(L)(PCy₃) (2a/b) into the bis-carbonyl adducts RuHCl(CO)₂(L)(PCy₃) (7a/b) under CO, conversion of 7a/b into the more reactive triflates RuH(OTf)(CO)₂(L)(PCy₃) (8a/b), and reaction of 8a/b with equimolar NaOMe. Dihydride 6b was also prepared, by reaction of 8b with NaH.

Full text of DOI:10.1021/om201288p

Article
pubs.acs.org/Organometallics
Reactions of Grubbs Catalysts with Excess Methoxide: Formation of
Novel Methoxyhydride Complexes
Nicholas J. Beach, Justin A. M. Lummiss, Jennifer M. Bates, and Deryn E. Fogg*
Centre for Catalysis Research and Innovation and Department of Chemistry, University of Ottawa, Ottawa, Ontario, Canada K1N
6N5
S
* Supporting Information
ABSTRACT: On exposure to NaOMe (≥3 equiv) in CH2Cl2−MeOH at 23
°C, the first-generation Grubbs catalyst RuCl2(PCy3)2(CHPh) (1a) is
immediately transformed into the six-coordinate methoxyhydride complexes
RuH(OMe)(CO)2(PCy3)2 (4a) and RuH(OMe)(CO)(H2)(PCy3)2 (5a).
Complex 5a can be recycled into 4a under conditions conducive to removal
of H2. The second-generation catalyst RuCl2(IMes)(PCy3)(CHPh) (1b;
IMes = N,N′-bis(mesityl)imidazol-2-ylidene) reacts more slowly, requiring
several hours even at 20 equiv of NaOMe, and terminates at five-coordinate RuH(OMe)(CO)(IMes)(PCy3) (3b). Experiments
in the presence of added PCy3 reveal that consumption of 1a, but not 1b, proceeds via the four-coordinate intermediate formed
by equilibrium loss of phosphine, a function of the lability of the PCy3 ligand at ambient temperatures. The poor accessibility of
such an intermediate for 1b at 23 °C retards salt metathesis and inhibits further reaction of 3b. For the bis(PCy3) analogue 3a,
fast transformation into 4a is proposed to involve reversible loss of PCy3, coordination of methanol, σ-metathesis of methanol at
the hydride site to liberate H2, and β-elimination/decarbonylation of bound methoxide. Competitive uptake of H2 by 3a yields
six-coordinate 5a (the dihydrogen adduct of 3a). Independent routes to RuH(OMe)(CO)2(L)(PCy3) (4a/b; a, L = PCy3; b, L =
IMes) were developed: these involved sequential transformation of RuHCl(CO)(L)(PCy3) (2a/b) into the bis-carbonyl adducts
RuHCl(CO)2(L)(PCy3) (7a/b) under CO, conversion of 7a/b into the more reactive triflates RuH(OTf)(CO)2(L)(PCy3) (8a/
b), and reaction of 8a/b with equimolar NaOMe. Dihydride 6b was also prepared, by reaction of 8b with NaH.
carbonyl adduct 2a, among other Ru species (ca. 50% 2a; we
suspected that the unidentified coproducts might arise from
thermolytic degradation of 1a). Given that complexes 2a/b are
high-productivity catalysts in the mechanistically related context
of olefin hydrogenation,7 we hypothesized that the poor
isomerization activity associated with primary alcohols might be
due to formation of deactivated polycarbonyl species in the
presence of excess alkoxide.
Within part of a broader program of study directed at
evaluating the behavior of Ru-alkylidene complexes with
reactive [ER]−/HER species, we thus wished to clarify the
reactivity of 1a and 1b toward excess methoxide (>2 equiv) in
the presence of methanol. We chose to explore this chemistry
at ambient temperatures, both to intercept the relevant Ru
species (rather than “downstream” products arising from
extraneous thermal decomposition pathways) and to gain a
clearer picture of the minimum conditions required for loss of
the benzylidene functionality. Here we report strikingly
different outcomes for the first- and second-generation Grubbs
catalysts. While 1a is immediately converted into RuH(OMe)-
(CO)2(PCy3)2 (4a) and RuH(OMe)(CO)(H2)(PCy3)2 (5a) at
23 °C, complex 1b undergoes much slower transformation into
a rare, unexpectedly stable five-coordinate methoxyhydride
species, RuH(OMe)(CO)(IMes)(PCy3) (3b), with no sign of
INTRODUCTION
■
The Grubbs metathesis catalysts RuCl2(L)(PCy3)(CHPh)
(e.g., 1a: L = PCy3; 1b: L = IMes, N,N′-bis(mesityl)imidazol-2-
ylidene) are now widely recognized as powerful agents for
further, nonmetathetical transformations.1,2 In some cases, this
expanded reactivity is due to the action of 1 and its alkylidene
and/or methylidene derivatives: in others, it reflects the
operation of further ruthenium catalysts generated in situ,
whether by chance or design.3 While a greater understanding of
the underlying inorganic transformations of these important
complexes would clearly be advantageous, overwhelming
interest in their organic applications has tended to overshadow
their inorganic reaction chemistry. Many of their fundamental
reactivity patterns thus remain obscure, studies of the
transformation of 1 into the catalytically important hydrides
RuHCl(CO)(L)(PCy3) (2) (particularly 2a; L = PCy3)
notwithstanding.4,5
A potentially important example of this obscurity is found in
the reactions of the Grubbs catalysts with alkoxides. Schmidt
has demonstrated that postmetathesis treatment of 1a with
isopropanol and NaOH (5−10 equiv vs Ru) yields efficient
catalysts for double-bond isomerization. Primary alcohols,
however, were strikingly less effective.6 While metal hydrides
are formed in either case, one obvious difference lies in the
resistance of secondary alkoxides to Ru-mediated decarbon-
ylation. In contrast, Mol and co-workers4a have shown that
treating 1a with 1 equiv of methoxide at 70−75 °C yields the
Received: December 31, 2011
Published: March 13, 2012
© 2012 American Chemical Society
2349
dx.doi.org/10.1021/om201288p | Organometallics 2012, 31, 2349−2356

Organometallics
Article
a
Scheme 1. Products Observed Following Reaction of Grubbs Catalysts with Excess Methoxide
a
Reactions at 23 °C in CH2Cl2−MeOH. PCy3, toluene, and formaldehyde also observed; see text. The trans-disposition of the hydride and CO
ligands in 5a (favored by donor−acceptor push−pull interactions) is suggested by analogy to the structure of 4a.
the expected 4b/5b. Routes to the new complexes 3b and 4a/b
from convenient hydride precursors were developed to confirm
the identities of observed products (3b, 4a) and the absence of
expected products (4b).
Scheme 2. Proposed Mechanisms for Transformation of 1a/
b into Five-Coordinate 3a/b
a
RESULTS AND DISCUSSION
■
Reactions of Grubbs Catalysts with Methoxide Ion.
Addition of excess methanolic NaOMe to a solution of 1a in
CH2Cl2 (Scheme 1a) caused an instant color change from
purple to pale yellow, evolution of small bubbles presumed to
be H2, and complete loss of the NMR signals for 1a within 5
min, without need for elevated temperatures. No evidence was
seen of methoxyhydride RuH(OMe)(CO)(PCy3)2 (3a), an
intermediate inferred from the corresponding, slower reaction
of 1b discussed below (Scheme 1b). Instead, the sole Ru
products observed were RuH(OMe)(CO)2(PCy3)2 (4a) and
RuH(OMe)(CO)(H2)(PCy3)2 (5a), formed in ca. 2:1 ratio.
Formaldehyde and free PCy3 were also detected (ca. 10%
each), as well as toluene, although quantification of the latter
was hampered by interfering signals from the cyclohexyl groups
or the internal standard Ph3PO. Co-formation of ca. 25%
paramagnetic material is indicated by integration against
Ph3PO, vide infra.
The distribution of Ru products did not change on longer
reaction (4 h), but stripping off the solvent and redissolving the
residue in C6D6 effected transformation of 5a into 4a, via
uptake of a further equivalent of methanol (see mechanistic
section). Traces of known8 dihydride RuH2(CO)2(PCy3)2 (6a)
(<5%) were also formed. The identity of 4a was confirmed by
independent synthesis from hydride precursors; dihydrogen
adduct 5a was characterized by in situ NMR analysis, including
T1(min) measurements. Facile β-elimination from methoxide,
discussed in more detail below, disfavors formation of stable
bis(alkoxide) or deprotonated carbyne products, as formed in
the corresponding reactions of 1a with excess tert-butoxide9 or
phenoxides.10−12
Consumption of 1b, in comparison, required several hours
even at relatively high proportions of methoxide (4 h at 20
equiv). As shown in Scheme 1b, reaction also halted at an
earlier stage, yielding five-coordinate RuH(OMe)(CO)(IMes)-
(PCy3) (3b) (a rare example of a coordinatively unsaturated
alkoxyhydride complex),13 rather than 4b and/or 5b. Addition
of excess PCy3 had no effect on the rate of transformation of 1b
into 3b, although such treatment retarded consumption of the
first-generation catalyst 1a. We attribute the difference to the
much higher lability of the PCy3 ligand in 1a, which gives
equilibrium access to four-coordinate RuCl2(PCy3)(CHPh)
(1a′) in solution. Salt metathesis via this sterically accessible
species (see path II, Scheme 2) is sufficiently faster that the
corresponding reaction of the parent 1a (path I) does not
compete. The slow salt metathesis of 1b, as well as the
resistance of 3b to further reaction, is consistent with the very
a
Product is isolable only for 3b (L = IMes). For ensuing reactions of
3a/3a′, see text. For L = PCy3, both of the pathways depicted are
feasible, but path II is kinetically dominant. For L = IMes, only path I
is available, and reaction is slow.
low room-temperature lability of the PCy3 ligand in these IMes
complexes,16 which significantly increases the steric impedi-
ment to reaction. The strong binding characteristic of
phosphine ligands trans to an NHC ligand is known to retard
dissociative reaction pathways in catalysis, particularly at
ambient temperatures. Such behavior has been documented
in hydrogenation and isomerization via the hydride complex
2b14,15 and in metathesis via 1b.16,17
A proposed mechanism for the transformation of 1 into 3
thus involves initial reaction of 1 via exchange of both chloride
ligands, followed by proton transfer from bound methoxide to
benzylidene (see A/A′; Scheme 2), and coordination of
formaldehyde to give Ru-benzyl intermediate B/B′. Deinsertion
of CO and liberation of the benzyl group as toluene (and
recoordination of PCy3, in the case of 3a′) would then generate
3a/b. Labeling evidence consistent with such a pathway was
reported by Dinger and Mol4a in the transformation of 1a into
2a with a single equivalent of methoxide. More recently, Leung
and co-workers invoked the analogous attack by bound
methoxide on benzylidene on treatment of RuCl[N(iPr2PS)2]-
(PCy3)(CHPh) with NaOMe.18 These reactions fall into a
broader class of proton migrations onto benzylidene from E-R
groups on an adjacent ligand19 (notwithstanding the moder-
ately electrophilic character inferred computationally for Ru
CH2 systems).20 Migration need not be restricted to β- or α-E-
H groups: Owen and co-workers recently reported attack on
2350
dx.doi.org/10.1021/om201288p | Organometallics 2012, 31, 2349−2356

Organometallics
Article
benzylidene by a γ-dihydroborate moiety in a scorpionate
derivative of 1a,21 underscoring the point that ligand flexibility,
in conjunction with a reactive E−H bond, can expand the scope
of this alkylidene transformation pathway. Ru-NHC deriva-
tives22,23 show substantially higher tolerance toward adjacent
N−H functionalities than first-generation complexes,24 perhaps
indicating that migration proceeds via a dissociative pathway
(vide infra).
Conversion of the intermediate 3a (or, more probably, 3a′)
into methoxydicarbonyl 4a necessitates installation of a further
equivalent of methoxide via reaction with methanol. The fact
that 3b does not evolve indicates that the associative pathway is
not productive in this case. Scheme 3 depicts a plausible
(although this will be attenuated by the cis-hydride effect),26
which circumvents the requirement for phosphine loss found in
the present chemistry. Alternatively, an associative pathway
involving outer-sphere, σ-bond metathesis of methanol may be
enabled by the acidity of bound H2.27 Either is consistent with
the reported [PCy3]-independence of the reaction.5
Paramagnetic Byproducts. A common challenge in
organometallic chemistry is the formation of paramagnetic
products that go undetected by direct NMR methods. In the
work above, observation of free PCy3 led us to suspect the
presence of such coproducts. We quantified the proportion of
paramagnetic Ru by integration against Ph3PO as an internal
standard. A ca. 25% decrease in total integration was evident for
the reactions of both 1a and 1b: we attribute this to the
formation of paramagnetic species, rather than fluxional
diamagnetic products, as low-temperature 31P{1H} NMR
analysis revealed no additional signals. Notably, however, 4a
can be prepared in 85% isolated yield from a nonbenzylidene
precursor (see below), tending to implicate the benzylidene
functionality in the Ru(II) to Ru(III) oxidation. It is unclear
whether C−H activation of the PCy3 and/or IMes ligands also
contributes: both are well precedented.28,29 The four-
coordinate species generated by phosphine loss is almost
certainly a key vector for decomposition, as suggested by
parallel experiments with the labile, phosphine-free “third-
generation” Grubbs catalyst RuCl2(IMes)(py)2(CHPh).
This complex was completely consumed within 15 min of
treating with methoxide, but only ca. 15% of a hydride product
was observed, which itself disappears within 1 h.
Scheme 3. Proposed Mechanism for Transformation of 3a
into Six-Coordinate 4a and 5a
Synthetic Routes to Novel Complexes. Preparation of
Five-Coordinate Methoxyhydride RuH(OMe)(CO)(IMes)-
(PCy3) (3b). To confirm the identity of 3b, we undertook its
synthesis on a preparative scale from the well-behaved hydride
precursor 2b. Addition of methanolic NaOMe (20 equiv;
Scheme 4) to a solution of 2b in CH2Cl2 at 23 °C caused an
pathway for 3a, involving coordination of methanol and σ-
metathesis25 of the O−H and Ru−hydride bonds, followed by
β-hydride elimination and CO deinsertion as before. Scaveng-
ing of H2 by unreacted 3a would account for the competitive
formation of dihydrogen adduct 5a. Reversible binding of
dihydrogen enables release and recycling of 3a, as indicated by
the conversion of 5a into 4a on workup noted above. Transient
signals assigned to 3a are observed by in situ NMR analysis, at
shifts very similar to those for 3b, on freeze−pump−thaw
degassing solutions of 5a. Hydrogen-bonding interactions
between incoming methanol and the hydride and/or methoxide
ligands may account for the dramatically faster reaction of
methanol with 3a, vs the benzylidene complex 1a (the latter
reaction requires hours even at 60 °C).5
An anticipated competing pathway involves direct β-
elimination and deinsertion from 3a/b to afford dihydrides
RuH2(CO)2(L)(PCy3)2 (6a/b). While traces of 6a and 6b
form on workup, this pathway is clearly a minor one. The
minimal formation of 6a is probably due to the rapid
conversion of 3a into 4a and 5a, reflecting the abundance
and noninnocence of the methanol cosolvent. Importantly,
however, the resistance of 3b to transformation into 6b implies
that β-elimination again requires phosphine loss, despite the
formal coordinative unsaturation of 3b.
Scheme 4. Synthesis of Five-Coordinate Methoxyhydride
Complex 3b
immediate color change from orange-yellow to red-orange.
NMR analysis revealed complete conversion to the new
methoxyhydride species RuH(OMe)(CO)(IMes)(PCy3) (3b)
within 15 min. Quantitative conversion was confirmed in
separate NMR-scale experiments by integration against Ph3PO
as internal standard, as noted above. Isolation of 3b was
frustrated by formation of traces of 4b and 6b on concentrating
to dryness, but detailed NMR analysis supports the proposed
structure. In particular, the upfield location and doublet
2
multiplicity of the hydride signal (−23.61 ppm; JHP = 22
Hz; Table 1; cf. the very similar data for chloride analogue 2b)
provide unequivocal evidence for a square-pyramidal complex
in which an apical hydride ligand lies cis to a single basal
phosphine. The latter gives rise to a 31P{1H} singlet at 50.2
ppm. The hydride signal correlates (HMBC) with the IMes
carbene carbon (δC 193.9 ppm, d, 2JCP = 103.2 Hz) and a single
carbonyl ligand (δC 204.7 ppm, d, 2JCP = 8.7 Hz), although not
with the broad OCH3 signal (63.5 ppm, ω0.5 25 Hz).
Assignment of the latter was confirmed by a DEPT-135
experiment (−30 °C, C7D8) and HMQC correlation with the
Finally, it may be noted that the inhibited methanolysis of
3b, relative to 3a, contrasts with behavior earlier established for
the corresponding dihydrogen complexes RuHCl(H2)(L)-
(PCy3), in which the IMes derivative reacted only marginally
slower with methanol and NEt3 than did its PCy3 analogue.5
The difference may be due to the lability of the H2 ligand
2351
dx.doi.org/10.1021/om201288p | Organometallics 2012, 31, 2349−2356

Organometallics
Article
a
Table 1. Key NMR and IR Data for Hydride Complexes Discussed
δH (Ru−X)
IR (ν)
complex
NMR solvent
δP
H (2JHP
)
OCH3
δCO (2JCP
)
CO
Ru−H
RuHCl(CO)(L)(PCy3)
L = PCy3, 2a33
L = IMes, 2b15,33
C6D6
C6D6
46.9
47.8
−24.21 (t, 18 Hz)
−24.82 (d, 21 Hz)
202.1 (t, 14)
202.3 (d, 14)
1905
1894
1862
1881
RuH(OMe)(CO)(L)(PCy3)
bc
c
,
L = IMes, 3b
C6D6
C6D6
C6D6
50.2
53.8
54.0
−23.61 (d, 22 Hz)
−4.25 (t, 20 Hz)
−4.18 (d, 25 Hz)
4.22
4.10
3.96
204.7 (d, 9 Hz)
1875
1890
1939
1948
RuH(OMe)(CO)2(L)(PCy3)
b
L = PCy3, 4a
203.7 (t, 7 Hz)
2006
1891
2013
1896
201.4 (t, 12 Hz)
202.9 (d, 12 Hz)
199.5 (d, 7 Hz)
b
L = IMes, 4b
RuH(OMe)(CO)(H2)(L)2
b
d
d
d
d
L = PCy3, 5a
CH2Cl2
MeOH
72.0
−7.45 (br s)
n.d.
n.d.
n.d.
n.d.
Ru(H)2(CO)2(L)(PCy3)
L = PCy3, 6a8
C6D6
C6D6
68.3
68.9
−7.9 (t, 23 Hz)
206.2
2004
1994
1995
1951
n.d.
b
L = IMes, 6b
−7.37 (d, 25 Hz)
204.7 (d, 8 Hz)
1898
RuHCl(CO)2(L)(PCy3)
b
b
b
L = PCy3, 7a8
C6D6
C6D6
49.8
48.2
−4.9 (t, 20 Hz)
201.6 (t, 7 Hz)
200.7 (t, 12 Hz)
202.3 (d, 13 Hz)
197.2 (d, 7 Hz)
2016
1942
2035
1913
1869
b
L = IMes, 7b
−4.77 (d, 23 Hz)
1954
RuH(OTf)(CO)2(L)(PCy3)
b
L = PCy3, 8a
C6D6
C6D6
53.7
51.3
−4.02 (t, 19 Hz)
−4.07 (d, 23 Hz)
202.7 (t, 14 Hz)
201.7 (t, 7 Hz)
204.4 (d, 14 Hz)
196.7 (d, 6 Hz)
2046
1966
2049
1935
1917
1972
b
L = IMes, 8b
a
NMR chemical shifts in ppm; coupling constants in Hz; IR bands in cm−1. Values at 23 °C unless otherwise noted. NMR samples in 10:1 CH2Cl2−
b
c
MeOH were spiked with C6D6 as a deuterium lock. References are given to literature values in the solvents indicated. This work. Cf. values for
d
transient species 3a at −30 °C in C7D8: δP 47.8 ppm (s); δH −22.94 (d, 19 Hz). At 23 °C in C6D6, δH −22.81 (br t, 2JHP = 17.4 Hz). Measurement
of IR and 13C NMR data was hampered by the low proportion of these species.
methoxy proton singlet (4.22 ppm). 1H NOESY-NMR analysis
reveals a through-space interaction between the methoxy
protons and hydride. Rapid rotation about the Ru−CIMes
bond in 3b is indicated by the equivalence of the IMes
“backbone” protons, as well as the mesityl p-Me nuclei, despite
the difference in environment above and below the basal plane
of the square pyramid.30
Synthesis of Six-Coordinate RuH(OMe)(CO)2(L)(PCy3)
(4a/b) and RuH2(CO)2(IMes)(PCy3) (6b) from RuH(OTf)-
(CO)2(L)(PCy3) (8a/b). High-yield routes to 4a/b and 6b were
devised to support characterization of these previously
unreported complexes (Scheme 5). We initially envisaged
synthesis of 4 from the known8 bis(carbonyl) complex
RuHCl(CO)2(PCy3)2 (7a) and its IMes analogue 7b.
Complexes 7 were conveniently prepared in ca. 85% isolated
yield via reaction of 2a/b with CO. As reactions of 7a with
methanolic NaOMe in THF proved slow (<20% in 4 h), we
converted 7a/b into their more reactive triflates RuH(OTf)-
(CO)2(L)(PCy3) (8a/b) by reaction with AgOTf in THF (8a:
70%; 8b: 83%). Treatment of 8a/b with NaOMe−MeOH in
THF effected complete conversion to yellow RuH(OMe)-
(CO)2(L)(PCy3) (4a/b) within 30 min.31 Sodium triflate was
removed by stripping the reaction mixture to dryness,
extracting with CH2Cl2, and filtering through Celite.
Reprecipitation with hexanes afforded 4a/b as light yellow
powders in excellent yields (ca. 85% each).
A convenient route to RuH2(CO)2(IMes)(PCy3) (6b) from
8b was also developed, via reaction with excess NaH at 50 °C in
THF. Reaction was complete within 45 min. While high
solubility in hexanes and diethyl ether frustrated reprecipitation,
crude 6b exhibits spectroscopic features closely similar to those
reported for 6a (originally prepared by the Chaudret group by
treating RuH2(H2)2(PCy3)2 with CO; data are provided in
Table 1).8
Scheme 5. Synthetic Routes to 4a/b and 6b from 2a/b
The structures of the new complexes 4a/b, 6b, 7b, and 8a/b
were established by one- and two-dimensional NMR experi-
ments, supported by IR spectroscopy32 and by elemental
analysis for all but 6b, which proved highly sensitive toward
decomposition even in the solid state. Coordinative saturation
2352
dx.doi.org/10.1021/om201288p | Organometallics 2012, 31, 2349−2356

Organometallics
Article
in each is indicated by the downfield location of the hydride
signal (between −4 and −8 ppm; Table 1), the triplet or
doublet multiplicities of which indicate retention of the PCy3
ligand(s) originally present. Trans-disposition of the two PCy3
groups, or of the PCy3 and IMes groups, is confirmed from the
singlet multiplicity of the 31P NMR signal (for 4a, 8a) or from
EXPERIMENTAL SECTION
■
General Procedures. Reactions were carried out at room
temperature (23 °C) under argon using standard Schlenk or glovebox
techniques, unless otherwise stated. Dry, oxygen-free solvents were
obtained using a Glass Contour solvent purification system and stored
over Linde 4 Å molecular sieves. C6D6 was degassed by consecutive
freeze/pump/thaw cycles and dried over molecular sieves (Linde 4 Å).
Methanol was distilled from Mg(OMe)2 under Ar and stored over
molecular sieves (Linde 3 Å). Sodium methoxide solutions were
prepared by digesting Na metal in methanol immediately before use.
RuCl2(PCy3)2(CHPh) (1a),35 RuCl2(IMes)(PCy3)(CHPh)
(1b),36,37 and RuHCl(CO)(L)(PCy3) (2a: L = PCy3, 2b: L =
IMes)33 were prepared according to literature procedures. NMR
spectra were recorded on a Bruker Avance 300 or Avance 500
spectrometer, at 298 K unless otherwise specified. Chemical shifts are
2
the magnitude of JCP coupling to the carbene carbon (4b: 96
Hz; 6b: 70 Hz; 7b: 88 Hz; 8b: 84 Hz). In contrast to 6b, the
symmetry of which results in equivalent carbonyl carbon (and
hydride) signals, complexes 4a/b, 7a/b, and 8a/b contain
inequivalent carbonyl groups. Each appears as a 13C{1H} NMR
doublet or triplet, the 2JCP values for which (6−15 Hz) confirm
cis-disposition relative to the PCy3 ligand(s). The expected
HMBC correlations are seen between the hydride and the
carbonyl and carbene ligands. For methoxides 4a/b, assignment
of the methoxy carbons (ca. 67 ppm) is confirmed by DEPT-
135 analysis and HMQC correlation with the methoxy methyl
singlet at ca. 4.0 ppm. A NOESY correlation between the latter
and the hydride signal confirms the cis-disposition of these
ligands. Finally, the triflate CF3 groups in 8a/b exhibit
essentially identical NMR values (a 13C{1H} quartet at ca.
120 ppm (1JCF = 319 Hz) and a 19F{1H} singlet at −77 ppm).
1
reported relative to TMS (13C, H), 85% external H3PO4 (31P), or
1
CFCl3 (19F) at 0 ppm. H and 13C NMR spectra were referenced to
the carbon or residual proton signal of the deuterated solvent; 19F
spectra, to CF3CO2H at −76.55 ppm. In the NMR assignments of
IMes derivatives given below, a/b labels indicate corresponding,
inequivalent nuclei on the same mesityl ring, as indicated by HMQC
or HMBC correlations. IR spectra were measured as Nujol mulls
between NaCl plates using a Bomem MB100 spectrometer or as
powders using a Varian 640-IR reflectance IR spectrometer.
Microanalyses were carried out by Guelph Chemical Laboratories
Ltd., Guelph, Ontario.
CONCLUSIONS
■
NMR-Scale Reactions of Grubbs Catalysts with Methoxide.
In a representative reaction, a solution of 1a (10 mg, 0.012 mmol) was
dissolved in CH2Cl2 (0.7 mL, with a 50 μL spike of C6D6) in a J.
Young NMR tube. To this was added NaOMe as a solution in MeOH
(9.7 μL of a 3.75 M solution, 3 equiv), and the reactions were
monitored by NMR analysis (in some cases with Ph3PO added as an
internal standard for integration). No solvent suppression was used for
the H NMR spectra, as the alkylidene and hydride regions (10 to 30
ppm and 0 to −30 ppm, respectively) could be viewed without
interference.
RuCl2(PCy3)2(CHPh) (1a) + Methoxide: Observation of 4a
and 5a. The solution changed color from deep purple to pale yellow
over 5 min. 31P{1H} NMR (CH2Cl2−MeOH−C6D6): δ 52.1 (s, 4a,
48%), 72.0 (s, 5a, 19%); 11.1 (s, PCy3, 13%); the deficit is attributed
to paramagnetic material (see text). 1H NMR (CH2Cl2−MeOH−
C6D6; hydride region): δ −4.59 (4a), −7.45 (5a). For the synthesis
and full characterization of 4a, see below.
The foregoing describes the rapid reaction of the first-
generation Grubbs catalyst 1a with excess methoxide and
methanol at room temperature. Major products are the
coordinatively saturated methoxyhydride complexes RuH-
(OMe)(CO)2(PCy3)2 (4a) and RuH(OMe)(CO)(H2)(PCy3)2
(5a). Formation of such species may account for the poor
isomerization activity found when an excess of primary
alkoxides is used to trigger CC isomerization following 1a-
mediated metathesis. Reaction of the second-generation
catalyst 1b with methanolic methoxide is much slower and
terminates at the five-coordinate methoxyhydride complex
RuH(OMe)(CO)(IMes)(PCy3) (3b). The identities of the
new complexes 3b and 4a/b were confirmed by independent
synthesis from RuHCl(CO)(IMes)(PCy3) (2b) or hydridotri-
flates RuH(OTf)(CO)2(L)(PCy3) (8a/b), respectively; iso-
lation of RuH(OMe)(CO)(PCy3)2 (3a) is hampered by its
much higher reactivity. The slower rate of formation of 3b at
room temperature, and its stability once formed, reflect the low
lability characteristic of phosphine ligands trans to an N-
heterocyclic carbene. This constrains salt metathesis to a non-
dissociative pathway for 1b, rather than (as with 1a) proceeding
via a sterically accessible four-coordinate species formed by
equilibrium loss of PCy3. The stability of 3b toward reaction
with methanol indicates that for this subsequent reaction the
associative pathway is either inaccessible or prohibitively slow.
Precipitation of NaCl from solution (in which the proportion of
CH2Cl2 dominates by 10:1 over MeOH) contributes to the
greater driving force of the initial salt metathesis reaction.
Notable in this chemistry is the facility with which these
“robust” benzylidene complexes decompose into hydride
species under exceptionally mild conditions. Consumption of
1a by alkoxide occurs within minutes at room temperature; loss
of 1b, while slower, is complete within a few hours. Reaction
conditions that can give rise to adventitious alkoxides,
particularly in the presence of methanol cosolvent (a favored
reaction medium for olefin metathesis reactions of biologically
relevant substrates),34 should thus be recognized as profoundly
detrimental to catalytic performance.
1
RuCl2(IMes)(PCy3)(CHPh) (1b) + Methoxide: Observation
of 3b. A color change from red to pale orange-yellow occurred over 4
h. 31P{1H} NMR (CH2Cl2−MeOH−C6D6): δ 49.6 (3b, 73%), 11.1 (s,
PCy3, 10%); the deficit is attributed to paramagnetic material (see
1
text). H NMR (CH2Cl2−MeOH−C6D6; hydride region): δ −25.02
ppm (3b). For the synthesis of 3b using 20 equiv of NaOMe, with full
details of NMR characterization, see below.
In Situ Observation of 5a. Complex 5a was generated in situ (as
a 2:1 mixture of 4a and 5a) as described above. Sweeping the
atmosphere with Ar resulted in partial transformation of 5a into 3a
(ca. 15% vs the original 1a). This was readily reversed by freeze−
pump−thaw degassing (3×) and backfilling with H2. NMR data for 5a:
31P{1H} NMR (121.5 MHz, 10:1 CH2Cl2−MeOH): δ 72.0 (s). H
1
NMR (300.1 MHz, 10:1 CH2Cl2−MeOH): δ −7.45 (br s). Hydride
T1(min) (C7D8, H2, 263 K, 500.1 MHz): 37.5 ms (dH−H = 0.95 Å for
fast-spinning H2).26,38 Decoalescence was not observed down to −90
°C in C7D8.
Synthesis of RuH(OMe)(CO)(IMes)(PCy3), 3b. Adding a solution
of NaOMe in methanol (300 μL, 2.75 M, 0.83 mmol, 20 equiv) to a
vigorously stirred solution of RuHCl(CO)(IMes)(PCy3) (2b) (31 mg,
0.041 mmol) in 3 mL of CH2Cl2 caused a color change from orange-
yellow to red-orange over 15 min. In situ 31P{1H} NMR analysis
indicated solely 3b. The solvent was stripped off, and the residue was
redissolved in benzene, filtered through Celite, and stripped to
dryness. Yield: 24 mg (78%). 31P{1H} NMR analysis of the residue
revealed, in addition to 3b, small amounts of Ru(H)2(CO)2(IMes)-
2353
dx.doi.org/10.1021/om201288p | Organometallics 2012, 31, 2349−2356

Organometallics
Article
(PCy3) (6b) (5%), RuH(OMe)(CO)2(IMes)(PCy3) (4b) (2%), and
free PCy3 (5%), which impede microanalysis. Data for 3b: 31P{1H}
NMR (121.5 MHz, C7D8): δ 50.2 (s). H NMR (300.1 MHz, C6D6):
129.2 (s, Mes m-CH), 121.5 (s, CHN), 37.9 (d, 1JCP = 20.9 Hz, C1
of Cy), 30.1 (s, Cy), 28.1 (d, JCP = 10.1 Hz, Cy), 27.0 (s, C4 of Cy),
21.2 (s, Mes p-CH3), 18.6 (s, Mes o-CH3). IR (Nujol, cm−1): ν(CO)
1995 (s), 1951 (s); ν(Ru−H) 1898 (w).
1
δ 6.85 (s, 2H, Mes m-CHb) 6.83 (s, 2H, Mes m-CHa), 6.26 (s, 2H, 
CHN), 4.22 (s, 3H, OCH3), 2.42−2.38 (two overlapping s, 12H, Mes
o-CH3), 2.14 (s, 6H, Mes p-CH3), 2.2−1.1 (m, Cy; accurate
integration impeded by overlap with Cy signals for 6b, 4b), −23.61
(d, 2JHP = 22.2 Hz, 1H, RuH). 13C{1H} NMR (125.8 MHz, C7D8, 243
K): δ 204.7 (d, 2JCP = 8.7 Hz, CO), 193.9 (d, 2JCP = 103.2 Hz, NCN),
138.3 (s, Mes p-C), 138.0 (br, Mes o-C), 137.3 (br, Mes o-C), 136.4 (s,
Mes i-C), 129.0 (s, Mes m-CH), 122.1 (s, CHN), 63.5 (br s, ω0.5 25
Hz, OCH3), 34.2 (d, 1JCP = 16.6 Hz, C1 of Cy), 31.1 (s, Cy), 30.2 (s,
Cy) 28.6 (m, Cy), 27.4 (s, Cy), 21.5 (s, Mes p-CH3), 19.3 (s, Mes o-
CH3). IR (Nujol, cm−1): ν(CO) 1875 (s); ν(Ru−H) 1890 (w).
Preparation of RuH(OMe)(CO)2(L)(PCy3), 4a/b. L = PCy3, 4a.
Addition of NaOMe as a solution in methanol (49 μL, 3.58 M,
0.18 mmol) to RuH(OSO2CF3)(CO)2(PCy3)2 (8a) (150 mg,
0.173 mmol) in THF (5 mL), with stirring, caused a color
change from pale brown to yellow over 30 min and deposition
of a white precipitate. The reaction mixture was stripped to
dryness, and the residue was taken up in CH2Cl2 (15 mL). The
mixture was filtered through Celite, concentrated to ca. 0.5 mL,
treated with hexanes (5 mL), and chilled to −35 °C. A light
yellow powder deposited, which was filtered off, washed with
cold hexanes (3 × 2 mL), and dried under vacuum. Yield: 110
Preparation of RuHCl(CO)2(IMes)(PCy3), 7b. (Known8 7a was
prepared similarly, in 84% yield.) An orange-yellow solution of
RuHCl(CO)(IMes)(PCy3) (2b) (320 mg, 0.546 mmol) in benzene (5
mL) was stirred under 1 atm of CO for 1 h, after which the colorless
solution was concentrated (ca. 0.5 mL) and hexanes were added to
precipitate the white product. This was reprecipitated from benzene−
hexanes, filtered off, washed with cold hexanes (3 × 2 mL), and dried
under vacuum. Yield: 270 mg (81%). 31P{1H} NMR (121.5 MHz,
1
C6D6): δ 48.2 (s). H NMR (300.1 MHz, C6D6): δ 6.87 (s, 2H, Mes
m-CH), 6.84 (s, 2H, Mes m-CH), 6.28 (s, 2H, CHN), 2.36 (s, 6H,
Mes o-CH3), 2.34 (s, 6H, Mes o-CH3), 2.21 (s, 6H, Mes p-CH3), 2.3−
1.1 (m, 33H, Cy), −4.77 (d, 2JHP = 22.8 Hz, 1H, RuH). 13C{1H} NMR
(125.8 MHz, C6D6): δ 202.3 (d, 2JCP = 12.6 Hz, CO), 197.2 (d, 2JCP
=
2
6.6 Hz, CO), 184.6 (d, JCP = 88.4 Hz, NCN), 139.4 (m, Mes i-C),
138.5 (s, Mes p-C), 136.9 (m, Mes o-C), 136.6 (m, Mes o-C), 129.4
1
(m, Mes m-CH), 122.7 (overlapping s, CHN), 34.5 (d, JCP = 20.0
Hz, C1 of Cy), 29.3 (d, J = 12.4 Hz (or overlapping s), Cy), 28.0 (d,
JCP = 9.0 Hz, Cy), 27.9 (d, JCP = 9.0 Hz, Cy), 26.8 (s, C4 of Cy), 21.2
(s, Mes p-CH3), 18.8 (s, Mes o-CH3), 18.7 (s, Mes o-CH3). IR
(powder, cm−1): ν(CO) 2035 (s), 1913 (s); ν(Ru−H) 1954 (w).
Anal. Calcd for C41H59ClN2O2PRu: C, 63.18; H, 7.63; N, 3.59. Found:
C, 63.09; H, 7.53; N, 3.63.
1
mg (85%). 31P{1H} NMR (121.5 MHz, C6D6): δ 53.8 (s). H
Preparation of RuH(OSO2CF3)(CO)2(L)(PCy3), 8a/b. L = PCy3,
8a. Solid AgOSO2CF3 (90 mg, 0.35 mmol) was added to
RuHCl(CO)2(PCy3)2 (7a) (250 mg, 0.35 mmol) in THF (15
mL) in a foil-wrapped vessel and stirred for 1 h. The solvent
was stripped off under vacuum, and the residue extracted with
CH2Cl2 (3 × 5 mL). The combined extracts were filtered
through Celite, concentrated (ca. 0.5 mL), and treated with
hexanes to precipitate the pale beige powder. This was chilled
(−35 °C), filtered off, washed with cold hexanes (3 × 2 mL),
and dried under vacuum. Yield: 200 mg (70%). 31P{1H} NMR
(121.5 MHz, C6D6): δ 53.7 (s). 1H NMR (300.1 MHz, C6D6):
NMR (300.1 MHz, C6D6): δ 4.10 (s, 3H, OCH3), 2.3−1.2 (m,
2
13
66H, Cy), −4.25 (t, JHP = 20.2 Hz, 1H, RuH). C{1H} NMR
2
(125.8 MHz, C6D6): δ 203.7 (t, JCP = 6.6 Hz, CO), 201.4 (t,
2JCP = 11.6 Hz, CO), 66.4 (s, OCH3), 34.4 (vt, 1JCP = 10 Hz, C1
of Cy), 29.8 (d, JCP = 2.6 Hz (or two overlapping s), Cy), 28.3
(overlapping m, Cy), 27.1 (s, C4 of Cy). IR (Nujol, cm−1):
ν(CO) 2006 (s), 1891 (s); ν(Ru−H) 1939 (w). Anal. Calcd for
C39H70O3P2Ru: C, 62.46; H, 9.41. Found: C, 62.07; H, 9.77.
L = IMes, 4b. The light yellow powder was prepared as for 4a, from
RuH(OSO2CF3)(CO)2(IMes)(PCy3) (8b) (154 mg, 0.173 mmol).
Yield: 115 mg (86%). 31P{1H} NMR (121.5 MHz, C6D6): δ 54.0 (s).
1H NMR (300.1 MHz, C6D6): δ 6.91 (s, 2H, Mes m-CHa), 6.89 (s,
2H, Mes m-CHb), 6.32 (s, 2H, CHN), 3.96 (s, 3H, OCH3), 2.37 (s,
2
δ 2.4−1.0 (m, 66H, Cy), −4.02 (t, JHP = 18.8 Hz, 1H, RuH).
2
13C{1H} NMR (125.8 MHz, C6D6): δ 202.7 (t, JCP = 13.5 Hz,
2
1
CO), 201.7 (t, JCP = 7.0 Hz, CO), 119.8 (q, JCF = 319.2 Hz,
OSO2CF3), 34.7 (br s, C1 of Cy), 29.7 (d, JCP = 12.2 Hz (or
b
a
6H, Mes o-CH3 ), 2.30 (s, 6H, Mes o-CH3 ), 2.20 (s, 6H, Mes p-CH3),
2.2−1.0 (m, 33H, Cy), −4.18 (d, 2JHP = 25.3 Hz, 1H, RuH). 13C{1H}
NMR (125.8 MHz, C6D6): δ 202.9 (d, 2JCP = 12.4 Hz, CO), 199.5 (d,
2JCP = 6.9 Hz, CO), 187.5 (d, 2JCP = 95.6 Hz, NCN), 139.4 (s, Mes i-
C), 138.2 (s, Mes p-C), 137.4 (s, Mes o-Ca), 136.2 (s, Mes o-Cb), 129.2
(s, Mes m-CHb), 128.9 (s, Mes m-CHa), 122.3 (m, CHN), 67.4 (s,
OCH3), 34.1 (d, 1JCP = 18.8 Hz, C1 of Cy), 29.4 (d, JCP = 6.9 Hz (or
overlapping s), Cy), 27.7 (m, Cy), 26.8 (s, C4 of Cy). F{1H}
19
NMR (282.4 MHz, C6D6): δ −77.2 (s, CF3). IR (powder,
cm−1): ν(CO) 2046 (s), 1966 (s); ν(Ru−H) 1917 (w). Anal.
Calcd for C39H67F3O5P2RuS: C, 53.96; H, 7.78. Found: C,
54.22; H, 8.20.
L = IMes, 8b. Reaction was as for 8a, using RuHCl(CO)2(IMes)-
(PCy3) (7b) (250 mg, 0.321 mmol) as precursor. Yield of the pale
beige powder: 190 mg (83%). In the NMR assignments, a/b labels
indicate corresponding, inequivalent nuclei on the same mesityl ring,
as indicated by HMQC or HMBC correlations; a prime label is used to
differentiate the two Mes rings. 31P{1H} NMR (121.5 MHz, C6D6): δ
3
3
two overlapping s), Cy), 28.3 (d, JCP = 9.9 Hz, Cy), 28.3 (d, JCP
=
10.1 Hz, Cy), 27.0 (s, C4 of Cy), 21.2 (s, Mes p-CH3), 18.4
(overlapping s, Mes o-CH3). IR (powder, cm−1): ν(CO) 2013 (s),
1896 (s); ν(Ru−H) 1948 (w). Anal. Calcd for C42H62N2O3PRu: C,
65.09; H, 8.06; N, 3.61. Found: C, 64.76; H, 7.96; N, 3.67.
1
51.3 (s). H NMR (300.1 MHz, C6D6): δ 6.94 (s, 1H, Mes m-CHa′),
Preparation of Ru(H)2(CO)2(IMes)(PCy3), 6b. Solid NaH (30
mg, 1.3 mmol) was added to a solution of RuH(OSO2CF3)-
(CO)2(IMes)(PCy3) (8b) (110 mg, 0.123 mmol) in THF (1.0
mL), and the reaction mixture was heated to 50 °C. A color change
from pale brown to light yellow occurred over 45 min, accompanied by
complete transformation to 6b. The solvent was stripped off, and the
residue taken up in benzene (5 mL) and filtered through Celite.
Additional, unassigned NMR signals (<5% total integration) were
observed when the filtrate was stripped to dryness and redissolved in
C6D6. Attempts to obtain pure 6b by reprecipitation from benzene−
hexanes, benzene−diethyl ether, or neat hexanes were frustrated by
high solubility, and satisfactory microanalysis could not be obtained.
31P{1H} NMR (121.5 MHz, C6D6): δ 68.9 (s). 1H NMR (300.1 MHz,
C6D6): δ 6.88 (s, 4H, Mes m-CH), 6.28 (s, 2H, CHN), 2.24 (s,
12H, Mes o-CH3), 2.19 (s, 6H, Mes p-CH3), 2.0−1.1 (m, 33H, Cy),
6.89 (s, 1H, Mes m-CHa), 6.84 (s, 1H, Mes m-CHb), 6.73 (s, 1H, Mes
m-CHb′), 6.32 (s, 1H, CHN), 6.22 (s, 1H, CHN), 2.57 (s, 3H,
a
a
Mes o-CH3 ′), 2.34 (s, 3H, Mes o-CH3 ), 2.19 (s, 3H, Mes p-CH3),
b
2.12 (s, 3H, Mes p-CH3′), 2.02 (overlapping s; 6H, Mes o-CH3 , o-
b
2
CH3 ′), 2.2−1.0 (m, 33H, Cy), −4.07 (d, JHP = 23.1 Hz, 1H, RuH).
13C{1H} NMR (125.8 MHz, C6D6): δ 204.4 (d, 2JCP = 13.9 Hz, CO),
2
2
196.7 (d, JCP = 5.5 Hz, CO), 182.3 (d, JCP = 83.8 Hz, NCN), 139.9
(s, Mes i-C), 139.7 (s, Mes p-C′), 138.3 (s, Mes p-C), 138.1 (s, Mes o-
Ca′), 137.5 (s, Mes o-Ca), 137.3 (s, Mes i-C′), 135.3−135.2
(overlapping s, Mes o-Cb, o-Cb′), 130.4 (s, Mes m-CHa′), 129.4 (s,
Mes m-CHa), 129.1 (s, Mes m-CHb′), 128.5 (s, Mes m-CHb), 123.5 (s,
CHN), 119.9 (q, JCF = 320.4 Hz, OSO2CF3), 34.2 (d, JCP = 15.7
Hz, C1 of Cy), 29.3 (overlapping s, Cy), 27.8 (d, JCP = 10.2 Hz, Cy),
27.7 (d, JCP = 10.4 Hz, Cy), 26.7 (s, C4 of Cy), 21.0 (coincident s, Mes
p-CH3, p-CH3′), 19.1 (s, Mes o-CH3), 18.4−18.3 (overlapping s, Mes
o-CH3). 19F{1H} NMR (282.4 MHz, C6D6): δ −77.1 (s, CF3). IR
(powder, cm−1): ν(CO) 2049 (s), 1935 (s); ν(Ru−H) 1972 (w).
1
1
2
−7.37 (d, JHP = 24.9 Hz, 2H, RuH). 13C{1H} NMR (125.8 MHz,
2
2
C6D6): δ 204.7 (d, JCP = 7.7 Hz, CO), 190.1 (d, JCP = 69.8 Hz,
NCN), 139.6 (s, Mes i-C), 138.0 (s, Mes p-C), 136.1 (s, Mes o-C),
2354
dx.doi.org/10.1021/om201288p | Organometallics 2012, 31, 2349−2356

Organometallics
Article
Anal. Calcd for C42H59F3N2O5PRuS: C, 56.49; H, 6.66; N, 3.14.
Found: C, 56.65; H, 6.68; N, 3.10.
(12) Caskey, S. R.; Stewart, M. H.; Ahn, Y. J.; Johnson, M. J. A.;
Kampf, J. W. Organometallics 2005, 24, 6074−6076.
(13) A prior example of a five-coordinate Ru-alkoxyhydride complex,
RuH(OEt)(CO)(PtBu2Me)2, was found to decompose rapidly in
solution. See: (a) Poulton, J. T.; Sigalas, M. P.; Felting, K.; Streib, W.
E.; Eisenstein, O.; Caulton, K. G. Inorg. Chem. 1994, 33, 1476−1485.
Several coordinatively saturated methoxyhydrides have been isolated:
(b) FeH(OMe)[K4-P(CH2CH2PMe2)3]: Field, L. D.; Messerle, B. A.;
Smernik, R. J. Inorg. Chem. 1997, 36, 5984−5990. (c) mer-cis-
IrHCl(OMe)(PEt3)3: Blum, O.; Milstein, D. Angew. Chem., Int. Ed.
Engl. 1995, 34, 229−231. (d) OsHCl(NO)(OMe)(PiPr3)2: Werner,
H.; Michenfelder, A.; Schulz, M. Angew. Chem., Int. Ed. Engl. 1991, 30,
596−598. (e) cis-[IrH(OMe)(PMe3)4]PF6: Milstein, D.; Calabrese, J.
C.; Williams, I. D. J. Am. Chem. Soc. 1986, 108, 6387−6389. (f)
Cp*2ZrH(OMe): Manriquez, J. M.; McAlister, D. R.; Sanner, R. D.;
Bercaw, J. E. J. Am. Chem. Soc. 1976, 98, 6733−6735. For polynuclear
examples, see: (g) Buil, M. L.; Esteruelas, M. A.; Modrego, J.; Onate,
E. New J. Chem. 1999, 23, 403−406. (h) Lee, K. K. H.; Wong, W. T. J.
Chem. Soc., Dalton Trans. 1996, 1707−1720.
ASSOCIATED CONTENT
■
S
* Supporting Information
Kinetics plots showing phosphine-dependence studies, and
NMR spectra for the new complexes 3b, 4a/b, 6b, 7b, and 8a/
b. This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: dfogg@uottawa.ca.
Notes
The authors declare no competing financial interest.
(14) Dharmasena, U. L.; Foucault, H. M.; dos Santos, E. N.; Fogg, D.
E.; Nolan, S. P. Organometallics 2005, 24, 1056−1058.
ACKNOWLEDGMENTS
■
This work was supported by NSERC of Canada and the
Canada Foundation for Innovation (CFI). N.J.B. thanks
NSERC for a postgraduate scholarship.
(15) Lee, H. M.; Smith, D. C. Jr.; He, Z.; Stevens, E. D.; Yi, C. S.;
Nolan, S. P. Organometallics 2001, 20, 794−797.
(16) Sanford, M. S.; Love, J. A.; Grubbs, R. H. J. Am. Chem. Soc.
2001, 123, 6543−6554.
(17) Fogg, D. E.; Foucault, H. M. Ring-Opening Metathesis
Polymerization. In Comprehensive Organometallic Chemistry III;
Crabtree, R. H.; Mingos, D. M. P., Eds.; Elsevier: Oxford, 2007;
Vol. 11, pp 623−652.
REFERENCES
■
(1) Alcaide, B.; Almendros, P.; Luna, A. Chem. Rev. 2009, 109, 3817−
3858.
(2) Alcaide, B.; Almendros, P. Chem.Eur. J. 2003, 9, 1259−1262.
(3) Tandem catalysis originating in the Grubbs complexes has been
reviewed: (a) Shindoh, N.; Takemoto, Y.; Takasu, K. Chem.Eur. J.
2009, 15, 12168−12179. (b) Wasilke, J.-C.; Obrey, S. J.; Baker, R. T.;
Bazan, G. C. Chem. Rev. 2005, 105, 1001−1020. (c) Fogg, D. E.; dos
Santos, E. N. Coord. Chem. Rev. 2004, 248, 2365−2379.
(4) Heating the Grubbs catalysts with primary alkoxides, or alcohols
and base, generated RuHCl(CO)(L)(PCy3) (2a: L = PCy3; 2c: L =
H2IMes) in 30−50% yield. See: (a) Dinger, M. B.; Mol, J. C.
Organometallics 2003, 22, 1089−1095. (b) Dinger, M. B.; Mol, J. C.
Eur. J. Inorg. Chem. 2003, 2827−2833. Trace 2c also forms on
washing 1a with methanol; see: (c) Trnka, T. M.; Morgan, J. P.;
Sanford, M. S.; Wilhelm, T. E.; Scholl, M.; Choi, T.-L.; Ding, S.; Day,
M. W.; Grubbs, R. H. J. Am. Chem. Soc. 2003, 125, 2546−2558.
Thermolysis of Fischer carbenes permits isolation of 2a in 69% yield:
(d) Louie, J.; Grubbs, R. H. Organometallics 2002, 21, 2153−2164 or
2c in quantitative yield. (e) Arisawa, M.; Terada, Y.; Takahashi, K.;
Nakagawa, M.; Nishida, A. J. Org. Chem. 2006, 71, 4255−4261. In
tandem ROMP−hydrogenation by various Grubbs catalysts, hydro-
genation is catalyzed by 2a/b generated in the presence of H2, amine
base, and methanol. See: (f) Camm, K. D.; Castro, N. M.; Liu, Y.;
Czechura, P.; Snelgrove, J. L.; Fogg, D. E. J. Am. Chem. Soc. 2007, 129,
4168−4169. (g) Drouin, S. D.; Zamanian, F.; Fogg, D. E.
Organometallics 2001, 20, 5495−5497 Catalyst-specific protocols
give improved yields of 2: see ref 5. Thus, sequential hydrogenolysis
and methanolysis of 1a maximizes formation of 2a (72% net yield); in
contrast, reaction with methanol and NEt3 maximizes conversion of 1b
into 2b (83%).
(18) Cheung, W.-M.; Chiu, W.-H.; Yi, X.-Y.; Zhang, Q.-F.; Williams,
I. D.; Leung, W.-H. Organometallics 2010, 29, 1981−1984.
(19) In striking contrast, attack of bound phenoxide on the alkylidene
proton results in α-elimination of phenol and formation of a
ruthenium carbyne (see above and refs 10−12).
(20) Occhipinti, G.; Jensen, V. R. Organometallics 2011, 30, 3522−
3529.
(21) Rudolf, G. C.; Hamilton, A.; Orpen, A. G.; Owen, G. R. Chem.
Commun. 2009, 553−555.
(22) Wilson, G. O.; Porter, K. A.; Weissman, H.; White, S. R.; Sottos,
N. R.; Moore, J. S. Adv. Synth. Catal. 2009, 351, 1817−1825.
(23) Jong, H.; Patrick, B. O.; Fryzuk, M. D. Organometallics 2011, 30,
2333−2341.
(24) Occhipinti, G.; Bjorsvik, H.-R.; Toernroos, K. W.; Jensen, V. R.
Organometallics 2007, 26, 5803−5814.
(25) Both σ-bond metathesis and σ-complex-assisted metathesis are
plausible: we favor the latter, given the evidence for a dissociative
pathway. See: Perutz, R. N.; Sabo-Etienne, S. Angew. Chem., Int. Ed.
2007, 46, 2578−2592.
(26) (a) Jessop, P. G.; Morris, R. H. Coord. Chem. Rev. 1992, 121,
155−284. (b) Kubas, G. J. Metal Dihydrogen and σ-Bond Complexes;
Kluwer/Plenum: Dordrecht, 2001.
(27) Clapham, S. E.; Hadzovic, A.; Morris, R. H. Coord. Chem. Rev.
2004, 248, 2201−2237.
(28) For leading early examples of C−H activation of cyclo-
hexylphosphine ligands at Ru centres, see: (a) Christ, M. L.; Sabo-
Etienne, S.; Chaudret, B. Organometallics 1995, 14, 1082−1084.
(b) Borowski, A.; Sabo-Etienne, S.; Christ, M. L.; Donnadieu, B.;
Chaudret, B. Organometallics 1996, 15, 1427−1434. (c) Leitner, W.;
Six, C. Chem. Ber. 1997, 130, 555−558. (d) Six, C.; Gabor, B.; Gorls,
H.; Mynott, R.; Philipps, P. Organometallics 1999, 18, 3316−3326.
Exhaustive dehydrogenation of the cyclohexyl rings affords para-
magnetic products. See: (e) Amoroso, D.; Yap, G. P. A.; Fogg, D. E.
Can. J. Chem. 2001, 79, 958−963.
(29) For selected examples of activation of IMes ligands at electron-
rich Ru centers, see: (a) Abdur-Rashid, K.; Fedorkiw, T.; Lough, A. J.;
Morris, R. H. Organometallics 2004, 23, 86−94. (b) Chilvers, M. J.;
Jazzar, R. F. R.; Mahon, M. F.; Whittlesey, M. K. Adv. Synth. Catal.
2003, 345, 1111−1114. (c) Jazzar, R. F. R.; Macgregor, S. A.; Mahon,
M. F.; Richards, S. P.; Whittlesey, M. K. J. Am. Chem. Soc. 2002, 124,
(5) Beach, N. J.; Camm, K. D.; Fogg, D. E. Organometallics 2010, 29,
5450−5455.
(6) Schmidt, B. J. Org. Chem. 2004, 69, 7672−7687.
(7) Beach, N. J.; Blacquiere, J. M.; Drouin, S. D.; Fogg, D. E.
Organometallics 2009, 28, 441−447.
(8) Christ, M. L.; Sabo-Etienne, S.; Chaudret, B. Organometallics
1994, 13, 3800−3804.
(9) Sanford, M. S.; Henling, L. M.; Day, M. W.; Grubbs, R. H. Angew.
Chem., Int. Ed. 2000, 39, 3451−3453.
(10) Conrad, J. C.; Amoroso, D.; Czechura, P.; Yap, G. P. A.; Fogg,
D. E. Organometallics 2003, 22, 3634−3636.
(11) Coalter, J. N.; Bollinger, J. C.; Eisenstein, O.; Caulton, K. G.
New J. Chem. 2000, 24, 925−927.
2355
dx.doi.org/10.1021/om201288p | Organometallics 2012, 31, 2349−2356

Organometallics
Article
4944−4945. For an early review of bond activation within NHC
ligands, see: Crudden, C. M.; Allen, D. P. Coord. Chem. Rev. 2004, 248,
2247−2273.
(30) Rotation about the metal−CNHC bond at room temperature is
commonly observed for aryl-NHC complexes. See, for example:
(a) Dible, B. R.; Sigman, M. S. Inorg. Chem. 2006, 45, 8430−8441.
(b) Ritleng, V.; Barth, C.; Brenner, E.; Milosevic, S.; Chetcuti, M. J.
Organometallics 2008, 27, 4223−4228. (c) Lee, J. P.; Ke, Z.; Ramirez,
M. A.; Gunnoe, T. B.; Cundari, T. R.; Boyle, P. D.; Petersen, J. L.
Organometallics 2009, 28, 1758−1775. (d) Kotyk, M. W.; Gorelsky, S.
I.; Conrad, J. C.; Carra, C.; Fogg, D. E. Organometallics 2009, 5424−
5431.
(31) Related chemistry was suggested by a referee as an alternative
route to 3a/b, via reaction of 2a/b with AgOTf, then NaOMe. This
was explored for 2b, but is thwarted by the rapid decomposition of 2b
in the presence of AgOTf.
(32) Two strong ν(CO) bands and one weak ν(Ru−H) band were
observed in the IR spectrum for each of 4a/b, 7a/b, and 8a/b.
(33) Beach, N. J.; Dharmasena, U. L.; Drouin, S. D.; Fogg, D. E. Adv.
Synth. Catal. 2008, 350, 773−777.
(34) Camm, K. D.; Fogg, D. E. From Drug Cocktails to Tissue
Engineering: Synthesis of ROMP Polymers for Biological Applications.
In NATO Sci. Ser. II; Imamoglu, Y.; Dragutan, V., Eds.; Springer
Verlag: Berlin, 2007; Vol. 243, pp 285−303.
(35) Schwab, P.; Grubbs, R. H.; Ziller, J. W. J. Am. Chem. Soc. 1996,
118, 100−110.
(36) Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett. 1999,
1, 953−956.
(37) Huang, J.; Stevens, E. D.; Nolan, S. P.; Petersen, J. L. J. Am.
Chem. Soc. 1999, 121, 2674−2678.
(38) See ref 26a and: (a) Bautista, M. T.; Cappellani, E. P.; Drouin,
S. D.; Morris, R. H.; Schweitzer, C. T.; Sella, A.; Zubkowski, J. J. Am.
Chem. Soc. 1991, 113, 4876−4687. (b) Maltby, P. A.; Schlaf, M.;
Steinbeck, M.; Lough, A. J.; Morris, R. H.; Klooster, W. T.; Koetzle, T.
F.; Srivastava, R. C. J. Am. Chem. Soc. 1996, 118, 5396−5407.
2356
dx.doi.org/10.1021/om201288p | Organometallics 2012, 31, 2349−2356