Bulky N-Phosphino-Functionalized N-Heterocyclic Carbene Ligands: Synthesis, Ruthenium Coordination Chemistry, and Ruthenium Alkylidene Complexes for Olefin Metathesis

Article abstract of DOI:10.1021/acs.inorgchem.5b00513

Ruthenium chemistry and applications in catalytic olefin metathesis based on N-phosphino-functionalized N-heterocyclic carbene ligands (NHCPs) are presented. Alkyl NHCP Ru coordination chemistry is described, and access to several potential synthetic prec

Full text of DOI:10.1021/acs.inorgchem.5b00513

Article
pubs.acs.org/IC
Bulky N‑Phosphino-Functionalized N‑Heterocyclic Carbene Ligands:
Synthesis, Ruthenium Coordination Chemistry, and Ruthenium
Alkylidene Complexes for Olefin Metathesis⊥
Christopher C. Brown,*,† Frank Rominger,‡ Michael Limbach,† and Peter Hofmann†,‡
‡
†Catalysis Research Laboratory (CaRLa) and Organisch-Chemisches Institut, University of Heidelberg, Im Neuenheimer Feld 584,
D-69120 Heidelberg, Germany
S
* Supporting Information
ABSTRACT: Ruthenium chemistry and applications in catalytic olefin metathesis based on
N-phosphino-functionalized N-heterocyclic carbene ligands (NHCPs) are presented. Alkyl
NHCP Ru coordination chemistry is described, and access to several potential synthetic
precursors for ruthenium alkylidene complexes is outlined, incorporating both trimethylsilyl
and phenyl alkylidenes. The Ru alkylidene complexes are evaluated as potential olefin
metathesis catalysts and were shown to behave in a latent fashion. They displayed catalytic
activity at elevated temperatures for both ring closing metathesis and ring opening
metathesis polymerization.
Specifically, we have an interest in developing cis-chelating
analogues of the Grubbs second-generation catalyst system.
Our efforts are centered on the use of bidentate bulky ligands
to enforce the cis binding mode of the donor atoms. We have
previously reported on the synthesis, characterization, and
catalytic activity of two classes of systems. Our original report
centered around dicationic, dinuclear bis(di-tert-
butylphosphino)methane (dtbpm) ruthenium olefin metathesis
catalysts (F1). These systems, containing a four-membered
bisphosphine chelate structure, showed remarkably high activity
in ring opening metathesis polymerization (ROMP), leading to
at least an order of magnitude of improvement for ROMP of
substrates such as cis-cyclooctene and cyclopentene. The
mechanistic details of these efficient ROMP-catalysts were
investigated and deciphered both in the solution and in the gas
phase.11 In the course of our bisphosphine studies, the five-
membered bidentate bisphosphine bis(di-tert-butylphosphino)-
ethane (dtbpe) system was also employed, leading to (F2),
which showed a decrease in activity when compared to the
related dtbpm system F1.10 Such bisphosphines may be
considered the cis-phosphine coordination analogues of first
generation Grubbs catalysts (A).
We also investigated a new series of structurally related
potential olefin metathesis catalyst precursors with mixed
bidentate N-heterocyclic carbene−phosphine ligand systems
(E).10 The NHC and phosphine donors in those studies are
joined via a methylene bridge, thus giving stable five-membered
chelate structures. One may consider these NHCP-congeners E
the cis-coordination analogues of a second generation Grubbs
catalyst B. Notable differences between the highly active bis-
phosphine system F1 and G are evident, such as the increasing
of the chelate ring system from four-membered to a less
INTRODUCTION
■
The introduction of the first generation of Grubbs catalysts
such as A,1 beginning in the 1990s, proved to be a significant
discovery which allowed for the simplification of multistep
organic synthesis, for improved selectivities in organic trans-
formations, and the homogeneously catalyzed polymerization
of a wide variety of functionalized cyclic olefinic substrates. A
large manifold of ruthenium alkylidene complexes for olefin
metathesis have been prepared by replacing, for example, one
or both of the phosphines with N-heterocyclic carbene ligands
(NHCs) (B)2 and pyridines,3 of the chlorides with bromides,
iodides, or pseudohalides such as alkoxides (C)4 or by varying
the alkylidene moiety (D).5 The investigation of NHC
chelating ligand systems has, however, received significantly
less attention. Notable examples include the employment of
axially chiral 1,1′-binaphthyl- or 1,1′-biaryl-substituted NHC-
ruthenium catalysts for asymmetric olefin metathesis,6 and the
catalysts introduced for Z-selective olefin metathesis by Grubbs
et al. featuring NHC ligands with C−H-activated N-adamantyl
groups, leading to crowded five-membered chelate complexes
(D).
Although numerous phosphine functionalized NHC ligands
(NHCPs) have been published since the synthesis of the first
representative example by Herrmann et al. in 1996,7 few
systems possess electron-rich, bulky dialkyl-substituted phos-
phino groups.8 By far, most research related to the use of
NHCPs as spectator ligands in transition metal catalysis has
focused on cross-coupling reactions. A small number of papers
on the ruthenium coordination chemistry of NHCP ligands
have been published.8a,9 Closely related to the present work is,
however, a report recently published by our group concerning
bulky N-phosphino-methyl functionalized N-heterocyclic car-
bene chelate ligands.10 Our report is the only example, to the
best of our knowledge, of a study into the catalytic olefin
metathesis viability of a ruthenium NHCP system.
Received: March 6, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.5b00513
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. Selected olefin metathesis catalysts.
strained five-membered chelate ring and reduced steric bulk
proximal to the metal center. Studies are currently underway
into the evaluation of the properties of G, and the cationic
versions of E, the direct NHCP analogues of F2. Consequently,
it was an open question whether the preparation and evaluation
of four-membered ruthenium NHC-phosphine complexes, such
as neutral mononuclear H, may provide improvements in
stability over bis-phosphine F1 or of catalytic activity over
NHC-phosphine complexes such as (G). For complexes of the
type H, their four-membered chelate structuresif these
species are accessiblemay provide the possibility of strained
chelate ring opening (hemilability) and generation of active Ru
14 VE intermediates which are widely believed to initiate the
olefin coordination step in Grubbs second based olefin
metathesis (vide infra).
Scheme 1. Hemilabile Behavior of Ru-NHCP Complexes of
Type H where R = Me and R′ = tBu
The first synthesis of an NHCP ligand system for Ru-
complexes of type H (R = C2H5, R′ = tBu) with a direct
nitrogen−phosphorus bond was achieved in 2007 by our
group.11a Soon after in 2010 an elegant more general synthesis
of such ligands was published by Marchenko et al., followed up
by slight synthetic variations for a broader scope of ligands in
2012.12b,c These syntheses provided the opportunity to
investigate their metal coordination chemistry. Our first
communication into this area for Ru described the unexpected
reaction between a ruthenium alkylidene and an NHCP
ligand,13 following investigations involving group 10 and 11
metals.11,12c,14 While such NHCP ligands have a general
tendency to act as bridging ligands between metal centers, this
has only been observed for group 11 metals, and otherwise, the
desired four-membered chelate structures have been found.
Our previous report on the subject of N-phosphino function-
alized NHC ligands included a considerable computational
study into the behavior of a NHCP ligand and a ruthenium
center. While the reader may refer to this publication for a full
description of our results, one computational finding relevant to
the present work is briefly described and presented in Scheme
1. Beginning with a five-coordinate Ru-NHCP alkylidene
dichloride species, X1, we showed computationally that
phosphine dissociation leading to X2, was a reasonably facile
reaction, while carbene-C dissociation to X5 was energetically
prohibitive, as expected from the known chemistry of NHC vs
phosphine ligands. Species X2 can be considered comparable to
the active 14 VE species of a second generation Grubbs type
catalyst system. From species X2 the catalytic cycle may
proceed via olefin coordination, X3, and metallacyclobutane
formation, X4, leading to catalytic olefin metathesis.
We report herein the synthesis of bidentate NHCP ligands of
the form 2a−c, a convenient entry into their ruthenium
coordination chemistry and the preparation of the correspond-
ing ruthenium alkylidene complexes. We describe the synthesis
of a sufficiently reactive ruthenium precursor for the formation
of these ruthenium alkylidenes from diazo reagents. Trime-
thylsilyl-alkylidene complexes are presented first, synthesized
from the commercially available trimethylsilyldiazomethane,
and their behavior with respect to catalysis is described.
Furthermore, an example of a benzylidene complex has been
prepared and is compared to its trimethylsily substituted
congeners.
RESULTS AND DISCUSSION
■
Ligand Synthesis. Our research group and others have
recently developed a simple, high yielding, and scalable
synthesis of N-phosphino-functionalized N-heterocyclic car-
bene ligands (NHCPs). The NHCP salt precursor, 1a, and its
corresponding free NHCPs, 2a and 2c, have been previously
described in the literature, while compounds 1b, 1c, and 2b
have been newly prepared during the course of this study, using
slight modifications of reported procedures (Scheme 2). The
newly prepared NHCP compounds showed a similar ease of
synthesis with the exception of the N-tert-butyl-N′-di-tert-butyl-
B
DOI: 10.1021/acs.inorgchem.5b00513
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Scheme 2. Synthesis of NHCP Salts and Free NHCP Ligands 2a−c
a
Yields are given in brackets for newly prepared compounds.
phosphino-functionalized N-heterocyclic carbene 2c (Scheme
2), which could be only isolated as a viscous oily compound
which hampers complete purification. Subsequent reactivity,
however, was always unaffected. As mentioned above, the
C2H5-analogue of 2a (R = Et instead of Me) was also
accessible, although via a route different from the one shown in
Scheme 2, as disclosed earlier,12c but this compound was not
part of the present study.
confirmed via single crystal X-ray crystallography (Figure 3).
The structure of 3a shows that it adopts a distorted square
The new NHCP salt 1b has been fully characterized. The
solid state molecular structure of 1b displays the expected
characteristics and is shown in Figure 2. Furthermore, NMR
Figure 2. ORTEP plot of the cation of imidazolium triflate salt 1b.
Thermal ellipsoids are at the 30% probability level. Hydrogen atoms
and the triflate anion are omitted for clarity. Selected bond length (Å):
P1−N2 = 1.788(5) and angle (deg): C1−N2−P1 = 119.46 (44).
Figure 3. ORTEP plot of ruthenium complex 3a. Thermal ellipsoids
are at the 30% probability level. Hydrogen atoms are omitted for
clarity. Selected bond lengths (Å) Ru1−C1 = 1.996(7); Ru1−P1 =
2.303(2); Ru1−P2 = 2.225(2); Ru1−Cl1 = 2.405(2); Ru1−Cl2 =
2.464(2); P1−N2 = 1.743(6) and angles (deg) P1−Ru1−C1 =
67.6(2); P1−N2−C1 = 100.3(4).
spectroscopy of 1b had a 31P NMR signal at 120.8 ppm, which,
upon deprotonation, shifted upfield to 97.2 ppm as expected.
The previously reported NHCP compounds used in this study
otherwise match the literature characterization data.12a It was
expected, outside of noninnocent behavior as we have
previously described, that such NHCP ligands will form four-
membered rings upon ruthenium coordination.13
pyramidal geometry about the ruthenium center. The NHCP
ligand coordinates via the expected four-membered chelate
structure with two trans-disposed chlorides completing the base
of the square pyramidal configuration. One coordinated PPh3
forms the apex of the distorted square pyramid. The angles
about the base sum to 355.3° showing the slight distortion. The
NHCP-carbon−ruthenium bond distance was found to be
1.996(7) Å with a NHCP-phosphorus−ruthenium bond
distance of 2.3032(19) Å. The Ru−Cl bond distances are
2.4636(18) Å and 2.4051(18) Å trans to carbon and
phosphorus, respectively. The Ru−PPh3 bond distance is
2.2247(19) Å. The Ru-NHCP bite angle was determined to be
67.6(2)°, which confirms the expected bidentate nature of our
NHCP ligands.
NHCP Coordination. A high yielding synthetic entry into
mononuclear 16 VE ruthenium dichloride chemistry using
Ru(PPh3)3Cl2 as a convenient starting point was developed.
Mixing an equimolar solution of Ru(PPh3)3Cl2 and 2a in THF
afforded 3a in 86% isolated yield within 1 h at room
temperature. The initial red-brown color of Ru(PPh3)3Cl2 is
quickly lost, and a deep purple solution is formed. The product
precipitates from THF forming microcrystalline bulk material.
Analysis of an NMR scale reaction shows the complete
consumption of 2a and the liberation of free PPh3 via 31P NMR
spectroscopy. The reaction can be scaled up easily and can be
run as efficiently on a multigram scale. Analysis of isolated 3a
displays two coupled doublets in the 31P{1H} NMR. A doublet
at 104.3 ppm corresponds to the coordinated NHCP
phosphorus fragment, while a doublet at 77.3 ppm corresponds
to one remaining coordinated PPh3. One coordinated PPh3 is
Complexes 3b and 3c were prepared using a similar synthetic
procedure as for 3a. 3b and 3c were isolated in 79% and 56%
yields, respectively. Changing the supporting NHCP ligand
i
t
from R = Me to Pr to Bu systematically reduces the isolated
yield due to the increased solubility and difficulty of PPh3
separation. However, monitoring of the reaction via NMR
1
also implied by the H NMR of 3a relative to one coordinated
NHCP ligand (2a). The molecular structure of 3a was
C
DOI: 10.1021/acs.inorgchem.5b00513
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
The synthesis of 3a, 3b, and 3c stands in contrast to our
attempts toward the synthesis of N-aryl-NHCP ruthenium
complexes. We have previously reported that the reaction of
Ru(PPh3)3Cl2 with N-mesityl-NHCPs leads uncontrollably to
an octahedral bis-chelation ruthenium trans-dichloride com-
plex.12c It was clear that alternative synthetic pathways will thus
be required to obtain the desired monochelation N-aryl-NHCP
ruthenium complexes.
Scheme 3. Synthesis of 3a−c
Synthetic Variations. Indeed, complexes 3a−c provided a
convenient entry into 16 VE Ru(II) coordination chemistry,
although complexes 3a−c proved to be substitutionally rather
inert. Screening reactions of 3a−c with trimethylsilyldiazo-
methane and phenyldiazomethane showed only small con-
versions to what could be considered the desired products 5a−
c and 6a−c via 1H and 31P NMR spectroscopy. On the basis of
these observations it was decided that a more labile ligand than
PPh3 would be desirable. Pyridine and substituted pyridines
have been shown to be widely useful in transition metal
catalysis as labile spectator ligands in a variety of applications,
e.g., in catalytic olefin metathesis and in transition metal
mediated hydrogenation, because they are more weakly bound
to the metal center and thus ease ligand substitution.3a,15
Using complex 3a as a test system in order to assess the
efficacy of PPh3-pyridine ligand exchange reactions and
subsequent metal complex reactivity, we examined the role of
pyridine, 2-methylpyridine (2-picoline), and 3-bromopyridine.
These three examples should provide an electronic (3-
bromopyridine), and a steric (2-methylpyridine) modification,
in contrast to pyridine itself. The reaction of 3a with neat
pyridine showed an immediate color change from purple to
orange and the precipitation of an orange solid. Examination of
the crude reaction mixture, 4aH, via 31P NMR spectroscopy
displayed no further 31P{1H} coupling, the presence of
uncoordinated PPh3, and several signals between 120 and
105 ppm, likely corresponding to a coordinated NHCP ligand
in deuterated dichloromethane as solvent. Analysis of a purified
sample of 4aH in deuterated CD2Cl2 showed four singlet signals
a
Yields are given in brackets for newly prepared compounds.
spectroscopy shows similar conversions. 3b displays two
doublets in the 31P{1H} NMR spectrum, 108.9 and 77.7
ppm, indicative of a similar coordination mode about the
ruthenium center as 3a. The 1H NMR spectrum shows a slight
broadening of the aromatic proton signals attributed to the
PPh3 ligand. 3b was characterized via single crystal X-ray
diffraction (Figure 4, left) and reveals an analogous geometry as
3a. The NHCP-carbene−ruthenium bond distance was found
to be 2.024(4) Å, while the NHCP-phosphorus−ruthenium
bond distance was found to be 2.3063(12) Å. A PPh3 was again
the apex of a slightly distorted square pyramid with a Ru−PPh3
bond distance of 2.2199(12) Å. 3c displays similar spectro-
scopic characteristics as 3a and 3b with one notable difference
in its solution 1H NMR. While the 31P{1H} NMR displays two
1
doublets at 110.8 and 72.6 ppm, the H NMR is significantly
broadened especially for the aromatic protons of the PPh3. This
would suggest inhibited rotation about the Ru−PPh3 bond.
This observation is only significant for the most bulky Ru-
NHCP complex 3c (Figure 4, right). An NHCP-carbene−
ruthenium bond distance of 2.048 Å, the average for two
molecules in the asymmetric unit, and the NHCP-phosphorus−
ruthenium bond of 2.280 Å, again the average for two
molecules in the asymmetric unit, were not found to be
significantly different for 3c compared to those found in 3a and
3b. Furthermore, the Ru−PPh3 bond distance found in 3c is
not significantly different from those distances found in 3a and
3b.
1
in the 31P NMR spectrum. The H NMR also showed four
distinct sets of signals. The NMR spectra of an identical sample
Figure 4. ORTEP plot of ruthenium complexes 3b and 3c. Thermal ellipsoids are at the 30% probability level. Hydrogen atoms are omitted for
clarity. Selected 3b bond lengths Ru1−C1 = 2.024(4); Ru1−P1 = 2.306(1); Ru1−P2 = 2.220(1); Ru1−Cl1 = 2.416(1); Ru1−Cl2 = 2.469(1); P1−
N2 = 1.749(3) and angles (deg) P1−Ru1−C1 = 67.2(1); P1−N2−C1 = 99.6(3) 3c bond lengths Ru1−C1 = 2.04(1); Ru1−P1 = 2.281(4); Ru1−P2
= 2.227(4); Ru1−Cl1 = 2.419(3); Ru1−Cl2 = 2.445(4); P1−N2 = 1.72(1) and angles (deg) P1−Ru1−C1 = 67.7(4); P1−N2−C1 = 101.9(8).
D
DOI: 10.1021/acs.inorgchem.5b00513
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Scheme 4. Synthesis of Ruthenium Pyridine Complexes 4a−c
a
Yields are given in brackets for newly prepared compounds.
recorded in deuterated pyridine, however, displayed far fewer
signals. In deuterated pyridine only two signals are visible in the
31P{1H} NMR spectrum at 113.9 and 107.5 ppm. Furthermore,
1H NMR spectroscopy revealed proton signals which
corresponded to only two compounds. For example, two
signals at 4.26 and 4.68 ppm indicated two different N−CH3
groups. From this we conclude that different mixtures of
isomers are formed, the relative concentration of which is
strongly dependent on the solvent. In deuterated pyridine two
isomers are present in an approximately 70:30 ratio based on
1H NMR. One may imagine several isomers of a complex such
as 4aH particularly if in noncoordinating solvents one
equivalent of pyridine is not coordinated to the ruthenium
center. Our observations support that in dichloromethane the
complex is likely in equilibrium with a species missing one
pyridine by dissociation, or at least with one pyridine not
strongly bound, and that the addition of pyridine forces
coordinative saturation of the metal center, thus reducing the
number of isomers. The solid state structure of 4aH which was
unambiguously assigned using single crystal X-ray diffraction,
and elemental analysis further supports our structural assign-
ment. The diffraction study revealed an octahedral 18 VE
ruthenium species (Figure 5). The structure showed two trans-
disposed pyridines, a bidentate equatorial NHCP ligand, and
two cis-chlorides with the pyridine ligands showing slight
bending away from the NHCP ligand. The NHCP-carbene
ruthenium bond length was determined to be 2.009(3) Å, the
NHCP-phosphorus ruthenium bond length was found to be
2.3308(7) Å, and the Ru−Cl bond lengths were determined to
be 2.5200(7) Å and 2.4673(7) Å trans to the carbene and
phosphorus, respectively. The C−Ru−P bite angle of the
NHCP ligand was determined to be 66.16(7)°. These values
represent only slight deviations from the bond lengths and bite
angle found in 3a. The ruthenium-pyridine Ru−N distances are
2.099(2) Å and 2.132(2) Å, which are similar to those found in
analogous ruthenium complexes, such as SIMesRuCl2py2.16 We
carried out screening reactions with pyridine complex 4aH to
form the desired complexes 5a and 6a. In test reactions with
trimethylsilyldiazomethane and phenyldiazomethane only a
marginal increase in reactivity was observed over 3a;
furthermore, 4aH was observed to be relatively insoluble in
Figure 5. ORTEP plot of ruthenium complex 4aH. Thermal ellipsoids
are at the 30% probability level. Hydrogen atoms are omitted for
clarity. Selected bond lengths (Å) Ru1−C1 = 2.009(3); Ru1−P1 =
2.3308(7); Ru1−N21 = 2.099(2); Ru1−N31 = 2.132(2); Ru1−Cl1 =
2.5200(7); Ru1−Cl2 = 2.4673(7); P1−N2 = 1.742(2) and angles
(deg) P1−Ru1−C1 = 66.16(7); P1−N2−C1 = 99.1(2).
common solvents. The reactions of 4aH with diazo reagents
consistently resulted in the synthesis of some so far unidentified
ruthenium products and substantial amounts of remaining
starting material (4aH). Considering the observation that 4aH
had little improvement in terms of reactivity we did not carry
out further studies into the behavior of 4aH.
The mixing of 3a with 2-methylpyridine (2-picoline), either
neat or using dichloromethane as a solvent, did not result in an
immediate reaction. In fact, as opposed to pyridine, 3a does not
dissolve in 2-methylpyridine and dichloromethane is required
to solubilize 3a. Prolonged heating and mixing of a mixture of
3a and 2-methylpyridine also resulted in limited conversion, no
new isolable products, and eventually to the production of
ruthenium black. Presumably, ortho-substituted pyridines are
too bulky to coordinate efficiently to 3a and to replace PPh3.
In an effort to assess the electronic affects of electron
withdrawing substituents on pyridine, we explored the
formation of ruthenium 3-bromopyridine complexes. The
mixing of 3a with 3-bromopyridine resulted in an immediate
color change from purple to orange, analogous to what was
observed during the synthesis of 4aH. The NMR of an isolated
sample of 4aBr once again revealed a complex spectrum. This
E
DOI: 10.1021/acs.inorgchem.5b00513
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
observation is not entirely surprising given the expected
increased lability of the 3-bromopyridine ligands relative to
pyridine. The molecular structure was confirmed by single
crystal X-ray diffraction. 4aBr displayed a distorted octahedral
geometry. The molecular structure (Figure 6) was similar to
Figure 7. ORTEP plot of ruthenium complex 4cBr. Thermal ellipsoids
are at the 30% probability level. Hydrogen atoms are omitted for
clarity. Selected bond lengths: Ru1−C1 = 2.060(4); Ru1−P1 =
2.296(1); Ru1−N21 = 2.257(4); Ru1−N31 = 2.169(3); Ru1−Cl1 =
2.432(1); Ru1−Cl2 = 2.417(1); P1−N2 = 1.740(4) and angles (deg)
P1−Ru1−C1 = 66.4(1); P1−N2−C1 = 99.5(3).
ligands complete the axial coordination sites of the octahedral
framework. An NHCP-carbene−ruthenium bond distance of
2.060(4) Å was found, with NHCP-phosphorus−ruthenium
and the two ruthenium-chloride bond distances of 2.296(1) Å,
2.417(1) Å, and 2.433(1) Å, respectively. In this case, the
ruthenium 3-bromopyridine Ru−N bond distances of 2.169(3)
Å and 2.257(4) Å trans to the NHCP carbene and phosphorus
correspondingly were found. These data show that there is a
considerable lengthening of the ruthenium−pyridine bonds in
4cBr in comparison to that found in 4aBr. The NMR spectra of
both 4bBr and 4cBr are equally complex to that observed for
4aBr.
Synthesis of Ruthenium Alkylidene Complexes.
Following a similar synthetic procedure as that used in the
initial synthesis of Grubbs precatalysts, we employed
derivatized diazomethane reagents, specifically trimethylsilyl-
diazomethane and phenyldiazomethane, as synthetic precursors
for Ru alkylidene species. Initially we explored the synthesis of
ruthenium complexes bearing trimethylsilyl alkylidene function-
alities derived from commercially available trimethylsilyldiazo-
methane for synthetic ease and efficiency. While less commonly
used as an olefin metathesis initiator than benzylidene- and
indenylidene-based complexes, there are reports of the effective
use of trimethylsilyl alkylidenes in catalytic olefin metathesis.17
Complexes 4aBr, 4bBr, and 4cBr all showed improved reactivity
with diazomethane reagents in small scale reactivity screening
when compared to the previously prepared complexes in this
study; therefore, complexes 4aBr, 4bBr, and 4cBr were more
closely examined as precursors to ruthenium-alkylidene
complexes.
Figure 6. ORTEP plot of ruthenium complex 4aBr. Thermal ellipsoids
are at the 30% probability level. Hydrogen atoms are omitted for
clarity. Selected bond lengths: Ru1−C1 = 1.99(1); Ru1−P1 =
2.347(3); Ru1−N21 = 2.087(8); Ru1−N31 = 2.127(8); Ru1−Cl1 =
2.471(3); Ru1−Cl2 = 2.526(3); P1−N2 = 1.738(9) and angles (deg)
P1−Ru1−C1 = 66.4(3); P1−N2−C1 = 99.8(7).
that found for 4aH. The NHCP ligand is situated in a plane with
the cis-positioned chloride ligands. Two trans-pyridines
complete the octahedron. The NHCP-carbene−ruthenium
bond distance is 1.990(11) Å, the NHCP-phosphorus−
ruthenium bond distance was determined to be 2.347(3) Å,
and the Ru−Cl bond distances were found to be 2.526(2) Å
and 2.472(2) Å trans to the carbene and phosphine,
respectively. The NHCP-Ru bite angle amounts to 66.4(3)°,
and the ruthenium 3-bromopyridine Ru−N bond distances
were found to be 2.087(8) Å and 2.128(8) Å. Comparing the
experiential ruthenium−pyridine bond distances between 4aH
and 4aBr shows that the electron withdrawing 3-bromopyridine
has little to no effect on the ruthenium−pyridine bond
distances. Complex 4aBr shows improved solubility over
complex 4aH.
We also investigated the synthesis of complexes 4bBr and
4cBr, more bulky analogues of 4aBr. Dissolving 3b in a
minimum amount of 3-bromopyridine resulted in a quick
reaction to form a new orange complex 4bBr. However, so far it
was not possible to obtain suitable single crystals for a solid
state structure determination; therefore, we cannot definitively
assign the molecular structure, but NMR spectroscopy suggests
a similar coordination sphere as 4aBr. Finally, dissolving 3c in a
minimum amount of 3-bromopyridine also yielded a new
orange complex, 4cBr. The molecular geometry of this species
could be again assigned via single crystal X-ray diffraction to
reveal a nearly ideal octahedral geometry. Remarkably, the
structure of 4cBr in the solid state (Figure 7) differs compared
to that found for 4bBr. 4cBr displays a plane consisting of a
NHCP ligand and two 3-bromopyridine ligands. Two chloride
A solution of 4aBr in dichloromethane cooled to −40 °C was
reacted with a commercial 2 M solution of trimethylsilyldiazo-
methane (in hexanes) with warming to room temperature.
Nearing room temperature the solution began to bubble with a
slow change of color from purple to green. Isolating the green
product, 5a, from THF and analysis via NMR spectroscopy
revealed the formation of one new complex. Most indicatively,
a doublet at 18.61 ppm in the proton NMR suggests the
formation of a ruthenium alkylidene bond as desired.
F
DOI: 10.1021/acs.inorgchem.5b00513
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Scheme 5. Synthesis of Ruthenium Trimethylsilyl-Alkylidene Complexes 5a−c and 6c
a
Yields are given in brackets for newly prepared compounds.
Furthermore, a new TMS signal is visible at 0.38 ppm, one
signal for the NHCP methyl substituent is observed at 3.52
ppm, two signals for the imidazole backbone, and two dissimilar
tBu signals for the NHCP ligand were present as expected. The
31P NMR showed one singlet for the di-tert-butylphosphine
fragment of the NHCP chelate ligand. Crystals suitable for a
single crystal X-ray diffraction study allowed for the
unambiguous determination of the molecular structure and
revealed a distorted square pyramidal arrangement (Figure 8).
As in the case of 4aBr we explored the preparation of
ruthenium TMS-alkylidene complexes using 4bBr and 4cBr. The
reaction of 4bBr with trimethylsilyldiazomethane, with con-
ditions similar to those used to prepare 5a, yielded a new green
complex (5b). Analysis of the sample with NMR once again
1
showed the formation of a new alkylidene signal in the H
NMR at 18.56 ppm. A single crystal analyzed via X-ray
crystallography revealed a structure similar to that found for 5a.
Similar metrical parameters as in 5a characterize the molecular
structure of 5b (Figure 9). The ruthenium−alkylidene bond
Figure 8. ORTEP plot of ruthenium complex 5a. Thermal ellipsoids
are at the 30% probability level. Hydrogen atoms are omitted for
clarity. Selected bond lengths: Ru1−C1 = 1.987(6); Ru1−P1 =
2.288(2); Ru1−C30 = 1.829(6); Ru1−Cl1 = 2.438(1); Ru1−Cl2 =
2.395(2); P1−N2 = 1.737(4) and angles (deg) P1−Ru1−C1 =
67.2(2); P1−N2−C1 = 99.0(3).
Figure 9. ORTEP plot of ruthenium complex 5b. Thermal ellipsoids
are at the 30% probability level. Hydrogen atoms are omitted for
clarity. Selected bond lengths: Ru1−C1 = 2.010(2); Ru1−P1 =
2.2962(4); Ru1−C30 = 1.829(2); Ru1−Cl1 = 2.4400(5); Ru1−Cl2 =
2.3773(5); P1−N2 = 1.732(2) and angles (deg) P1−Ru1−C1 =
66.81(5); P1−N2−C1 = 99.7(1).
The NHCP ligand along with two chloride ligands forms the
base of the square pyramidal structure with the alkylidene
group positioned as the apex. An NHCP-carbene−ruthenium
bond distance of 1.987(6) Å was determined, the NHCP-
phosphorus−ruthenium bond distance is 2.288(2) Å, and the
ruthenium chloride bonds distances were found to be 2.438(1)
Å and 2.395(2) Å trans to the carbene and phosphorus,
respectively. The NHCP bite angle amounts to 67.2(2)°, while
the sum of the angles at the metal in the base of the square
pyramid is 352.7°, showing a slight distortion of the complex.
The ruthenium alkylidene bond distance of 1.829(6) Å is one
of only two known structurally characterized examples of
ruthenium TMS-alkylidene complexes, the other example
having the TMS-alkylidene unit at the apex of a ruthenium
porphyin complex.18 Regardless, the ruthenium alkylidene bond
distance of 1.829(6) Å is very similar to the ruthenium
alkylidene bond distance of 1.839(3) Å found in the second
generation Grubbs catalyst.19
length is 1.829(2) Å, while the NHCP bite angle was found as
66.81(5)°. The distortion of the square pyramidal structure was
similar to the case of 5b, with the sum of the angles around the
basal Ru-position of the square pyramid equal to 352.63°.
While we were unable to obtain a molecular structure of 5c,
the NMR data of 5c were fully consistent with a molecular
geometry as those found in 5a and 5b. The 1H NMR revealed a
t
new alkylidene signal at 18.98 ppm, dissymmetric Bu signals
for the di-tert-butylphosphino fragment, and one new TMS
signal for the trimethylsilylalkylidene.
The synthesis of 5c appeared to be the most straightforward,
high yielding, and scalable route to a ruthenium TMS-
alkylidene and therefore was used first to synthesize Ru-
NHCP benzylidene complexes. The synthesis of 5c could be
undertaken on the gram scale, while the synthesis of 5a and 5b
G
DOI: 10.1021/acs.inorgchem.5b00513
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 10. Performance of 6c (left) and 5c (right) in the RCM of diethyl diallylmalonate. Condition (0.1 M, cat. 1 mol %, bromobenzene 0.75 mL).
Reaction monitored every 3 min using high temperature NMR.
did not scale well beyond the 200 mg scale (without further
optimization). The mixing of 4cBr and phenyldiazomethane in a
dichloromethane solution cooled to −40 °C resulted in the
vigorous bubbling of the reaction mixture and in a color change
from orange to deep green. Using procedures similar to those
used in the synthesis of 4cBr yielded a new product 6c. NMR
analysis showed the clean formation of a single new species. It
displayed a new 1H NMR resonance at 15.77 ppm typical for a
ruthenium-alkylidene fragment. This signal differs significantly
from that observed in 5c, 3.21 ppm upfield of that observed for
4cBr. The 31P NMR signal for 6c was detected at 130 ppm
compared to a 31P signal at 137 ppm for 5c.
We have thus far been unable to prepare the complexes 6a
and 6b. Following both identical as well as modified procedures
varying the temperature, solvent, and reaction time screening or
modifying the workup strategies compared to those employed
in the synthesis of 5a−c and 6c resulted in mixtures of
unidentified products. Importantly, no new 1H alkylidene
signals were observed. The reason for the contrasting reactivity
of 5c versus 5a and 5b is elusive at this point. Two notable
observations are that 5c is significantly more soluble than the
poorly soluble complexes 5a and 5b. Furthermore, the solid
state structure of 5c showed the pyridine ligands in the plane of
the NHCP ligand, while the solid state structure of 5a showed
the pyridine ligands in the apical positions of the octahedral
structure.
Screening of NHCP Ruthenium Alkylidene Complexes
in Catalysis. As mentioned above, the motivation to
synthesize Ru alkylidene complexes like 6a−c was based
upon their similarity to Grubbs II precatalysts. Contrasting
those, our systems contain a chelate-enforced cis-coordination
mode of an NHC carbon and a phosphine donor as parts of a
presumably strained four-membered Ru−P−N−C chelate-ring,
which may open a pathway from a pentacoordinate, square
pyramidal 16 VE precatalysts like 6a−c to tetracoordinate,
catalytically active 14 VE species via Ru−P dissociation. These
four-coordinate, highly reactive intermediates created from, e.g.,
6a−c, would obviously be structurally identical to the active
species formed from Grubbs II precatalysts by phosphine (e.g.,
PCy3) dissociation. As we knew from preliminary density
functional theory (DFT) calculations on model systems and
from established chemistry that the alternative Ru−C bond
dissociation of the NHCP chelate rings should not compete
with P-dissociation, we had set out to make appropriate Ru-
NHCP-alkylidenes as described.
To compare the catalytic efficacy of our Ru-NHCP
complexes, we focused our efforts on complexes 5c and 6c
due to easy access to larger amounts of these complexes. This
choice also was expected to allow an evaluation of the influence
of the alkylidene functionality and its impact on catalysis. To
benchmark the catalytic potential of 5c and 6c, we investigated
the ring closing metathesis (RCM) of diethyl diallylmalonate,
based on the standard catalytic conditions suggested by Grubbs,
and the ring opening metathesis polymerization (ROMP)/ring
expansion (RE) of cyclooctene, a reaction in which we have a
long-standing interest.20
Both complexes 5c and 6c did not effect the RCM catalysis
of diethyl diallylmalonate at room temperature. Using variable
temperature NMR to investigate this RCM test reaction, we
found that some catalysis is observed beginning at 60 °C,
whereas at 80 °C we were able to observe appreciable catalytic
activity. For the high temperature NMR studies bromobenzene-
d5 was used as the deuterated reaction solvent due to its
convenient temperature range and the good solubility of the
ruthenium complexes. The RCM reaction was monitored at 80,
90, and 100 °C using high temperature NMR and measuring
spectra approximately every 3 min. Figure 10 shows that over
the course of 1 h the conversion is 15%, 27%, and 48% at 80,
90, and 100 °C, respectively for 6c and 10%, 16%, and 27% at
80, 90, and 100 °C, respectively, for 5c. The rate of conversion
over the course of 1 h is approximately linear with a slight fall
off of activity over time at 100 °C, which may be attributed to
catalyst decomposition. This pattern holds such that at 120 °C
the RCM of diethyl diallylmalonate is 94% complete within the
span of 1 h, catalyzed by 6c. These findings demonstrate that
our catalyst system is behaving in a latent fashion. Actually, few
olefin metathesis systems have been shown to behave this way.
One notable recent example is a report by Cazin and co-
workers.21 Furthermore, these data demonstrate that while a
TMS substituted alkylidene is less competent at the RCM of
diethyl diallylmalonate than a phenyl substituted alkylidene, it
can still provide access to an active catalyst system via a
commercially available alkylidene precursor.
The ROMP/RE of cis-cyclooctene is a reaction with
significant industrial interest.20b,22 It is used to produce an
industrially important polymer in the polymer blending field.23
H
DOI: 10.1021/acs.inorgchem.5b00513
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 11. ROMP/RE of cis-cyclooctene. Solid line: conversion of cis-cyclooctene with respect to time. Dashed line: selectivity for 1,9-
cyclohexadecadiene with respect to time. Yield and selectivity determined by GC. 1 mol % catalyst loading, 0.1 M concentration of cycloctene in
bromobenzene.
The controlled RE of cyclooctene, on the other hand,
represents an entry point into the production of high value
macrocyclic olefins, attractive precursors in the fragrance and
flavor industry. A key example is the dimerization of
cyclooctene to 1,9-cyclohexadecadiene. A substantial challenge
is the controlled dimerization avoiding subsequent telomerisa-
tion and/or polymerization. The current heterogeneous process
for the formation of 1,9-cyclohexadecadiene produces signifi-
cant amounts of byproducts, and selective homogeneous
alternatives would be advantageous. It has been noted that
dissymmetric ligands, such as those reported herein, may be
important in the improvement of the desired C16-selectivity.20a
Similar to our observations with the RCM of diethyl
diallylmalonate, the ROMP/RE of cyclooctene also required
elevated temperatures. Bromobenzene was again utilized as the
solvent because of its ability to solubilize the catalyst and its
high boiling point. Running the ROMP/RE of cyclooctene with
catalyst 6c in Teflon-sealed reaction vessels, sampling revealed
the reaction profile shown in Figure 11. At 100 °C over the
course of 5 h 60% conversion was obtained, while at 120 °C a
maximum conversion of 75% was obtained after 4 h. These
observations also demonstrate that our catalyst system 6c
behaves in a latent fashion. Furthermore, at these conditions we
achieve a peak selectivity of 21% for 1,9-cyclohexadecadiene at
100 °C with decreasing selectivity over time. While the
selectivity is lower than one of our previously reported
dissymmetric NHC systems, this compares favorably with
respect to heterogeneous Re2O7, with a selectivity of 6% at
70% conversion, and with heterogenized Grubbs type systems,
which typically display conversions in the range of 20% under
flow conditions.20b Upon completion of the reaction, removal
of the solvent, and analysis of the resulting material by GPC
analysis, we found that all samples were composed of a broad
mixture of cyclooctene oligomers/polymers. Relevant polymer
data are presented in Table 1 showing broad PDIs, very low Mn
and Mw values ranging from approximately 2000−43000 g/mol.
Conclusions. Herein we have described an entry into
square pyramidal 16 VE ruthenium alkylidene complexes via
appropriate precursor compounds, using N-phosphino-func-
Table 1. Data for the Polymer Obtained by Polymerization
of cis-Cyclooctene with 6c
temp [°C]
Mn [g/mol]
Mw [g/mol]
PDI
100
120
714
425
42.700
3.460
59.8
8.1
tionalized NHC ligands (NHCP) forming four-membered
chelate structures, and preliminary catalytic screening of some
of these species is reported. Ruthenium NHCP dichloride PPh3
complexes, prepared in high yielding syntheses, which can be
easily scaled up, formed the basis for this novel ruthenium
coordination chemistry. From these complexes the successful
synthesis of the desired ruthenium alkylidene complexes has
been demonstrated. Ruthenium NHCP complexes have been
shown to catalyze both the ring closing metathesis of diethyl
diallylmalonate and the metathesis ring expansion of cis-
cyclooctene. The new ruthenium-NHCP-alkylidene complexes
provide a second generation of the initial bis-phosphine olefin
metathesis catalysts initially reported in 1999 by Hofmann and
co-workers.11b Ongoing efforts in our laboratory are directed
toward extending the structural manifold of this family of four-
membered chelating (NHCP)Ru-alkylidenes by synthesis,
guided by further DFT studies, toward more extensive catalyst
screening, understanding Ru-NHCP catalytic activity, and
subsequent catalyst optimization. The results will be presented
in due course.
General Considerations. All manipulations were carried
out under an atmosphere of dry argon using standard Schlenk
techniques or in an MBraun glovebox. Solvents were either
purchased dry from Aldrich or dispensed from an MBraun SPS-
800 solvent system. Solvents were stored over either sodium
ingots or molecular sieves (4 Å). Solvents were degassed either
via 3-freeze−pump−thaw cycles or sparging with argon for 30
min. Previously reported phosphorylated NHC salts, their
corresponding free NHCs were prepared according to literature
procedures.12a,24 All other solvents and reagents where
purchased from Aldrich or ABCR chemicals. The cyclooctene
was distilled from sodium prior to use, and the diethyl diallyl
I
DOI: 10.1021/acs.inorgchem.5b00513
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
malonte was used as received. NMR spectra were recorded
using a Bruker 200 or a Bruker 600 spectrometer. Chemical
shifts are given in ppm referenced to solvent (1H, 13C) and
relative to 85% H3PO4 for 31P NMR. Abbreviations used are s =
singlet, d = doublet, sept = septet, br = broad. J-coupling
constants are reported in Hertz (Hz). Elemental analyses were
performed by the “Mikroanalytisches Laboratorium der
for a total substrate concentration of 0.1 M. The Teflon sealed
vessel was heated to 100 °C and samples taken at 5 min, 30
min, 1 and 2 h under a stream of Argon. The aliquot was
filtered through a small amount of silica gel and analyzed via
GC.
Selectivity was defined as follows: selectivity in product i. (Si)
= (number of mole of cyclooctene converted in product (i)/
(total number of cyclooctene converted).
Chemischen Institute der Universitat Heidelberg”. Gas
̈
Synthesis of N-Isopropyl-N′-di(tert-butyl)phosphino-imi-
dazolium Triflate (1b). To a solution of N-isopropylimidazole
(7.71 g, 70 mmol) and CF3OSO2Na (12.65 g, 73.5 mmol) in
THF (75 mL) was added at −10 °C di(tert-butyl)-
chlorophosphine (13.28 g, 73.5 mmol) dissolved in THF (50
mL). The reaction mixture was allowed to warm to room
temperature and stirred for 24 h. The solvent was removed in
vacuo, the solid residue was extracted in CH2Cl2 (100 mL), the
insoluble solid was filtered off, washed with CH2Cl2 (3 × 10
mL), and the filtrate was concentrated in vacuo until crystal
started to form then cooled to −40 °C to yield colorless crystals
chromatography was performed on an Agilent 6890N modular
GC base equipped with a split-mode capillary injection system
and a flame ionization detector using a BGB-5 capillary column
(BGB Analytik Vertrieb 20530-025; 30 m × 0.25 mm; He flow
1.0 mL/min, program: initial 50 °C for 5 min, ramp 10 °C/min,
300 °C for 15 min, ramp 10 °C/min, 320 °C for 8 min).
Starting materials and products had following retention times:
cyclooctene (tR = 10.8 min), dodecane (tR = 17.7 min),
bromobenzene (tR = 12.4 min), 1,9-cyclohexadecadiene (tR =
25.8 min), and 1,9,17-cyclotricosatriene (tR = 34.5 min). The
response correction is detailed in the Supporting Information.
GPC analysis was performed by BASF SE. The GPC was
calibrated using polystyrene standards in the range of M = 580
bis M 6 870 000. The particulars are as follows: elution solvent
THF, temperature of 35 °C, flow rate of 1 mL/min, injection
volume of 100 μL, and a concentration of 2 mg/mL. The
detector used was a DRI HP 1100 UV Agilent 1100 MWD
[254 nm].
For the X-ray diffraction studies data sets were collected on a
Bruker APEX-II Quazar CCD diffractometer with Mo Kα
radiation (λ = 0.71073 Å). Intensities were corrected for
Lorentz and polarization effects, an empirical absorption
correction was applied using SADABS1 based on the Laue
symmetry of the reciprocal space. Structure solved by direct
methods and refined against F2 with a Full-matrix least-squares
algorithm using the SHELXTL (Version 2008/4 or 2013/3)
software package.25
CCDC 1045646 (1b), 1045647 (3a), 1045648 (3b)
1045649 (3c), 1045650 (4aH), 1045651 (4aBr) 1045652
(4cBr), 1045653 (5a), and 1045654 (5b) contain the
supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
Catalysis. The ring closing metathesis reactions were
performed in line with the protocols described by Grubbs
and co-workers. An example procedure is as follows: A stock
solution of catalyst 6c was prepared. 6c (10 mg, 0.0189 mmol)
was dissolved in CD2Cl2-d2 (1 mL). 50 μL of this solution was
added to an NMR along with bromobenzene-d5 (0.75 mL) and
diethyl diallylmalonate (23 μL, 0.094 mmol). The NMR tube
was immediately inserted into the NMR spectrometer
prewarmed to the desired temperature. The displayed NMR
temperature was correlated to internal temperature using
ethylene glycol in accordance to the procedures described by
Bruker Instruments. For example, for a catalytic run at 80 °C
the spectrometer was set to 74 °C.
1
1
to give 1b (24.06 g, 85%). H NMR (CD2Cl2): 8.98 (s, H,
1
1
NCHN), 7.67 (s, H, Cimidazole), 7.56 (s, H, Cimidazole), 4.94
(sept, J = 6.7 Hz, 1H, CH(CH3)3), 1.58 (d, J = 6.7 Hz, 6H,
CH(CH3)3), 1.25 (d, J = 13.6 Hz, 18H, P(C(CH3)3)2). 19F
NMR (CD2Cl2): − 78.9. 31P{1H} NMR (CD2Cl2): 120.5.
13C{1H} NMR (CD2Cl2): 23.1 (s, CH(CH3)2), 28.6 (d, 15.5
Hz, C(CH3)3), 35.4 (d, 30.1 Hz, C(CH3)3), 54.05 (s, C(CH3)2
partly overlapping CD2Cl2 signal), 121.3 (q, 320.1 Hz, CF3SO3
quartet only partly observable), 121.6 (s, NCHCHN), 127.3 (s,
NCHCHN), 140.8 (br, NCHN) X-ray quality crystals could be
produced by the cooling of a solution of 1b in CH2Cl2 to −40
°C. Elemental analysis C15H28N2F3O3PS calculated: C 44.55%,
H 6.98%, N 6.93% found: C 44.65%, H 7.05%, N 6.87%.
Colorless crystal (polyhedron), dimensions 0.180 × 0.080 ×
0.050 mm3, crystal system orthorhombic, space group P212121,
Z = 4, a = 16.3820(11) Å, b = 10.5135(7) Å, c = 11.8675(8) Å,
α = 90°, β = 90°, γ = 90°, V = 2044.0(2) Å3, ρ = 1.314 g/cm3, T
= 200(2) K, θmax = 25.722°, radiation Mo Kα, λ = 0.71073 Å,
0.5° ω-scans with CCD area detector, covering the asymmetric
unit in reciprocal space with a mean redundancy of 6.03 and a
completeness of 100.0% to a resolution of 0.82 Å, 13 547
reflections measured, 3885 unique (R(int) = 0.0529), 3172
observed (I > 2σ(I)), μ = 0.28 mm−1, Tmin = 0.83, Tmax = 0.96,
226 parameters refined, hydrogen atoms were treated using
appropriate riding models, Flack absolute structure parameter
−0.01(5), goodness of fit 1.09 for observed reflections, final
residual values R1(F) = 0.065, wR(F2) = 0.134 for observed
reflections, residual electron density −0.56 to 0.59 e·Å−3.
Synthesis of N-tert-Butyl-N′-di(tert-butyl)phosphino-imi-
dazolium Triflate (1c). To a solution of N-tert-butylimidazole
(8.69 g, 70.0 mmol) and CF3OSO2Na (12.65 g, 73.5 mmol) in
THF (75 mL) was added at −10 °C di(tert-butyl)-
chlorophosphine (13.28 g, 73.5 mmol) dissolved in THF (50
mL). The reaction mixture was allowed to warm to room
temperature and stirred for 24 h. The solvent was removed in
vacuo, the solid residue was dissolved in CH2Cl2 (100 mL), the
insoluble solid was filtered off, washed with CH2Cl2 (3 × 10
mL), and the filtrate was concentrated in vacuo until crystals
started to form and then it was cooled to −40 °C to yield
colorless crystals to give 1c (26.66 g, 91%) 1H NMR (CD2Cl2):
8.69 (m, 1H, NCHN), 7.85 (m, 1H, CimidazoleH), 7.68 (m, 1H,
The catalytic ring expansion of cis-cyclooctene was
performed as in previous work in the area by Limbach and
co-workers.20a An example procedure is as follows: A stock
solution of catalyst 6c was prepared. 6c (10 mg, 0.0189 mmol)
was dissolved in bromobenzene (10 mL). Five milliliters of the
catalyst solution was added to a Teflon sealed reaction vessel
followed by addition bromobenzene (5 mL), cis-cyclooctene
(110 mg, 1 mmol), and dodecane (internal standard, 170 mg)
C
imidazoleH), 1.72 (s, 9H, NC(CH3)3), 1.26 (d, 18H, P(C-
(CH3)3)2. 19F NMR (CD2Cl2): − 78.8 31P{1H} NMR
(CD2Cl2): 120.2. 13C{1H} NMR (CD2Cl2): 28.6 (d, 16.0 Hz,
J
DOI: 10.1021/acs.inorgchem.5b00513
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
P(C(CH3)3)2), 29.9 (s, N(C(CH3)3), 35.5 (d, 30.0 Hz,
P(C(CH3)3)2), 61.6 (s, N(C(CH3)3), 121.4 (q, 321.4 Hz,
CF3SO3 quartet only partly observable), 122.7 (s, NCHCHN),
126.9 (s, NCHCHN), 139.7 (d, 33.6 Hz, NCHN). Elemental
analysis C16H30N2F3O3PS calculated: C 45.93%, H 7.23%, N
6.69% Found: C 45.78%, H 7.29%, N 6.92%.
Several attempts were made to get better EA data. A
representative example is given here.
Violet crystal (polyhedron), dimensions 0.08 × 0.07 × 0.05
mm3, crystal system monoclinic, space group P21/n, Z = 4, a =
9.9759(7) Å, b = 16.3832(12) Å, c = 23.1814(17) Å, α = 90°, b
= 95.333(2)°, γ = 90°, V = 3772.3(5) Å3, ρ = 1.462 g/cm3, T =
200(2) K, θmax = 23.81°, radiation Mo Kα, λ = 0.71073 Å, 0.5°
ω-scans with CCD area detector, covering the asymmetric unit
in reciprocal space with a mean redundancy of 5.41 and a
completeness of 100.0% to a resolution of 0.88 Å, 33690
reflections measured, 5796 unique (R(int) = 0.0940), 4147
observed (I > 2σ(I)), μ = 0.95 mm−1, Tmin = 0.93, Tmax = 0.95,
417 parameters refined, hydrogen atoms were treated using
appropriate riding models, goodness of fit 1.08 for observed
reflections, final residual values R1(F) = 0.061, wR(F2) = 0.132
for observed reflections, residual electron density −1.11 to 0.83
e·Å−3.
Synthesis of N-Isopropyl-N′-di(tert-butyl)phosphino-imi-
dazol-2-ylidene (2b). To a suspension of the salt 1b (6.07 g,
15 mmol) in Et2O (50 mL) a solution of sodium
hexamethyldisilazanide (2.75 g, 15 mmol) in Et2O (50 mL)
was added dropwise at −5 °C over 30 min. The reaction
mixture was stirred at 0 °C for 30 min. The solvent was
removed in vacuo and the residue was extracted with pentane
(100 mL). Pentane was removed in vacuo and stored at −40
°C. The product could not be crystallized due to its very high
solubility, and 2b was isolated (2.86 g, 11.2 mmol, 75%) as a
colorless solid at −40 °C or as a liquid at room temperature
1
Synthesis of 3b. Ru(PPh3)3Cl2 (9.35 g, 9.75 mmol) was
dissolved in THF (50 mL) to which was added 2b (2.68 g, 10.5
mmol) dissolved in THF (25 mL). The solution was stirred for
1 h. Half of the solution was removed in vacuo, filtered on a
glass sinter, and washed with pentane (40 mL) to yield 3b
(5.34 g, 7.75 mmol, 79%) as a purple solid. X-ray quality single
crystals could be isolated from the layering of CH2Cl2 and
and freshly used due to its instability. H NMR (THF-d8):
6.93−7.02 (m, 2H, CimidazoleH), 4.55 (sept, J = 6.7 Hz, 1H,
CH(CH3)2), 1.43 (d, J = 6.7 Hz, 6H, CH(CH3)2), 1.22 (d, J =
12.3 Hz, 18H, P(C(CH3)3)2). 31P{1H}P NMR (THF-d8): 97.2.
13C{1H} NMR (THF-d8): 221.6 (br, NCHN), 128.1 (d, J = 34
Hz, NCHCHN), 127.3 (d, 7.9 Hz, NCHCHN), 53.0 (s,
C(CH3)2), 35.3 (d, J = 24.4, C(CH3)3), 29.6 (d, 16.5 Hz,
C(CH3)3), 24.5 (s, CH(CH3)2).
1
cyclohexane. H NMR (CD2Cl2): 7.65 (br, 6H, P(C6H5)3),
7.22−7.33 (m, 9H, P(C6H5)3), 7.07 (s, 1H, CimidazoleH), 6.80 (s,
1H, CimidazoleH), 4.94 (sept, J = 6.8 Hz, 1H, CH(CH3)2), 1.26−
1.31 (m, 12H, P(C(CH3)3)2 and CH(CH3)2), 1.05 (d, J = 14.7
Hz, 9H, P(C(CH3)3)2), 0.77 (d, J = 6.7 Hz, 3H, CH(CH3)2).
31P{1H} NMR (CD2Cl2): 108.8 (d, J = 37.3 Hz, P(C(CH3)3)2),
77.7 (d, J = 37.3 Hz, P(C6H5)3). 13C{1H} NMR (CD2Cl2):
173.9 (m, NCN), 134.1 (d, J = 9.2 Hz, P(C6H5)3), 129.5 (d, J =
2.5 Hz, P(C6H5)3), 127.5 (d, J = 10.4 Hz, P(C6H5)3), 123.6 (d,
J = 8.0 Hz, CimidazoleH), 119.5 (br, CimidazoleH), 49.8 (s,
CH(CH3)2), 44.6 (d, J = 7.7 Hz, P(C(CH3)3)2), 36.9 (d, J = 3.5
Hz, P(C(CH3)3)2), 28.4 (d, J = 5.3 Hz, P(C(CH3)3)2), 26.2 (m,
P(C(CH3)3)2), 25.1 (s, CH(CH3)2), 20.4 (s, CH(CH3)2). Note
one CH2Cl2 and THF in NMR spectra. Elemental Analysis
C32H42Cl2N2P2Ru·CH2Cl2·THF calculated: C, 52.55%; H,
6.20%; N, 3.31% Elemental analysis C32H42Cl2N2P2Ru·
CH2Cl2·THF found: C 52.63%, H 6.39%, N 3.39%.
Synthesis of N-tert-Butyl-N′-di(tert-butyl)phosphino-imi-
dazol-2-ylidene (2c). To a suspension of the salt 1c (4.18 g, 10
mmol) in Et2O (50 mL) a solution of sodium hexamethyldi-
silazanide (1.83 g, 10 mmol) in Et2O (50 mL) was added
dropwise at −5 °C over 30 min. The reaction mixture was
stirred at 0 °C for 30 min. The solvent was removed in vacuo,
and the residue extracted with pentane (75 mL). Pentane was
removed in vacuo to give a viscous oily compound at room
temperature which solidified at −40 °C. The product could not
be crystallized due to its very high solubility, and 2c was
isolated as a crude colorless solid (2.24 g) at −40 °C. 1H NMR
(THF-d8): 6.93−7.02 (m, 2H, CimidazoleH), 4.55 (sept, J = 6.7
Hz, 1H, CH(CH3)2), 1.43 (d, J = 6.7 Hz, 6H, CH(CH3)2),
1.22 (d, J = 12.3 Hz, 18H, P(C(CH3)3)2). 31P{1H}P NMR
(THF-d8): 97.2.
Synthesis of 3a. Ru(PPh3)3Cl2 (1.44 g, 1.5 mmol) was
dissolved in THF (5 mL) to which was added 2a (357 mg, 1.58
mmol) dissolved in THF (5 mL). The solution was stirred for 1
h. The solution was filtered on a glass sinter and washed with
pentane (40 mL) to yield 3a (848 mg, 1.28 mmol, 86%) as a
purple solid. X-ray quality single crystals could be isolated from
Violet crystal (plate), dimensions 0.120 × 0.080 × 0.070
mm3, crystal system triclinic, space group P1, Z = 4, a =
̅
11.6983(5) Å, b = 17.3524(8) Å, c = 17.5483(8) Å, α =
85.5090(12)°, β = 82.1460(12)°, γ = 78.1590(12)°, V =
3449.0(3) Å3, ρ = 1.394 g/cm3, T = 200(2) K, θmax = 25.055°,
radiation Mo Kα, λ = 0.71073 Å, 0.5° ω-scans with CCD area
detector, covering the asymmetric unit in reciprocal space with
a mean redundancy of 2.71 and a completeness of 99.8% to a
resolution of 0.84 Å, 33075 reflections measured, 12 204 unique
(R(int) = 0.0651), 7929 observed (I > 2σ(I)), μ = 0.73 mm−1,
Tmin = 0.86, Tmax = 0.96, 832 parameters refined, hydrogen
atoms were treated using appropriate riding models, goodness
of fit 0.98 for observed reflections, final residual values R1(F) =
0.045, wR(F2) = 0.079 for observed reflections, residual
electron density −0.47 to 0.50 e·Å−3.
1
the layering of CH2Cl2 and cyclohexane. H NMR (CD2Cl2):
7.73 (br, 6H, P(C6H5)3), 7.23−7.35 (m, 9H, P(C6H5)3), 7.00
(m, 1H, CimidazoleH), 6.56 (m, 1H, CimidazoleH), 3.04 (s, 3H,
NCH3), 1.29 (d, J = 13.3 Hz, 9H, P(C(CH3)3)2), 1.02 (d, J =
14.9 Hz, 9H, P(C(CH3)3)2). 31P{1H} NMR (CD2Cl2): 104.2
(d, J = 35.7 Hz, P(C(CH3)3)2), 77.2 (d, J = 35.6 Hz,
P(C(CH3)3)2). 13C{1H} NMR (CD2Cl2): 174.9 (tpseudo, NCN),
133.9 (br, P(C6H5)3), 129.5 (m, P(C6H5)3), 127.5 (m,
P(C6H5)3), 125.0 (s, CimidazoleH), 122.9 (d, J = 8.2 Hz,
CimidazoleH), 44.6 (d, JC,P = 7.3 Hz, P(C(CH3)3)2), 36.4 (d,
Synthesis of 3c. Ru(PPh3)3Cl2 (9.96 g, 10.39 mmol) was
dissolved in THF (50 mL) to which was added 2c (3.01 g, 11.2
mmol) dissolved in THF (5 mL). The solution was stirred for 1
h. The majority of the solution was removed in vacuo, filtered
on a glass sinter, and washed with pentane (50 mL) to yield 3c
(4.05 g, 5.8 mmol, 56%) as a purple solid. X-ray quality single
crystals were isolated from the layering of CH2Cl2 and
JC,P = 4.5 Hz, P(C(CH3)3)2), 34.6 (s, NCH3), 28.2 (d, J = 5.3
Hz, P(C(CH3)3)2), 25.5 (m, P(C(CH3)3)2). Note one CH2Cl2
in X-ray structure. Elemental analysis C30H38Cl2N2P2Ru·
CH2Cl2 calculated: C 49.95%, H 5.41%, N 3.76%. Elemental
analysis C30H38Cl2N2P2Ru·CH2Cl2 found: C 51.18%, H 5.56%,
N 3.67%. EA analysis was consistently problematic with 3a.
K
DOI: 10.1021/acs.inorgchem.5b00513
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
cyclohexane. 1H NMR (CD2Cl2): 7.50−8.10 (br, 4H, P-
(C6H5)3), 7.03−7.49 (m, 12H, P(C6H5)3 and CimidazoleH),
6.45−6.85 (br, 1H, P(C6H5)3), 1.33−1.45 (m, 18H, N−
C(CH3)3 and P(C(CH3)3)2), 1.05 (d, J = 14.9 hz, 9H,
P(C(CH3)3)2). 31P{1H} NMR (CD2Cl2): 110.8 (d, J = 41.5 Hz,
P(C(CH3)3)2), 72.6 (d, J = 41.4 Hz, P(C6H5)3). 13C{1H} NMR
(CD2Cl2): 171.4 (m, NCN), 135.2 (m, P(C6H5)3), 130.1 (m,
P(C6H5)3), 128.0 (m, P(C6H5)3), 122.7 (d, J = 7.0 Hz,
appropriate riding models, goodness of fit 1.03 for observed
reflections, final residual values R1(F) = 0.038, wR(F2) = 0.073
for observed reflections, residual electron density −0.99 to 1.22
e·Å−3.
Synthesis of 4aBr. 3a (2.15 g, 3.25 mmol) was dissolved in 3-
bromopyridine (2−3 mL) without stirring. Within 30 min
orange crystals formed. The 3-bromopyridine was decanted and
Et2O (2 × 15 mL) was added, stirred, and filtered to isolate
4aBr (1.89 g, 2.64 mmol, 82%) as orange crystals.The crystals
initially formed were of sufficient quality for single crystal X-ray
diffraction. 1H NMR (C5D5N): 11.2, 8.83 (m, free 3-
bromopyridine), 8.74 (residual pyridine), 8.60 (m, free 3-
bromopyridine), 8.09, 7.91, 7.85−7.81 (m, free 3-bromopyr-
idine), 7.73, 7.59 (residual pyridine), 7.22 (residual pyridine),
7.18−7.12 (m, free 3-bromopyridine), 6.98, 6.60, 4.67, 4.25,
3.78, 3.61, 1.43 (d, 12.2 Hz, P(C(CH3)3)), 1.37 (d, 11.9 Hz,
P(C(CH3)3)), 1.06 (d, 13.5 Hz, P(C(CH3)3)), 1.05 (d, 13.7
Hz, P(C(CH3)3)). 31P{1H} NMR (C5D5N): 115.8, 109.3,
89.0, 64.6. 13C{1H} NMR (C5D5N): 184.1, 180.6, 179.4, 152.5,
149.7, 140.1, 130.0, 129.4, 129.2, 128.5, 127.1, 126.7, 122.5,
119.1, 42.8, 42.5, 42.4, 41.6, 39.3, 38.3, 37.2, 30.85, 30.81,
30.77, 30.51, 30.46. Elemental analysis C22H31Br2Cl2N4PRu
calculated: C 36.99%, H 4.37%, N 7.84% Found: C 37.04%, H
4.32%, N 8.04%.
CimidazoleH), 121.9 (d, J = 1.5 Hz, CimidazoleH), 59.7 (s, N-
C(CH3)3), 43.8 (d, J = 10.5 Hz, P-C(CH3)3), 37.3 (s, N−
C(CH3)3), 30.1 (s, N−C(CH3)3), 29.1 8 (d, J = 5.0 Hz, P−
C(CH3)3), 27.0 (d, J = 3.6 Hz, P−C(CH3)3). Elemental
analysis C33H44Cl2N2P2Ru calculated: C 56.41%, H 6.31%, N
3.99% Found: C 56.65%, H 6.61%, N 4.17%.
Violet crystal (plate), dimensions 0.090 × 0.080 × 0.060
mm3, crystal system monoclinic, space group P21, Z = 4, a =
10.5570(10) Å, b = 17.4081(17) Å, c = 19.9446(18) Å, α = 90°,
β = 91.141(2)°, γ = 90°, V = 3664.6(6) Å3, ρ = 1.427 g/cm3, T
= 200(2) K, θmax = 24.807°, radiation Mo Kα, λ = 0.71073 Å,
0.5° ω-scans with CCD area detector, covering the asymmetric
unit in reciprocal space with a mean redundancy of 3.33 and a
completeness of 99.5% to a resolution of 0.85 Å, 21833
reflections measured, 12376 unique (R(int) = 0.0712), 8056
observed (I > 2σ(I)), μ = 0.83 mm−1, Tmin = 0.75, Tmax = 0.96,
775 parameters refined, hydrogen atoms were treated using
appropriate riding models, Flack absolute structure parameter
0.13(3), goodness of fit 1.03 for observed reflections, final
residual values R1(F) = 0.076, wR(F2) = 0.106 for observed
reflections, residual electron density −1.13 to 1.14 e·Å−3.
Synthesis of 4aH. 3a (280 mg, 0.42 mmol) was dissolved in
pyridine (3 mL) without stirring. Within 30 min orange crystals
formed. The pyridine was decanted and hexanes (15 mL) was
added, stirred, and filtered to isolate 4aH (203 mg, 0.365 mmol,
87%) as orange crystals. X-ray quality single crystals were
Orange crystal (polyhedron), dimensions 0.17 × 0.12 × 0.11
mm3, crystal system monoclinic, space group Cc, Z = 4, a =
18.609(3) Å, b = 19.892(3) Å, c = 8.7676(15) Å, α = 90°, β =
113.047(3)°, γ = 90°, V = 2986.5(9) Å3, ρ = 1.589 g/cm3, T =
200(2) K, θmax = 25.95°, radiation Mo Kα, λ = 0.71073 Å, 0.5°
ω-scans with CCD area detector, covering the asymmetric unit
in reciprocal space with a mean redundancy of 2.90 and a
completeness of 97.7% to a resolution of 0.82 Å, 8498
reflections measured, 2864 unique (R(int) = 0.0640), 2499
observed (I > 2σ(I)), μ = 3.45 mm−1, Tmin = 0.59, Tmax = 0.70,
310 parameters refined, hydrogen atoms were treated using
appropriate riding models, Flack absolute structure parameter
0.46(3), goodness of fit 1.06 for observed reflections, final
residual values R1(F) = 0.045, wR(F2) = 0.086 for observed
reflections, residual electron density −0.81 to 0.72 e·Å−3.
Synthesis of 4bBr. 3b (1.26 g, 1.83 mmol) was dissolved in
3-bromopyridine (3 mL) without stirring. Within 30 min
orange crystals formed. The 3-bromopyridine was decanted and
Et2O (3 × 20 mL) was added, stirred, and filtered to isolate
1
isolated from the layering of CH2Cl2 and cyclohexane. H
NMR (C5D5N): 8.73 (m, C5D5Nresidual and C5H5N), 8.23 (m,
CHimidazole), 7.99 (s, CHimidazole), 7.98 (s, CHimidazole), 7.90 (s,
CHimidazole), 7.59 (m, C5D5Nresidual and C5H5N), 7.22 (m,
C5D5Nresidual and C5H5N), 4.68 (s, CH3), 4.26 (s, CH3), 1.03
(m, P(C(CH3)3)2). 31P{1H} NMR (C5D5N): 113.9, 107.5.
13C{1H} NMR (C5D5N): 179.0 (d, J = 13.9 Hz, NCN), 177.7
(d, J = 13.3 Hz, NCN), 150.1, 150.0, 135.8, 135.6, 128.1 (s,
CHimidazole), 128.0 (s, CHimidazole), 125.7 (d, J = 7.2 Hz,
CHimidazole), 125.2 (d, J = 7.2 Hz, CHimidazole), 123.9, 123.6, 41.3
(d, J = 2.2 Hz, P(C(CH3)3)2), 40.0 (d, J = 2.1 Hz,
P(C(CH3)3)2), 36.9 (s, CH3), 35.7 (s, CH3), 29.3 (d, J = 5.2
Hz, P(C(CH3)3)2), 29.0 (d, J = 5.2 Hz, P(C(CH3)3)2).
Elemental analysis C22H33Cl2N4PRu·2CH2Cl2 calculated: C
39.69%, H 5.31%, N 7.71% Found: C 40.36%, H 5.21%, N
7.99%. As in the elemental analysis of 3a repeated attempts
were made to obtain better EA data but results were
inconsistent, presumably due to residual solvent.
Orange crystal (polyhedron), dimensions 0.26 × 0.16 × 0.16
mm3, crystal system monoclinic, space group P21/c, Z = 4, a =
8.6967(7) Å, b = 15.7760(12) Å, c = 22.6971(17) Å, α = 90°, β
= 93.707(1)°, γ = 90°, V = 3107.5(4) Å3, ρ = 1.552 g/cm3, T =
199(2) K, θmax = 28.71°, radiation Mo Kα, λ = 0.71073 Å, 0.5°
ω-scans with CCD area detector, covering the asymmetric unit
in reciprocal space with a mean redundancy of 4.68 and a
completeness of 99.1% to a resolution of 0.74 Å, 38123
reflections measured, 7973 unique (R(int) = 0.0573), 6072
observed (I > 2σ(I)), μ = 1.09 mm−1, Tmin = 0.76, Tmax = 0.84,
332 parameters refined, hydrogen atoms were treated using
1
4bBr (1.28 g, 1.72 mmol, 94%) as orange crystals. H NMR
(CD2Cl2): 10.45, 10.37, 10.30, 9.24, 9.18, 8.97, 8.91, 8.67, 8.51,
8.20, 8.10, 7.83, 7.68, 7.27, 7.21, 7.09, 6.92, 6.87, 5.96 (m,
NCH(CH3)2), 4.23 (m, NCH(CH3)2), 1.56−1.33 (m, NCH-
(CH3)2) and P(C(CH3)3)2), 1.03 (m, NCH(CH3)2) and
P(C(CH3)3)2). 31P{1H} NMR (CD2Cl2): 122.1, 120.0, 119.9.
13C{1H} NMR (CD2Cl2): 184.3 (m, NCN), 158.4, 157.9,
157.4, 156.4, 154.5, 153.7, 151.0, 148.0, 138.5, 138.2, 137.9,
124.9, 124.7, 124.6, 124.1, 120.8, 119.2, 118.8, 118.1, 50.6, 49.4,
41.0, 40.3, 30.0 (d, 5.7 Hz), 29.7, 23.6, 23.5. Elemental analysis
C24H35Br2Cl2N4PRu calculated: C 38.83%, H 4.75%, N 7.55%
Found: C 38.79%, H 4.71%, N 7.64%.
Synthesis of 4cBr. 3c (1.44 g, 2.05 mmol) was dissolved in 3-
bromopyridine (3 mL) without stirring. Within 30 min orange
crystals formed. The 3-bromopyridine was decanted and Et2O
(10 mL) was added, stirred, and filtered to isolate 4cBr (1.43 g,
1.89 mmol, 92%) as orange crystals. The crystals initially
formed were of sufficient quality for single crystal X-ray
1
diffraction. H NMR (NC5D5): 8.84 (d, J = 2.5 Hz, free 3-
bromopyridine), 8.74 (s, residual pyridine), 8.60 (dd, J = 4.7
L
DOI: 10.1021/acs.inorgchem.5b00513
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
and 1.3 Hz, free 3-bromopyridine), 7.84−7.77 (m, free 3-
bromopyridine), 7.59 (s, residual pyridine), 7.55 (d, J = 2.1 Hz,
CHimidazole), 7.53 (m, CHimidazole), 7.49 (d, J = 2.2 Hz,
CHimidazole), 7.46 (d, J = 2.3 Hz, CHimidazole), 7.24 (d, J = 2.2
Hz, CHimidazole), 7.22 (s, residual pyridine), 7.14 (3-bromopyr-
idine), 1.94, 1.82, 1.77, 1.58, 1.55, 1.52, 1.50, 1.48, 1.46, 1.40,
1.36, 1.30, 1.26, 1.19, 1.15, 1.12, 1.09 31P{1H} NMR (CD2Cl2):
132.45, 132.37, 132.1, 122.82. 13C{1H} NMR (CD2Cl2): 181.5,
175.6, 152.5, 149.7, 140.1, 126.7, 124.2, 123.9, 122.5, 122.4,
60.1, 59.8, 59.77, 58.4, 42.9, 42.8, 42.6, 42.5, 42.4, 42.2, 42.1,
42.0, 41.95, 32.4, 32.2, 32.0 (d, J = 5.3 Hz), 31.8 (d, J = 5.4 Hz),
30.4 (d, J = 4.9 Hz), 30.3 (d, J = 4.4 Hz), 30.1 (d, J = 4.9 Hz),
30.0 (d, J = 4.6 Hz), 28.9. Elemental Analysis
C25H37Br2Cl2N4PRu calculated: C 39.70%, H 4.93%, N 7.41%
Found: C 40.15%, H 4.59%, N 7.65%.
328 parameters refined, hydrogen atoms were treated using
appropriate riding models, goodness of fit 1.04 for observed
reflections, final residual values R1(F) = 0.056, wR(F2) = 0.109
for observed reflections, residual electron density −0.69 to 0.61
e·Å−3.
Synthesis of 5b. 4bBr (350 mg, 0.47 mmol) was dissolved in
CH2Cl2 (20 mL) and cooled to −40 °C. Trimethylsilyldiazo-
methane (0.25 mL, 2 M hexanes solution)was added, and the
solution was allowed to warm to room temperature. Once
noticeable bubbling had ceased the solution was stirred for an
additional hour. The solvent was removed in vacuo. The solid
was slurried in a minimum amount of THF and filtered. 5b was
washed off the filter with CH2Cl2 (10 mL), and the solvent was
removed in vacuo. The sample was further purified via
crystallization from CH2Cl2 and cyclohexane to yield 5b (145
mg, 0.28 mmol, 60%) as green crystals. X-ray quality single
crystals were isolated from the layering of CH2Cl2 and
Orange crystal (needle), dimensions 0.940 × 0.130 × 0.060
mm3, crystal system triclinic, space group P1, Z = 2, a =
̅
1
8.987(2) Å, b = 13.992(3) Å, c = 14.689(3) Å, α = 82.541(4)°,
β = 86.686(4)°, γ = 75.254(4)°, V = 1770.6(7) Å3, ρ = 1.715 g/
cm3, T = 199(2) K, θmax = 27.743°, radiation Mo Kα, λ =
0.71073 Å, 0.5° ω-scans with CCD area detector, covering the
asymmetric unit in reciprocal space with a mean redundancy of
2.83 and a completeness of 97.6% to a resolution of 0.73 Å,
23079 reflections measured, 8142 unique (R(int) = 0.0523),
cyclohexane. H NMR (CD2Cl2): 18.56 (d, J = 6.3 Hz, 1H,
RuCH), 7.28 (m, 1H, CimidazoleH), 7.03 (d, J = 2.1 Hz, 1H,
C
imidazoleH), 4.61 (sept, J = 6.8 Hz, 1H, CH(CH3)2), 1.55 (d, J
= 15.3 Hz, 9H, P(C(CH3)3)2), 1.23 (d, J = 16.1 Hz, 9H,
P(C(CH3)3)2), 0.37 (s, 9H, Si(CH3)3) 31P{1H} NMR
(CD2Cl2): 141.8. 13C{1H} NMR (CD2Cl2): 215.9 (br), 165.5
(d, J = 20.6 Hz, NCN), 122.1 (d, J = 7.4 Hz, NCCN), 118.5 (d,
J = 2.5 Hz, NCCN), 49.0 (s,), 39.4 (d, J = 9.9 Hz), 37.1 (d, J =
4.6 Hz), 26.9 (dd, J = 15.4/4.3 Hz), 21.5 (d, J = 22.4 Hz), −3.3
(s). Elemental analysis C18H37Cl2N2PRuSi calculated: C
42.18%, H 7.28%, N 5.47% Found: C 42.43%, H 7.26%, N
5.42%.
5730 observed (I > 2σ(I)), μ = 4.05 mm−1, Tmin = 0.52, Tmax
=
0.77, 388 parameters refined, hydrogen atoms were treated
using appropriate riding models, goodness of fit 1.06 for
observed reflections, final residual values R1(F) = 0.051, wR(F2)
= 0.099 for observed reflections, residual electron density −0.97
to 1.02 e·Å−3.
Green crystal (polyhedron), dimensions 0.130 × 0.090 ×
Synthesis of 5a. 4aBr (950 mg, 1.33 mmol) was dissolved in
CH2Cl2 (20 mL) and cooled to −40 °C. Trimethylsilyldiazo-
methane (0.70 mL, 2 M hexanes solution) was added, and the
solution was allowed to warm to room temperature. Once
noticeable bubbling had ceased the solution was stirred for an
additional hour. The solvent was removed in vacuo. The solid
was slurried in a minimum amount of THF and filtered. 5a was
washed off of the filter with CH2Cl2 (10 mL) and the solvent
removed in vacuo. The sample was further purified via
crystallization from CH2Cl2 and cyclohexane to yield 5a (310
mg, 0.64 mmol, 47%) as green crystals. . X-ray quality single
crystals were isolated from the layering of CH2Cl2 and
0.060 mm3, crystal system triclinic, space group P1, Z = 2, a =
̅
9.0506(3) Å, b = 9.7152(3) Å, c = 15.2604(4) Å, α =
94.5103(15)°, β = 94.3406(15)°, γ = 113.3483(13)°, V =
1219.73(6) Å3, ρ = 1.395 g/cm3, T = 200(2) K, θmax = 32.245°,
radiation Mo Kα, λ = 0.71073 Å, 0.5° ω-scans with CCD area
detector, covering the asymmetric unit in reciprocal space with
a mean redundancy of 2.86 and a completeness of 100.0% to a
resolution of 0.67 Å, 24766 reflections measured, 8543 unique
(R(int) = 0.0299), 7381 observed (I > 2σ(I)), μ = 0.98 mm−1,
Tmin = 0.87, Tmax = 0.96, 237 parameters refined, hydrogen
atoms were treated using appropriate riding models, goodness
of fit 1.04 for observed reflections, final residual values R1(F) =
0.031, wR(F2) = 0.065 for observed reflections, residual
electron density −0.34 to 0.67 e·Å−3.
1
1
cyclohexane. H NMR (CD2Cl2): 18.61 (d, J = 6.2 Hz, H,
RuCH), 7.23 (m, 1H, CimidazoleH), 6.91 (d, 1H, CimidazoleH), 3.52
(s, 3H, NCH3), 1.55 (d, J = 15.1, 9H, P(C(CH3)3)2), 1.22 (d, J
= 15.9 Hz, 9H, P(C(CH3)2), 0.37 (s, Si(CH3)3). 31P{1H} NMR
(CD2Cl2): 142.2 13C{1H} NMR (CD2Cl2): 168.5 (d, J = 20.6
Hz, NCN), 125.4 (d, J = 2.4 Hz, NCCN), 123.3 (d, J = 7.8 Hz,
NCCN), 40.8 (d, JC,P = 9.5 Hz, P(C(CH3)3)2), 38.6 (d, JC,P =
5.1 Hz, P(C(CH3)3)2), 35.7 (s, NCH3), 28.5 (d, J = 4.0 Hz,
P(C(CH3)3)2), 28.2 (d, J = 5.0 Hz, P(C(CH3)3)2). Elemental
analysis C16H33Cl2N2PRuSi calculated: C 39.67%, H 6.87%, N
5.78% Found: C 40.24%, H 6.88%, N 5.69%.
Synthesis of 5c. 4cBr (420 mg, 0.56 mmol) was dissolved in
CH2Cl2 (15 mL) and cooled to −40 °C. Trimethylsilyldiazo-
methane (0.21 mL, 2 M hexanes solution) was added, and the
solution was allowed to warm to room temperature. Once
noticeable bubbling had ceased the solution was stirred for an
additional hour. The solvent was removed in vacuo. The solid
was slurried in a minimum amount of THF and filtered. 5c was
washed off the filter with CH2Cl2 (10 mL), and the solvent was
removed in vacuo. The sample was further purified via
crystallization from CH2Cl2 and cyclohexane to yield 5c (266
Green crystal (polyhedron), dimensions 0.110 × 0.100 ×
0.100 mm3, crystal system monoclinic, space group P21/n, Z =
4, a = 9.8131(13) Å, b = 15.449(2) Å, c = 19.426(3) Å, α = 90°,
β = 93.744(3)°, γ = 90°, V = 2938.8(7) Å3, ρ = 1.479 g/cm3, T
= 200(2) K, θmax = 25.680°, radiation Mo Kα, λ = 0.71073 Å,
0.5° ω-scans with CCD area detector, covering the asymmetric
unit in reciprocal space with a mean redundancy of 3.25 and a
completeness of 100.0% to a resolution of 0.82 Å, 18438
reflections measured, 5574 unique (R(int) = 0.0889), 3664
observed (I > 2σ(I)), μ = 1.18 mm−1, Tmin = 0.76, Tmax = 0.91,
1
mg, 0.51 mmol, 91%) as green crystals. . H NMR (CD2Cl2):
18.98 (d, J = 5.7 Hz, 1H, RuCH), 7.23 (m, 1H, CimidazoleH),
7.15 (d, J = 2.2 Hz, 1H, CimidazoleH), 1.60 (d, J = 15.3 Hz, 9H,
P(C(CH3)3)2), 1.43 (s, 9H, NC(CH3)3), 1.23 (d, J = 16.1 Hz,
9H, P(C(CH3)3)2), 0.38 (s, 9H, Si(CH3)3). 31P{1H} NMR
(CD2Cl2): 136.5. 13C{1H} NMR (CD2Cl2): 164.9 (m, NCN),
121.1 (m, CimidazoleH), 59.0 (s, N-C(CH3)3), 40.0 (d, J = 11.8
Hz, P-C(CH3)3), 38.2 (d, J = 7.8 Hz, P-C(CH3)3), 29.7 (s, N−
C(CH3)3), 28.1 (d, J = 4.1 Hz, P−C(CH3)3), 27.7 (d, J = 4.7
M
DOI: 10.1021/acs.inorgchem.5b00513
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Kohl, F. J.; Herrmann, W. A. Tetrahedron Lett. 1999, 40 (26), 4787−
4790.
Hz, P−C(CH3)3), −2.1 (s, Si(CH3)3). The alkylidene C was
not observed. Elemental analysis C19H39Cl2N2PRuSi calculated:
C 43.34%, H 7.47%, N 5.32% Found: C 43.34%, H 7.49%, N
5.56%.
(3) (a) Love, J. A.; Morgan, J. P.; Trnka, T. M.; Grubbs, R. H. Angew.
Chem., Int. Ed. 2002, 41 (21), 4035−4037. (b) Trnka, T. M.; Dias, E.
L.; Day, M. W.; Grubbs, R. H. ARKIVOC 2003, 2002, 28−41.
(4) (a) Conrad, J. C.; Amoroso, D.; Czechura, P.; Yap, G. P. A.; Fogg,
D. E. Organometallics 2003, 22 (18), 3634−3636. (b) Dias, E. L.;
Nguyen, S. T.; Grubbs, R. H. J. Am. Chem. Soc. 1997, 119 (17), 3887−
3897.
(5) (a) Katayama, H.; Nagao, M.; Ozawa, F. Organometallics 2003, 22
(3), 586−593. (b) Kingsbury, J. S.; Harrity, J. P. A.; Bonitatebus, P. J.;
Hoveyda, A. H. J. Am. Chem. Soc. 1999, 121 (4), 791−799. (c) Schwab,
P.; Grubbs, R. H.; Ziller, J. W. J. Am. Chem. Soc. 1996, 118 (1), 100−
110.
Synthesis of 6c. 4cBr (2.05 g, 2.71 mmol) was dissolved in
CH2Cl2 (50 mL) and cooled to −40 °C. Phenyldiazomethane
(0.4 g, 3.39 mmol, 1.25 equiv) was added, and the solution was
allowed to warm to room temperature. Once noticeable
bubbling had ceased the solution was stirred for an additional
hour. The solvent was removed in vacuo. The solid was slurried
in a minimum amount of THF and filtered. The sample was
further purified via crystallization from CH2Cl2 and cyclo-
hexane to yield 5c (1.12 g, 2.11 mmol, 78%) as green crystals. .
1H NMR (CD2Cl2): 15.77 (d, J = 10.2 Hz, 1H, RuCH), 8.39
(d, J = 10.2 Hz, 1H, RuCH), (d, J = 7.8 Hz, 2H, o-Ph), 7.70 (t, J
= 7.5 Hz, 1H, p-Ph), 7.45 (t, J = 7.8 Hz, 2H, m-Ph), (m, J = 2.1
Hz, 1H, CHImidazole), 7.19 (d, J = 7.5 Hz, 1H, CHImidazole), 1.62
(d, J = 14.7 Hz, 9H, P−C(CH3)3), 1.42(s, 9H, N−C(CH3)3),
1.30(d, J = 15.8 Hz, 9H, P−C(CH3)3). 31P{1H} NMR
(CD2Cl2): 129.8. 13C{1H} NMR (CD2Cl2): 299.1 (d, J =
14.0 Hz, CHPh), 167.3 (d, J = 16.5 Hz, NCHN), 150.9 (d, J =
14.0 Hz, CH-ipso-C6H5), 131.6 (s, Ph), 130.5 (s, Ph), 129.1 (s,
Ph), 121.9 (d, J = 7.0 Hz, CHImidazole), 121.5 (d, J = 2.4 Hz,
CHImidazole), 59.4 (s, N-C(CH3)3), 41.1 (d, J = 10.8 Hz, P-
C(CH3)3), 39.5 (d, J = 5.3 Hz, P-C(CH3)3), 30.0 (s, N−
C(CH3)3), 28.8 (d, J = 4.5 Hz, P−C(CH3)3), 28.2 (d, J = 4.0
Hz, P−C(CH3)3). Elemental analysis C19H39Cl2N2PRuSi
calculated: C 49.81%, H 6.65%, N 5.28% Found: C 49.82%,
H 6.89%, N 5.06%.
(6) Van Veldhuizen, J. J.; Campbell, J. E.; Giudici, R. E.; Hoveyda, A.
H. J. Am. Chem. Soc. 2005, 127 (18), 6877−6882.
(7) Herrmann, W. A.; Kocher, C.; Gooßen, L. J.; Artus, G. R. J. Chem.
̈
- Eur. J. 1996, 2 (12), 1627−1636.
́
(8) (a) Cabeza, J. A.; Damonte, M.; García-Alvarez, P.; Kennedy, A.
́
R.; Perez-Carreno, E. Organometallics 2011, 30 (4), 826−833.
̃
(b) Focken, T.; Raabe, G.; Bolm, C. Tetrahedron: Asymmetry 2004,
15 (11), 1693−1706. (c) Ho, C.-C.; Chatterjee, S.; Wu, T.-L.; Chan,
K.-T.; Chang, Y.-W.; Hsiao, T.-H.; Lee, H. M. Organometallics 2009,
28 (9), 2837−2847. (d) Naziruddin, A. R.; Hepp, A.; Pape, T.; Hahn,
F. E. Organometallics 2011, 30 (21), 5859−5866. (e) Wheaton, C. A.;
Bow, J.-P. J.; Stradiotto, M. Organometallics 2013, 32 (21), 6148−
6161. (f) Yang, C.; Lee, H. M.; Nolan, S. P. Org. Lett. 2001, 3 (10),
1511−1514.
́
(9) (a) Benhamou, L.; Wolf, J.; Cesar, V.; Labande, A. s.; Poli, R.;
Lugan, N. l.; Lavigne, G. Organometallics 2009, 28 (24), 6981−6993.
(b) Hahn, F. E.; Naziruddin, A. R.; Hepp, A.; Pape, T. Organometallics
2010, 29 (21), 5283−5288. (c) Cabeza, J. A.; Damonte, M.; García-
́
Alvarez, P.; Hernan
2012, 31 (1), 327−334.
(10) Salem, H.; Schmitt, M.; Herrlich, U.; Kuhnel, E.; Brill, M.;
́
dez-Cruz, M. G.; Kennedy, A. R. Organometallics
ASSOCIATED CONTENT
* Supporting Information
■
S
̈
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.5b00513.
Nagele, P.; Bogado, A. L.; Rominger, F.; Hofmann, P. Organometallics
̈
2013, 32 (1), 29−46.
(11) (a) Hansen, S. M.; Rominger, F.; Metz, M.; Hofmann, P. Chem.
- Eur. J. 1999, 5 (2), 557−566. (b) Hansen, S. M.; Volland, M. A. O.;
Experimental NMR data for compounds 1 and 3−5.
Tables of crystallographic data. Further details of GC
analysis (PDF)
(CIF)
Rominger, F.; Eisentrager, F.; Hofmann, P. Angew. Chem., Int. Ed.
̈
1999, 38 (9), 1273−1276. (c) Volland, M. A. O.; Straub, B. F.;
Gruber, I.; Rominger, F.; Hofmann, P. J. Organomet. Chem. 2001,
617−618 (0), 288−291. (d) Volland, M. A. O.; Rominger, F.;
Eisentrager, F.; Hofmann, P. J. Organomet. Chem. 2002, 641 (1−2),
̈
AUTHOR INFORMATION
Corresponding Author
*E-mail: chrisc.brown@alum.utoronto.ca.
220−226. (e) Volland, M. A. O.; Hansen, S. M.; Rominger, F.;
Hofmann, P. Organometallics 2004, 23 (4), 800−816.
■
(12) (a) Marchenko, A. P.; Koidan, H. N.; Huryeva, A. N.;
Zarudnitskii, E. V.; Yurchenko, A. A.; Kostyuk, A. N. J. Org. Chem.
2010, 75 (21), 7141−7145. (b) Marchenko, A. P.; Koidan, H. N.;
Pervak, I. I.; Huryeva, A. N.; Zarudnitskii, E. V.; Tolmachev, A. A.;
Notes
The authors declare no competing financial interest.
Kostyuk, A. N. Tetrahedron Lett. 2012, 53 (5), 494−496. (c) Nagele,
̈
ACKNOWLEDGMENTS
■
P.; Herrlich, U.; Rominger, F.; Hofmann, P. Organometallics 2013, 32
(1), 181−191.
C.C.B., M.L. and P.H. worked at CaRLa (Catalysis Research
Laboratory) of Heidelberg University, being cofinanced by the
University of Heidelberg, the State of Baden-Wuerttemberg,
and BASF SE. We gratefully acknowledge generous support
from these institutions.
(13) Brown, C. C.; Plessow, P. N.; Rominger, F.; Limbach, M.;
Hofmann, P. Organometallics 2014, 33 (23), 6754−6759.
(14) Kuhnel, E.; Shishkov, I. V.; Rominger, F.; Oeser, T.; Hofmann,
̈
P. Organometallics 2012, 31 (22), 8000−8011.
(15) Crabtree, R. H.; Felkin, H.; Morris, G. E. J. Organomet. Chem.
1977, 141 (2), 205−215.
DEDICATION
■
(16) Hong, S. H.; Wenzel, A. G.; Salguero, T. T.; Day, M. W.;
Grubbs, R. H. J. Am. Chem. Soc. 2007, 129 (25), 7961−7968.
(17) Stumpf, A. W.; Saive, E.; Demonceau, A.; Noels, A. F. J. Chem.
Soc., Chem. Commun. 1995, 11, 1127−1128.
⊥We dedicate this paper to the memory of Prof. Dr. Peter
Hofmann.
REFERENCES
(18) Deng, Q.-H.; Chen, J.; Huang, J.-S.; Chui, S. S.-Y.; Zhu, N.; Li,
G.-Y.; Che, C.-M. Chem. - Eur. J. 2009, 15 (41), 10707−10712.
■
(1) Schwab, P.; France, M. B.; Ziller, J. W.; Grubbs, R. H. Angew.
Chem., Int. Ed. Engl. 1995, 34 (18), 2039−2041.
(2) (a) Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett.
(19) Samojłowicz, C.; Bieniek, M.; Pazio, A.; Makal, A.; Wozniak, K.;
́
Poater, A.; Cavallo, L.; Woj
Eur. J. 2011, 17 (46), 12981−12993.
́
cik, J.; Zdanowski, K.; Grela, K. Chem. -
1999, 1 (6), 953−956. (b) Ackermann, L.; Furstner, A.; Weskamp, T.;
̈
N
DOI: 10.1021/acs.inorgchem.5b00513
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(20) (a) Kavitake, S.; Samantaray, M. K.; Dehn, R.; Deuerlein, S.;
Limbach, M.; Schachner, J. A.; Jeanneau, E.; Coperet, C.; Thieuleux, C.
Dalton Transactions 2011, 40 (46), 12443−12446. (b) Cabrera, J.;
Padilla, R.; Bru, M.; Lindner, R.; Kageyama, T.; Wilckens, K.; Balof, S.
L.; Schanz, H.-J.; Dehn, R.; Teles, J. H.; Deuerlein, S.; Muller, K.;
̈
Rominger, F.; Limbach, M. Chem. - Eur. J. 2012, 18 (46), 14717−
14724. (c) Bru, M.; Dehn, R.; Teles, J. H.; Deuerlein, S.; Danz, M.;
Muller, I. B.; Limbach, M. Chem. - Eur. J. 2013, 19 (35), 11661−
̈
11671.
(21) Songis, O.; Slawin, A. M. Z.; Cazin, C. S. J. Chem. Commun.
2012, 48 (9), 1266−1268.
(22) (a) Warwel, S.; Katker, H.; Rauenbusch, C. Angew. Chem., Int.
̈
Ed. Engl. 1987, 26 (7), 702−703. (b) Rost, A. M. J.; Schneider, H.;
Zoller, J. P.; Herrmann, W. A.; Kuhn, F. E. J. Organomet. Chem. 2005,
̈
690 (21−22), 4712−4718. (c) Rountree, S. M.; Lagunas, M. C.;
Hardacre, C.; Davey, P. N. Appl. Catal., A 2011, 408 (1−2), 54−62.
(23) Behr, A.; Neubert, P.Applied Homogeneous Catalysis; Wiley-
VCH: Hoboken, NJ, 2012.
(24) Hallman, P. S.; Stephenson, T. A.; Wilkinson, G. Tetrakis-
(Triphenylphosphine)Dichloro-Ruthenium(II) and Tris-
(Triphenylphosphine)-Dichlororuthenium(II). In Inorganic Syntheses;
John Wiley & Sons, Inc.: New York, 2007; Vol. 12, pp 237−240.
(25) (a) Sheldrick, G. M. SADABS; Bruker Analytical X-ray-Division:
Madison, WI, 2008. (b) Sheldrick, G. M. Acta Crystallogr., Sect. A:
Found. Crystallogr. 2008, 64, 112−122.
O
DOI: 10.1021/acs.inorgchem.5b00513
Inorg. Chem. XXXX, XXX, XXX−XXX