Pyridine-Stabilized Fast-Initiating Ruthenium Monothiolate Catalysts for Z-Selective Olefin Metathesis

Article abstract of DOI:10.1021/acs.organomet.7b00441

Pyridine as a stabilizing donor ligand drastically improves the performance of ruthenium monothiolate catalysts for olefin metathesis in comparison with previous versions based on a stabilizing benzylidene ether ligand. The new pyridine-stabilized ruthenium alkylidenes undergo fast initiation and reach appreciable yields combined with moderate to high Z selectivity in self-metathesis of terminal olefins after only a few minutes at room temperature. Moreover, they can be used with a variety of substrates, including acids, and promote self-metathesis of ω-alkenoic acids. The pyridine-stabilized ruthenium monothiolate catalysts are also efficient at the high substrate dilutions of macrocylic ring-closing metathesis and resist temperatures above 100 °C during catalysis.

Full text of DOI:10.1021/acs.organomet.7b00441

Article
pubs.acs.org/Organometallics
Pyridine-Stabilized Fast-Initiating Ruthenium Monothiolate Catalysts
for Z‑Selective Olefin Metathesis
Giovanni Occhipinti,* Karl W. Tornroos, and Vidar R. Jensen*
̈
́
Department of Chemistry, University of Bergen, Allegaten 41, N-5007 Bergen, Norway
S
* Supporting Information
ABSTRACT: Pyridine as a stabilizing donor ligand drastically
improves the performance of ruthenium monothiolate catalysts
for olefin metathesis in comparison with previous versions based
on a stabilizing benzylidene ether ligand. The new pyridine-
stabilized ruthenium alkylidenes undergo fast initiation and
reach appreciable yields combined with moderate to high Z
selectivity in self-metathesis of terminal olefins after only a few
minutes at room temperature. Moreover, they can be used with
a variety of substrates, including acids, and promote self-
metathesis of ω-alkenoic acids. The pyridine-stabilized ruthe-
nium monothiolate catalysts are also efficient at the high
substrate dilutions of macrocylic ring-closing metathesis and resist temperatures above 100 °C during catalysis.
Chart 1. Examples of Z-Selective Catalysts
INTRODUCTION
■
Transition-metal-catalyzed olefin metathesis is a powerful,
versatile, and green method for making carbon−carbon double
bonds and is widely used in organic synthesis.1,2 This reaction is
also exploited in several industrial processes ranging from value-
added processes of simple alkenes to the synthesis of complex
pharmaceuticals.3,4
Grubbs-type ruthenium-based catalysts have been the most
widely used so far because, in addition to their high activity,
they tolerate many functional groups, including alcohols and
carboxylic acids.5 Moreover, they are relatively robust toward
air and moisture and therefore also easy to handle and store. All
this means that the ruthenium-based catalysts may be used in a
broad range of synthetic applications.6−21 Nevertheless, apart
from particular cases, they do not give Z-configured alkenes in
practical yields. Thus, to fill this need, highly Z selective olefin
metathesis systems based on Mo,22 W,23 and Ru24−26 have
been developed in recent years (see Chart 1 for examples).
These catalysts, with the exception of the stereoretentive Ru
dithiolate complexes,27,28 get their Z selectivity from a “wall” of
steric bulk that essentially blocks one of the two faces of the
metallacyclobutane moiety in the rate-determining transition
state.29−31 Overall, these compounds allow the selective
synthesis of many useful cis-disubstituted alkenes in moderate
to high yields.32−35However, none of the existing Z-selective
systems can compete with the nonstereoselective Grubbs-type
catalysts in terms of robustness, catalytic activity, and functional
group tolerance, all of which still limit the scope of Z selective
metathesis despite the impressive progress in recent years.
Z-Selective ruthenium-based catalysts have been achieved by
following two conceptually different strategies (see Figure 1).
The first strategy exploits the steric properties of the N-
heterocyclic carbene (NHC) ligand in a side-pathway
mechanism,25,29 exemplified by the cyclometalated ruthenium
catalysts (e.g., 3, Chart 1) developed by Grubbs and co-
workers24,36,37 and by the Ru dithiolate systems (e.g., 4, Chart
1) developed by Hoveyda and co-workers.25,38 The Ru
dithiolates are highly stereoretentive but are much more active
Received: June 12, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.7b00441
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
dilutions of macrocyclic ring-closing metathesis (mRCM) is
poor.50,51
In this contribution we overcome the main limitations of the
monothiolate-substituted catalysts by replacing the chelating o-
isopropoxybenzylidene ligand with monodentate alkylidenes
and pyridine. The new catalysts initiate metathesis more quickly
and are also much less prone to mediate olefin isomerization.
The improved metathesis performance of the new catalysts is
particularly evident from the fact that they allow Z-selective
synthesis of α,ω-dicarboxylic acid alkenes via self-metathesis of
ω-alkenoic acids, an application that facilitates access to many
biologically active targets but that has not yet been achieved
with other Z-selective catalysts.52,53
RESULTS AND DISCUSSION
Figure 1. Geometry of the metallacyclobutane intermediate resulting
from bottom (GII and 5) and side (3 and 4) olefin binding pathways.
For the Z-selective catalysts 3−5 the substituents R1 and R2 of the
metallacyclobutane ring prefer to be oriented away from the sterically
demanding moieties, rendered in red. For the Z-selective catalysts, the
favored cis configuration of the metallacyclobutane is shown.
■
Syntheses of the Pyridine Complexes 11a,b and
14a,b. Our starting hypothesis was that the presence of two
sterically demanding ligands (i.e., the thiolate and the NHC) in
the 14-electron active catalyst is the main reason for the low
activity of monothiolate-substituted catalysts, an idea that is
consistent with the fact that the presence of an electron-
withdrawing group at the benzylidene ether ligand only affects
the catalytic activity marginally.49 However, attempts to
improve the activity by using smaller ligands resulted in less
stable and less Z-selective catalysts.48,49 Instead, we noted that
DFT studies suggest that the low catalytic activity of thiolate-
substituted catalyst 5 mainly is due to its slow initiation.30,48 In
particular, the calculations show that the dissociation of the
chelating ether functionality should be fast, while the bottleneck
of the reaction is the first metathesis step, which replaces the
starting benzylidene ether ligand. The high energy of the
transition state of this step is due to the steric repulsion
between the bulky thiolate and the o-isopropoxybenzylidene.30
Spurred by this insight, we inspected the transition state and
found that the o-isopropoxy moiety was the most important
contributor to the repulsion. This suggests that replacing the
chelating 2-isopropoxybenzylidene ligand by a unsubstituted
benzylidene and a suitable monodentate donor might lead to a
more active precatalyst.
and efficient in Z-selective in comparison to E-selective
reactions.27 Grubbs and co-workers have recently improved
the original design to boost the catalytic performance of the E-
selective transformations.28 The second strategy to achieve Z-
selective Ru-based catalysts consists of retaining the bottom-
pathway mechanism of the parent Grubbs-type catalysts39−42 at
the same time as using anionic ligands of very different
size.26,43−48
The only family of Z-selective catalysts following a bottom-
pathway mechanism are ruthenium monothiolate alkylidenes
such as complexes 5−7 (Chart 1), which are moderately to
highly selective and reach 96% Z selectivity in the self-
metathesis of sterically hindered allylic substrates such as
allyltrimethylsilane.26,46−48 These compounds share most of the
structural features and the bottom-pathway mechanism30 with
the Hoveyda−Grubbs-type catalysts (see the upper part of
Figure 1) and have been developed, via computational
prediction and experimental followup, by our group.46−48 The
catalysts are prepared from commercially available or easily
accessible nonselective Hoveyda−Grubbs-type catalysts via
one-step replacement of one of the anionic ligands (e.g.,
chloride or isocyanate) by a sterically demanding thiolate (e.g.,
2,4,6-triphenylbenzenethiolate).
Whereas the monothiolate-substituted catalysts so far have
been both less Z-selective and less active than side-pathway
catalysts, they may inherit highly desirable properties such as
robustness from the parent catalyst.49 For example, compound
6 (Chart 1) can be used in air with nondegassed and
nonpurified substrates and solvents and tolerates the presence
of acidic additives.47 To benefit from this robustness and
functional group tolerance in practical metathesis experiments,
however, the Z-product yields from the monothiolate-
substituted catalysts must be increased. The yields have so far
been hampered by low catalytic activity combined with a high
tendency to isomerize both substrates and products. With the
monothiolate-substituted catalysts, appreciable metathesis
activity is typically only obtained at high substrate concen-
trations and at moderately elevated temperatures, while
isomerization reactions reduce the metathesis efficiency at
elevated temperatures and at high substrate dilution. For
instance, the catalytic performance at the high substrate
We set out to investigate monodentate neutral donors such
as alkylphosphines, alkyl phosphites, and several N-heterocyclic
compounds, and after a preliminary screening of donor-
stabilized parent ruthenium-based catalysts, we identified
pyridine as a suitable ligand.
The pyridine-stabilized version of 5, i.e. complex 11a, was
successfully prepared by a two-step substitution starting from
Grubbs second-generation catalyst 9 (see Scheme 1). First,
tricyclohexylphosphine was replaced by two pyridine molecules
to give 10 using a previously reported literature procedure.54
Next, the target compound 11a was obtained, in low yield
(29%), by reacting 10 with potassium 2,4,6-triphenylbenzene-
thiolate (8a), which was prepared according to literature
procedures starting from the commercially available 2,4,6-
triphenylbenzenethiol.47 The low yield of the second step is a
consequence of the difficult separation of 11a from unidentified
alkylidene-free ruthenium species, some of which are
presumably products of decomposition of 11a (see the
Supporting Information for details).
In order to evaluate the effect of a larger and more donating
thiolate ligand in the framework of the new catalyst design, we
prepared 11b, which is based on a computationally designed
thiolate ligand, 2,4,6-tris(3,5-dimethylphenyl)benzenethiolate
(8b).55 Reaction of 10 with 8b delivered 11b in fair yield
B
DOI: 10.1021/acs.organomet.7b00441
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
a
Scheme 1. Synthesis of Complexes 11a,b
Figure 2. ORTEP-style drawing of one of the two unique complexes
of 11b with the anisotropic displacement ellipsoids drawn at the 30%
probability level. Hydrogen atoms and solvent molecules (pentane)
have been omitted for clarity. Selected geometrical parameters (bond
distances in Å and angles in deg): Ru1−C57 = 1.831, Ru1−Cl1 =
2.404, Ru1−S1 = 2.340, Ru1−N1 = 2.132, Ru−C36 = 2.054; Ru1−
S1−C1 = 107.6, S1−Ru1−Cl1 = 160.93, C36−Ru1−N1 = 167.3.
In order to probe the generality of our approach and the
influence of ligand variation, we decided to repeat the same
two-step protocol on the basis of another commercially
available catalyst, CatMetium RF357 (12, Scheme 2). The
a
Pyridine was the only neutral donor ligand for which Ru
monothiolate alkylidene complexes could be isolated and purified.
An analogue of 11b based on 3-Br-pyridine was successfully isolated,50
but this compound was harder to crystallize than 11b, and we were
unable to achieve a comparable level of purity for this 3-Br-pyridine
analogue.
Scheme 2. Synthesis of Complexes 14a,b
(60%). Crystals of 11b suitable for X-ray diffraction analysis
were grown at −34 °C from a solution of the complex, obtained
by first dissolving 11b in a minimum amount of toluene,
followed by addition of pentane until the solution became
slightly cloudy.
The molecular structure of 11b and the relevant bond
lengths and angles are shown in Figure 2. The complex has a
slightly distorted square pyramidal geometry, with the
benzylidene ligand occupying the apical position, oriented
toward the less sterically demanding chloride ligand to avoid
the bulky thiolate. A similar steric influence from the thiolate is
evident also on the pyridine ligand. The S1−Ru1−N1 bond
angle (92.3°) is wider than the corresponding angle involving
chlorine (Cl1−Ru1−N1 = 84.1°), showing that the pyridine is
bent toward the chloride to avoid the bulky thiolate. In
contrast, even with the thiolate in a cis position, the
ruthenium−pyridine bond distance is relatively short (2.132
Å) in comparison with, for example, those of similar pyridine-
stabilized ruthenium alkylidenes (2.128−2.205 Å).56 On the
other hand, the other rutheniumligand bonds are slightly
elongated, by 0.3 pm (alkylidene), 5 pm (NHC), 1 pm
(chloride), and 3 pm (thiolate), respectively, in comparison
with the corresponding bonds in 5. The largest difference is
observed for the NHC ligand, in agreement with the expected
larger trans influence of the pyridine in comparison with the
ether ligand. Finally, the bond angle at the sulfur atom (Ru1−
S1−C1 = 107.6°) is, remarkably, 5° sharper than that of 5,
despite the smaller thiolate ligand of the latter compound.
alkylidene ligand (2-thienylmethylidene) of complex 12 differs
slightly from that of 8, and 12 also contains an unsaturated
NHC with a dimethyl-substituted backbone (Me2IMes).
Starting from 12, we synthesized the new monothiolate-
substituted catalysts 14a,b as shown in Scheme 2. The bis-
pyridine intermediate 13 was prepared in very good yield
C
DOI: 10.1021/acs.organomet.7b00441
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(84%) by reacting 12 with an excess of pyridine (see the
Supporting Information for details). Next, the reaction of 13
with 8a or 8b delivered the target compound 14a or 14b in low
(37%) and very good (82%) yields, respectively. As for 11a
above, which is also based on the 2,4,6-triphenylbenzenethio-
late ligand, the low yield of 14a resulted from its difficult
isolation and purification. Crystals of 14b were grown from a
concentrated solution in pentane at −34 °C, and the X-ray
diffraction analysis thereof revealed that the structure of 14b
(Figure 3) is similar to that of 11b described above.
Whereas several cationic NHC-bearing, tetracoordinate ruthe-
nium alkylidenes have been reported and characterized by X-
ray diffraction,41,60−62 the only known neutral counterparts are
based on a trialkylphosphine in combination with two sterically
demanding alkoxide ligands.63,64
The fraction of B formed depends on the nature of the
pyridine-stabilized complex A and increases with temperature
and dilution; see Table 1 and Table S1 in the Supporting
Table 1. Dissociation of Pyridine from 11a,b and 14a,b in
C6D6 Solution (5 mM)
amt of B (%)
entry
complex
293 K
313 K
1
2
3
4
11a
11b
14a
14b
3.8
11.8
4.1
8.2
24.7
7.0
18.4
32.2
Information. In particular, more B is produced by the two
complexes having the bulkier and more electron donating
thiolate ligand, and at 313 K almost a fourth and a third of the
11b and the 14b complexes, respectively, have dissociated the
pyridine ligand. In comparison, the influence of the two
carbene ligands (i.e., the NHC and the alkylidene) is less clear.
For example, at 293 K the percentage of B increases in the
order 11a < 14a ≪ 11b ≪ 14b, while at 313 K 14a becomes
the complex least prone to dissociation, giving the order 14a <
11a ≪ 11b < 14b (see Table 1 and the Supporting Information
for more details).
Figure 3. ORTEP-style drawing of one of the two unique complexes
of 14b with the anisotropic displacement ellipsoids drawn at the 50%
probability level. Hydrogen atoms and solvent molecules (pentane)
have been omitted for clarity. Selected geometrical parameters (bond
distances in Å and angles in deg): Ru1−C6 = 1.832, Ru1−C11 =
2.066, Ru1−S1 = 2.335, Ru1−N1 = 2.141, Ru−Cl1 = 2.405; Ru1−
S1−C34 = 108.7, S1−Ru1−Cl1 = 159.63, C11−Ru1−N1 = 168.1.
In order to understand these trends and the factors favoring
pyridine dissociation, van’t Hoff plots have been generated to
determine the thermodynamic parameters (ΔH° and ΔS°; see
Figure 4 and Table 2). The dissociation constant (K) in C6D6
Dissociation Equilibrium of Pyridine Ligand in
1
Solution. The H NMR spectra (C6D6) of 11a,b and 14a,b
reveal that these 16-electron complexes A easily dissociate the
pyridine ligand in solution to give the corresponding 14-
electron species B, as depicted in Scheme 3. Specifically, the 1H
Scheme 3. Dissociation Equilibrium of the Pyridine Ligand
in C6D6 Solution
Figure 4. van’t Hoff plots of ln K vs 1000/T for the dissociation
equilibrium of pyridine in C6D6 solution (5 mM) of 11a,b and 14a,b.
NMR spectra contain two different signals for the alkylidene
proton along with the signals corresponding to the free pyridine
ligand. The pyridine signals are equimolar with the alkylidene
proton signal located at higher field strength (lower δ), which is
due to the 14-electron complex B. This lability is unusual for a
pyridine-stabilized 16-electron complex, because pyridine is flat
and has a high affinity for the ruthenium center58,59 and
because the dissociation produces an electronically as well as
coordinatively unsaturated ruthenium species B. These species,
albeit observed only in solution, are rare examples of stable,
neutral tetracoordinate NHC-based ruthenium alkylidenes.
Table 2. Experimentally Determined Thermodynamic
Parameters for Pyridine Dissociation from 11a,b and 14a,b
entry
complex
ΔH (kcal/mol)
ΔS (cal/mol)
1
2
3
4
11a
11b
14a
14b
14.51(6)
14.9(2)
10.3(2)
11.97(9)
26.1(2)
32.1(5)
12.1(5)
24.0(3)
D
DOI: 10.1021/acs.organomet.7b00441
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
solution (5 mM) has been evaluated at four different
temperatures (283, 293, 303, and 313 K) for all four complexes
using 1H NMR spectroscopy (see the Supporting Information).
In all cases the equilibrium was reached instantaneously,
suggesting a negligible or very low dissociation barrier. Figure 4
shows the plot of ln K against 1000/T, while the calculated
ΔH° and ΔS° values of dissociation, i.e., the slope and the
intercept of the lines, respectively, are reported in Table 2. Both
the enthalpy and the entropy of dissociation are positive for all
four complexes. In terms of enthalpy, dissociation costs more
than 10 kcal/mol in all cases, but the positive entropy changes
(12−32 cal/mol) imply that the Gibbs free energies of
dissociation are modest (4.5−6.5 kcal/mol) at 313 K,
consistent with the substantial fractions of pyridine dissociated
at that temperature (Table 1).
Table 3. Self-Metathesis of Neat Allylbenzene at 22 °C
a
b
a
entry
cat.
16/17
0.2
0.6
15
yield of 16 (%)
Z-16 (%)
1
2
3
4
5
6
5
2
3
83
87
81
86
63
80
6
11a
11b
14a
14b
41
33
69
73
51
63
>100
a
1
Determined by H NMR analysis of the quenched reaction mixture.
Isolated yields.
b
Even if the thermochemical parameters of Table 2 are all
positive, the thiolate-substituted catalysts (14) based on parent
complex 12 (CatMetium RF3) have lower enthalpies and
entropies of dissociation in comparison to those based on 9
(the second-generation Grubbs catalyst). Dissociation is in
particular associated with small thermochemical parameters for
complex 14a (ΔH° = 10.3(2) kcal/mol, ΔS° = 12.1(5) cal/
mol), which might suggest that dissociation of pyridine partially
is compensated for by dimerization of B (i.e., 14a without
pyridine). Reversible formation of dimers of 14-electron
complexes has been demonstrated by Piers and co-workers
for their tetracoordinate ruthenium phosphonium alkylidenes,65
and the tendency to form aggregates grows with increasing
complex concentration.65 Indeed, doubling the concentration
of the complex lowers the enthalpy and entropy of dissociation
(ΔH° = 7.0(2) kcal/mol, ΔS° = 1.6(4) cal/mol, at [14a] = 10
mM; see the Supporting Information for details). However,
even if these data suggest that the dissociated species B of 14a
do form aggregates, the 1H NMR spectra of 14a recorded at the
two different concentrations do not provide evidence for
aggregate formation. The corresponding spectra, recorded at
283 K and at 5 and 10 mM, respectively, may essentially be
superimposed and the signals associated with aggregates cannot
be distinguished from the rest of the signals. These spectra are
complex, and the aggregate-related signals are presumably
masked by other signals (see the Supporting Information for
details).
In the absence of spectral evidence for aggregation, the
contrast between the low dissociation entropy of 14a and those
of 11b (ΔS° = 32.1(5) cal/mol), 11a (ΔS° = 26.1(2) cal/mol),
and 14b (ΔS° = 24.0(3) cal/mol) is perhaps the most telling
piece of information. The much higher dissociation entropies of
11a,b and 14b suggest that their the tendency to form
aggregates of B must be low, which is consistent with the fact
that ΔH° and ΔS° of these complexes decrease much less than
those of 14a when the concentration of the complex is doubled
(see Table S1 in the Supporting Information).
Catalytic Performance in Olefin Metathesis. Complexes
11a,b and 14a,b were initially tested in the self-metathesis of
neat allylbenzene, a highly reactive substrate that is also prone
to double-bond alkene migration in the presence of ruthenium-
based olefin metathesis catalysts,26,66−68 at room temperature
using a catalyst loading of 1 mol % (see Table 3). As a
background for the evaluation of the new catalysts, the original
monothiolate-substituted catalysts 5 and 6 (containing a
chelating o-isopropoxybenzylidene ligand) were also tested
under identical conditions. The reaction was quenched after 5
min by dilution of the mixture with hexane followed by
filtration through a short pad of silica gel to remove the
catalyst.69
The pyridine-stabilized complexes 11a,b and 14a,b gave
isolated yields in the range 33−73%, which is clearly superior to
the single-digit yields of 5 and 6 (Table 3). In addition to being
more active, the pyridine-stabilized precatalysts are also much
more selective for metathesis, with a lower tendency to
isomerize the substrate, as shown by their higher ratios 16/17.
The catalytic performances of precursors such as 11a and 5
differ radically, even though they should give the same active
species on initiation. This strongly suggests that the faster
initiation of the pyridine-stabilized precatalysts is the cause of
their much lower tendency to isomerization. Olefin isomer-
ization during metathesis is believed to be promoted not by the
olefin metathesis catalyst but by different ruthenium-based
catalyst decomposition products or impurities present in the
precatalyst.47 Thus, rapid initiation of the metathesis catalyst
ensures that, for a certain time interval at the beginning of the
reaction, the ratio of metathesis-active to isomerization-active
ruthenium species will be high.
Since the species promoting isomerization of the substrate
also contribute to erosion of the stereoselectivity,47 increasing
the initiation rate of the catalyst may represent a strategy to
improve the efficiency of many stereoselective transformations.
Grubbs and co-workers have recently reported that faster-
initiating catalysts perform far better in E-stereoretentive cross-
metathesis than their slower-initiating counterparts.28 In
contrast, Nelson and Percy, studying a range of catalysts
based on leaving phosphine ligands as well as chelating
alkylidenes, reported that the tendency to isomerization did not
correlate with the initiation rate but was instead heavily affected
by the nature of the dissociating ligand.70 Their findings may
offer an alternative, or additional, explanation for the superiority
of 11a in comparison to 5.
In addition to shedding light on the role of the initiation rate,
the catalytic results in Table 3 show that yields and selectivities
also depend on the nature of the remaining ligands. For
example, the complexes 11a,b derived from the Grubbs second-
generation catalyst are less active and more prone to substrate
isomerization than 14a,b prepared from CatMetium RF3, but
the latter are less Z-selective (cf. entries 3 and 5 and entries 4
and 6 in Table 3). Similarly, the nature of the thiolate ligand
mainly affects the tendency to isomerization and the Z
selectivity, and in particular the complexes based on 2,4,6-
tris(3,5-dimethylphenyl)benzenethiolate are more selective for
metathesis as well as for the Z product in comparison to those
based on the smaller 2,4,6-triphenylbenzenethiolate. In
E
DOI: 10.1021/acs.organomet.7b00441
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
conclusion, the selectivity decreases in the order 11b > 11a and
14b > 14a. Overall, 14b stands out as the most active and
efficient catalyst.
Table 4. Self-Metathesis of ω-Alkenoic Acids in Toluene
Solution (1 M)
Next, the catalytic performance of 14b was tested in the self-
metathesis of six additional substrates, as depicted in Figure 5.
a
a
a
entry
1
cat.
n
time (min) conversn (%)
yield (%)
Z (%)
9
2
5
25
60
5
6
24
46
35
44
50
10
34
59
6
24
45
35
42
42
3
15
12
8
2
14b
2
80
70
54
n.d.
45
67
25
60
5
3
4
5
6
8
8
8
11b
14b
5
28
56
5
a
Determined by 1H NMR analysis of quenched reaction mixture
samples using hexamethylbenzene as internal standard. Samples of the
reaction mixture were taken at regular intervals and immediately
quenched using an excess of ethyl vinyl ether (EVE). Conversion is the
amount of substrate converted, whereas yield refers to the amount of
substrate converted into the self-metathesis product.
The reaction was conducted at room temperature in toluene
(1 M), under static vacuum (10−5 bar) and using 5 mol % of
catalyst.
Figure 5. Self-metathesis of neat 1-alkenes with catalyst 14b.
Entry 2 of Table 4 shows that the self-metathesis of 4-
pentenoic acid using 14b is fast and clean during the first few
minutes, but the activity drops rapidly with the progress of the
reaction, probably as a consequence of catalyst decomposition,
as indicated by the onset of double-bond migration and
diverging values for conversions and yields. For example, after 5
min 35% of the self-metathesis product was obtained with 80%
Z selectivity, while another 20 min only resulted in a modestly
improved yield, some product impurity (ca. 2%) due to double-
bond migration in the substrate, and a drop in Z selectivity.
Beyond 25 min of reaction time, no more self-metathesis
product is formed. More isomerization products are formed,
however, and the erosion of the Z selectivity continues. Such
erosion is, in fact, also seen for Grubbs second-generation
catalyst 9 (entry 1 in Table 4), which was subjected to the same
test as 14b. In contrast to the case for 14b, however, the
reaction using 9 is slow at the beginning but the rate increases
with the progress of the reaction, with the E isomer being the
major product.
Another series of self-metathesis experiments, under
conditions identical with those of the 4-pentenoic acid
conversions above, was carried out using a longer substrate,
10-undecenoic acid. Using catalyst 14b (entry 5), the reaction
was faster than for 4-pentenoic acid, and after 5 min more than
60% of the substrate had been converted, mostly to the self-
metathesis product (57%), with moderate Z-selectivity (67%).
Complex 11b also promotes this reaction, but with both lower
activity and Z selectivity in comparison to 14b. In contrast, 6,
which contains a chelating alkylidene ligand and which tolerates
acidic additives,47 is a poor metathesis catalyst for this substrate
and mainly promotes double-bond migration in the substrate
(entry 3).
The reactions were conducted under conditions identical with
those above for allylbenzene, except for the presence of
hexamethylbenzene as internal standard. After 5 min the
reaction was quenched with an excess of ethyl vinyl ether
(EVE) and then a sample of the reaction mixture was analyzed
1
by H NMR spectroscopy.
Of the six olefinic substrates tested, two, N-allylaniline and 2-
allyloxyethanol, gave low yields (14% and 2%, respectively) and
less than 80% of the Z product. For 2-allyloxyethanol, the low
solubility of 14b in this substrate, as indicated by the
heterogeneous appearance of the reaction mixture, to a large
extent explains the near-absence of product. Indeed, when the
same reaction was performed in toluene (1 M), under
otherwise identical conditions, a much higher yield (24%)
was obtained, albeit with a smaller fraction of the Z product
(70%).
The four remaining substrates (1-octene, allylacetate, 4-
phenyl-1-butene, and allylboronic acid pinacol ester) were all
converted to self-metathesis products in fair to good yields
(46−75%) in combination with moderate to high Z selectivities
(79−94%). Remarkably, these results were obtained in only 5
min and at room temperature and, with the exception of one
substrate (allylboronic acid pinacol ester), the degree of
isomerization, given by the difference between conversion
and yield, is negligible or very low.
Inspired by these results, the self-metathesis of substrates (ω-
alkenoic acids) bearing acidic functional groups that have
proved challenging for other Z-selective catalysts52,71 was also
tested (see Table 4). The cis products of this self-metathesis are
valuable and may, for example, be used as key intermediates in
the synthesis of several pharmaceutical drugs72−77 and natural
products.78,79 The latter compounds, which are traditionally
prepared via multistep procedures often developed for specific
target molecules,79,80 can be reached more directly via Z-
selective self-metathesis of ω-alkenoic acids.
Finally, catalyst 14b was also tested in macrocyclic ring-
closing metathesis (mRCM) of two different dienes, 18 and 19,
whose products are 18a, the Z isomer of which is the Yuzu
lactone,81,82 a 13-membered lactone used in the fragrance
industry, and the 14-membered lactone 19a, respectively
(Scheme 4). These lactones have previously been prepared
F
DOI: 10.1021/acs.organomet.7b00441
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 4. Macrocylic Ring-Closing Metathesis Using 14b
via Z-selective mRCM using both the cyclometalated
ruthenium-based catalyst 336 and the monoaryloxide−pyrrolide
(MAP) tungsten- and molybdenum-based catalysts 1 and 283
(see Scheme 4 and Chart 1 for the catalyst structures). The
results from these tests can be used in the evaluation of 14b.
The mRCM reactions of 14b were carried out under partially
optimized conditions,50 including the use of static vacuum
(10−5 bar), high substrate dilution (in toluene), elevated
temperature, and high catalyst loading (10%).
Macrocyclization of 18 is more challenging than that of 19,
because the product is a 13-membered lactone and is more
strained than the 14-membered 19a.83 The reactions were thus
conducted under slightly different conditions, with a reaction
temperature of 60 °C, a substrate concentration of 3 mM, and
quenching after 2 h being used for 19, whereas a higher
temperature (100 °C), a higher substrate dilution (1 mM), and
quenching after 1 h were used for the more difficult reaction of
diene 18.
corresponding potassium thiolate. Remarkably, the pyridine
molecule easily dissociates in solution, releasing a substantial
fraction of the 14-electron metathesis-active species even at low
temperature and at relatively high catalyst concentrations. The
facile pyridine dissociation and a sterically nondemanding
monodentate alkylidene ligand ensure that these precatalysts
initiate quickly. In comparison to the previously reported
catalysts based on the o-isopropoxybenzylidene chelate, the new
fast-initiating catalysts are much more selective toward olefin
metathesis, due to their reduced tendency to olefin isomer-
ization, while maintaining a comparable Z selectivity. They
reach a moderate to high Z selectivity with appreciable yields in
self-metathesis of terminal olefins after only a few minutes at
room temperature for most of the tested substrates. They
tolerate acidic substrates and mediate the first52,53 Z-selective
synthesis of α,ω-dicarboxylic acid alkenes via self-metathesis of
ω-alkenoic acids. Finally, they are also efficient at the high
substrate dilutions of macrocylic ring closing metathesis and
resist temperatures up to and above 100 °C.
The products of mRCM of 18 and 19 were purified using
silica gel chromatography, and the Z content was determined
36
1
ASSOCIATED CONTENT
* Supporting Information
by H NMR spectroscopy following literature procedures.
■
Lactones 18a and 19a were isolated in 38% and 56% yields,
respectively, with Z selectivities of 62% and 57%. Although
both the MAP catalysts (1 and 2) and the cyclometalated
ruthenium catalysts (3) have given higher Z selectivities, the
yields are comparable (see Scheme 4). The yields and reaction
times of 14b are particularly impressive in comparison with the
earlier generations of bottom-pathway monothiolate-substi-
tuted catalysts, which hardly promoted mRCM reactions at
all,50 demonstratring that even small changes in molecular
structure of a precatalyst may have dramatic consequences for
the catalytic performance. Specifically, as we have seen, the low
apparent activity of the o-isopropoxybenzylidene-coordinated
catalysts is due to slow initiation, and replacing the alkylidene
ether chelate with a nonchelating carbene and a pyridine
drastically improves the catalytic efficiency.
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.7b00441.
Experimental details, additional spectral and analytical
data, and crystallographic data for 11b and 14b (PDF)
Accession Codes
CCDC 1543473−1543474 contain the supplementary crystal-
lographic data for this paper. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/data_request/cif, or by email-
ing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
Corresponding Authors
*E-mail for G.O.: Giovanni.Occhipinti@uib.no.
*E-mail for V.R.J.: Vidar.Jensen@uib.no.
■
CONCLUSIONS
■
A new class of highly active and Z-selective Ru monothiolate
alkylidenes for olefin metathesis has been developed. These 16-
electron compounds are stabilized by a pyridine molecule and
are prepared in one step by reacting readily accessible pyridine-
stabilized third-generation Grubbs-type catalysts with the
ORCID
Giovanni Occhipinti: 0000-0002-7279-6322
Karl W. Tornroos: 0000-0001-6140-5915
̈
G
DOI: 10.1021/acs.organomet.7b00441
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(26) Occhipinti, G.; Hansen, F. R.; Tornroos, K. W.; Jensen, V. R. J.
̈
Vidar R. Jensen: 0000-0003-2444-3220
Am. Chem. Soc. 2013, 135, 3331−3334.
Notes
(27) Johns, A. M.; Ahmed, T. S.; Jackson, B. W.; Grubbs, R. H.;
Pederson, R. L. Org. Lett. 2016, 18, 772−775.
The authors declare no competing financial interest.
(28) Ahmed, T. S.; Grubbs, R. H. J. Am. Chem. Soc. 2017, 139, 1532−
1537.
ACKNOWLEDGMENTS
■
(29) Liu, P.; Xu, X.; Dong, X.; Keitz, B. K.; Herbert, M. B.; Grubbs,
R. H.; Houk, K. N. J. Am. Chem. Soc. 2012, 134, 1464−1467.
(30) Nelson, J. W.; Grundy, L. M.; Dang, Y.; Wang, Z.-X.; Wang, X.
Organometallics 2014, 33, 4290−4294.
The authors gratefully acknowledge financial support from the
Research Council of Norway via the GASSMAKS (grant
number 208335) and FORNY2020 (239288) programs and via
the Norwegian NMR Platform, NNP (226244). The authors
also thank Nataliya Kostenko for a preliminary screening of
monodentate donors and parent ruthenium-based catalysts
from which to start the synthesis of complexes 11 and 14 and
Theresa Schilling for preparing lactone 18a. Bjarte Holmelid is
thanked for assistance with the HRMS (DART and ESI−)
analyses.
(31) Koh, M. J.; Nguyen, T. T.; Lam, J. K.; Torker, S.; Hyvl, J.;
Schrock, R. R.; Hoveyda, A. H. Nature 2017, 542, 80−85.
(32) Furstner, A. Science 2013, 341, 1357−1364.
̈
(33) Shahane, S.; Bruneau, C.; Fischmeister, C. ChemCatChem 2013,
5, 3436−3459.
(34) Herbert, M. B.; Grubbs, R. H. Angew. Chem., Int. Ed. 2015, 54,
5018−5024.
(35) Werrel, S.; Walker, J. C. L.; Donohoe, T. J. Tetrahedron Lett.
2015, 56, 5261−5268.
REFERENCES
(36) Marx, V. M.; Herbert, M. B.; Keitz, B. K.; Grubbs, R. H. J. Am.
Chem. Soc. 2013, 135, 94−97.
■
(1) Hoveyda, A. H.; Zhugralin, A. R. Nature 2007, 450, 243−251.
(2) Handbook of metathesis, 2nd ed.; Grubbs, R. H., Wenzel, A. G.,
O’Leary, D. J., Khosravi, E., Eds.; Wiley-VCH: Weinheim, Germany,
2015; Vols. 1−3.
(3) Higman, C. S.; Lummiss, J. A. M.; Fogg, D. E. Angew. Chem., Int.
Ed. 2016, 55, 3552−3565.
(37) Marx, V. M.; Rosebrugh, L. E.; Herbert, M. B.; Grubbs, R. H.
Top. Organomet. Chem. 2014, 48, 1−17.
(38) Koh, M. J.; Khan, R. K. M.; Torker, S.; Yu, M.; Mikus, M. S.;
Hoveyda, A. H. Nature 2015, 517, 181−186.
(39) Vyboishchikov, S. E.; Buhl, M.; Thiel, W. Chem. - Eur. J. 2002, 8,
̈
3962−3975.
(4) Mol, J. C. J. Mol. Catal. A: Chem. 2004, 213, 39−45.
(5) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18−29.
(6) Clavier, H.; Grela, K.; Kirschning, A.; Mauduit, M.; Nolan, S. P.
Angew. Chem., Int. Ed. 2007, 46, 6786−6801.
(7) Samojlowicz, C.; Borre, E.; Mauduit, M.; Grela, K. Adv. Synth.
Catal. 2011, 353, 1993−2002.
(8) Borre, E.; Rouen, M.; Laurent, I.; Magrez, M.; Caijo, F.; Crevisy,
C.; Solodenko, W.; Toupet, L.; Frankfurter, R.; Vogt, C.; Kirschning,
A.; Mauduit, M. Chem. - Eur. J. 2012, 18, 16369−16382.
(9) Gleeson, E. C.; Jackson, W. R.; Robinson, A. J. Tetrahedron Lett.
2016, 57, 4325−4333.
(10) Chhabra, S.; Belgi, A.; Bartels, P.; van Lierop, B. J.; Robinson, S.
D.; Kompella, S. N.; Hung, A.; Callaghan, B. P.; Adams, D. J.;
Robinson, A. J.; Norton, R. S. J. Med. Chem. 2014, 57, 9933−9944.
(11) Guidone, S.; Songis, O.; Nahra, F.; Cazin, C. S. J. ACS Catal.
2015, 5, 2697−2701.
(12) Binder, J. B.; Raines, R. T. Curr. Opin. Chem. Biol. 2008, 12,
767−773.
(13) Sundararaju, B.; Sridhar, T.; Achard, M.; Sharma, G. V. M;
Bruneau, C. Eur. J. Org. Chem. 2013, 2013, 6433−6442.
(14) Bidange, J.; Fischmeister, C.; Bruneau, C. Chem. - Eur. J. 2016,
22, 12226−12244.
(15) Samojlowicz, C.; Bieniek, M.; Zarecki, A.; Kadyrov, R.; Grela, K.
Chem. Commun. 2008, 6282−6284.
(16) Weitekamp, R. A.; Atwater, H. A.; Grubbs, R. H. J. Am. Chem.
Soc. 2013, 135, 16817−16820.
(17) Astruc, D.; Diallo, A. K.; Gatard, S.; Liang, L. Y.; Ornelas, C.;
Martinez, V.; Mery, D.; Ruiz, J. Beilstein J. Org. Chem. 2011, 7, 94−103.
(18) Liu, X.; Basu, A. J. Organomet. Chem. 2006, 691, 5148−5154.
(19) Salinas, B.; Ruiz-Cabello, J.; Lechuga-Vieco, A. V.; Benito, M.;
Herranz, F. Chem. - Eur. J. 2015, 21, 10450−10456.
(20) Bertrand, A.; Hillmyer, M. A. J. Am. Chem. Soc. 2013, 135,
10918−10921.
(21) Torker, S.; Koh, M. J.; Khan, R. K. M.; Hoveyda, A. H.
Organometallics 2016, 35, 543−562.
(22) Flook, M. M.; Jiang, A. J.; Schrock, R. R.; Muller, P.; Hoveyda,
(40) Romero, P. E.; Piers, W. E. J. Am. Chem. Soc. 2005, 127, 5032−
5033.
(41) Wenzel, A. G.; Grubbs, R. H. J. Am. Chem. Soc. 2006, 128,
16048−16049.
(42) Benitez, D.; Tkatchouk, E.; Goddard, W. A., III Chem. Commun.
2008, 6194−6196.
(43) Conrad, J. C.; Parnas, H. H.; Snelgrove, J. L.; Fogg, D. E. J. Am.
Chem. Soc. 2005, 127, 11882−11883.
(44) Torker, S.; Muller, A.; Sigrist, R.; Chen, P. Organometallics 2010,
̈
29, 2735−2751.
(45) Teo, P.; Grubbs, R. H. Organometallics 2010, 29, 6045−6050.
(46) Jensen, V. R.; Occhipinti, G.; Hansen, F. U.S. Patent 8,716,488
B2, 2014.
(47) Occhipinti, G.; Koudriavtsev, V.; Tornroos, K. W.; Jensen, V. R.
̈
Dalton Trans. 2014, 43, 11106−11117.
(48) Smit, W.; Koudriavtsev, V.; Occhipinti, G.; Tornroos, K. W.;
̈
Jensen, V. R. Organometallics 2016, 35, 1825−1837.
(49) Jensen, V. R.; Occhipinti, G. Novel Organometallic Catalysts. U.S.
Patent US 9,303,100 B2, Apr 5, 2016.
(50) Jensen, V. R.; Occhipinti, G. Improved olefin metathesis catalysts.
Int. Patent Appl. WO 2017009232, 2017.
(51) Bates, J. M. Ruthenium Catalysts for Olefin Metathesis:
Understanding the Boomerang Mechanism and Challenges Associated
with Stereoselectivity. Ph.D. thesis, University of Ottawa, Ottawa, ON,
Canada K1N 6N5, 2014.
(52) Keitz, B. K.; Endo, K.; Herbert, M. B.; Grubbs, R. H. J. Am.
Chem. Soc. 2011, 133, 9686−9688.
(53) Highly stereoretentive cross-metathesis between (Z)-2-butene-
1,4-diol and carboxylic acid containing substrates has been reported
with ruthenium catecholthiolate catalysts.38 However, Z-selective self-
metathesis of 1-alkenecarboxylic acid containing substrates has not yet
been reported.
(54) Sanford, M. S.; Love, J. A.; Grubbs, R. H. J. Am. Chem. Soc.
2001, 123, 6543−6554.
̈
(55) Occhipinti, G.; Vitali, K.; Tonroos, K. W.; Jensen, V. R. To be
̈
A. H. J. Am. Chem. Soc. 2009, 131, 7962−7963.
(23) Jiang, A. J.; Zhao, Y.; Schrock, R. R.; Hoveyda, A. H. J. Am.
Chem. Soc. 2009, 131, 16630−16631.
submitted for publication.
(56) The Cambridge Structural Database (CSD) of Cambridge
Crystallographic Data Centre (CCDC), version 5.37, updated May
2016.
(24) Endo, K.; Grubbs, R. H. J. Am. Chem. Soc. 2011, 133, 8525−
8527.
(57) Kadyrov, R.; Wolf, D.; Azap, C.; Ostgard, D. J. Top. Catal. 2010,
53, 1066−1072.
(25) Khan, R. K. M.; Torker, S.; Hoveyda, A. H. J. Am. Chem. Soc.
2013, 135, 10258−10261.
H
DOI: 10.1021/acs.organomet.7b00441
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(58) Poater, A.; Cavallo, L. Beilstein J. Org. Chem. 2015, 11, 1767−
1780.
(59) Trzaskowski, B.; Grela, K. Organometallics 2013, 32, 3625−
3630.
(60) Romero, P. E.; Piers, W. E.; McDonald, R. Angew. Chem., Int. Ed.
2004, 43, 6161−6165.
(61) Dubberley, S. R.; Romero, P. E.; Piers, W. E.; McDonald, R.;
Parvez, M. Inorg. Chim. Acta 2006, 359, 2658−2664.
(62) Songis, O.; Slawin, A. M. Z.; Cazin, C. S. J. Chem. Commun.
2012, 48, 1266−1268.
(63) Sanford, M. S.; Henling, L. M.; Day, M. W.; Grubbs, R. H.
Angew. Chem., Int. Ed. 2000, 39, 3451−3453.
(64) Coalter, J. N.; Bollinger, J. C.; Eisenstein, O.; Caulton, K. G.
New J. Chem. 2000, 24, 925−927.
(65) Leitao, E. M.; van der Eide, E. F.; Romero, P. E.; Piers, W. E.;
McDonald, R. J. Am. Chem. Soc. 2010, 132, 2784−2794.
(66) Keitz, B. K.; Endo, K.; Patel, P. R.; Herbert, M. B.; Grubbs, R. H.
J. Am. Chem. Soc. 2012, 134, 693−699.
(67) Hassam, M.; Taher, A.; Arnott, G. E.; Green, I. R.; van Otterlo,
W. A. L. Chem. Rev. 2015, 115, 5462−5569.
(68) Higman, C. S.; Plais, L.; Fogg, D. E. ChemCatChem 2013, 5,
3548−3551.
(69) This particular quenching procedure was used here because it
facilitates the separation of the organic compounds from the least polar
Ru catalysts (14a,b and 11b) via silica gel filtration. Using EVE as a
quenching reagent instead results in a partial elution (leaching) of
colored ruthenium species.
(70) Nelson, D. J.; Percy, J. M. Dalton Trans. 2014, 43, 4674−4679.
(71) Keitz, B. K. In Handbook of metathesis, 2nd ed.; Grubbs, R. H.,
Wenzel, A. G., O’Leary, D. J., Khosravi, E., Eds.; Wiley-VCH:
Weinheim, Germany, 2015; Vol. 3.
(72) Burlison, J. A.; Blagg, B. S. J. Org. Lett. 2006, 8, 4855−4858.
(73) Kusuma, B. R.; Peterson, L. B.; Zhao, H.; Vielhauer, G.;
Holzbeierlein, J.; Blagg, B. S. J. J. Med. Chem. 2011, 54, 6234−6253.
(74) Perlman, N.; Albeck, A. Synth. Commun. 2000, 30, 4443−4449.
(75) Gensler, W. J.; Prasad, R. S.; Chaudhuri, A. P.; Alam, I. J. Org.
Chem. 1979, 44, 3643−3652.
(76) Jayasuriya, N.; Bosak, S.; Regen, S. L. J. Am. Chem. Soc. 1990,
112, 5844−5850.
(77) Jayasuriya, N.; Bosak, S.; Regen, S. L. J. Am. Chem. Soc. 1990,
112, 5851−5854.
(78) Hagiwara, H.; Adachi, T.; Nakamura, T.; Hoshi, T.; Suzuki, T.
Nat. Prod. Commun. 2012, 7, 913−915.
(79) Gensler, W. J.; Schlein, H. N. J. Am. Chem. Soc. 1955, 77, 4846−
4849.
(80) Matson, J. B.; Grubbs, R. H. Macromolecules 2010, 43, 213−221.
(81) Rodefeld, L.; Tochtermann, W. Tetrahedron 1998, 54, 5893−
5898.
(82) Furstner, A.; Guth, O.; Rumbo, A.; Seidel, G. J. Am. Chem. Soc.
̈
1999, 121, 11108−11113.
(83) Wang, C.; Yu, M.; Kyle, A. F.; Jakubec, P.; Dixon, D. J.; Schrock,
R. R.; Hoveyda, A. H. Chem. - Eur. J. 2013, 19, 2726−2740.
I
DOI: 10.1021/acs.organomet.7b00441
Organometallics XXXX, XXX, XXX−XXX