Synthesis and catalytic evaluation of ruthenium-arene complexes bearing imidazol(in)ium-2-thiocarboxylate ligands

Article abstract of DOI:10.1021/om2006529

Five new complexes with the generic formula [RuCl₂(p-cymene) (SOC·NHC)] (2-6) were isolated in high yields by reacting the [RuCl ₂(p-cymene)]₂ dimer with a range of imidazol(in)ium-2- thiocarboxylate zwitterions bearing cyclohexyl, 2,4,6-trimethylphenyl (mesityl), or 2,6-diisopropylphenyl groups on their nitrogen atoms in CH ₂Cl₂ at -20 °C. All the products were fully characterized by IR and NMR spectroscopy, and the molecular structures of [RuCl₂(p-cymene)(SOC·IMes)] (3) and [RuCl₂(p- cymene)(SOC·SIMes)] (5) were determined by X-ray diffraction analysis. Coordination of the NHC·COS ligands took place via the sulfur atom. A remarkable shielding of the methine proton on the p-cymene isopropyl group was observed by ¹H NMR spectroscopy for complexes 3-6. It is most likely caused by the aromatic ring current of a neighboring mesityl or 2,6-diisopropylphenyl substituent. The catalytic activity of compounds 2-6 was probed in the ring-opening metathesis polymerization (ROMP) of cyclooctene, in the atom transfer radical polymerization (ATRP) of methyl methacrylate, and in the synthesis of enol esters from 1-hexyne and 4-acetoxybenzoic acid. In all these reactions, the [RuCl₂(p-cymene)(SOC·NHC)] complexes displayed performances slightly inferior to those exhibited by [RuCl ₂(p-cymene)(NHC)] species that result from the reaction of [RuCl ₂(p-cymene)]₂ with NHC·CO₂ inner salts. However, they were significantly better catalyst precursors than the much more robust chelates of the [RuCl(p-cymene)(S₂C·NHC)]PF₆ type obtained by coordination of NHC·CS₂ betaines to the ruthenium dimer. These results suggest that the Ru-(SOC·NHC) motif undergoes a dethiocarboxylation under the experimental conditions adopted for the catalytic tests and leads to the same elusive Ru-NHC active species as the preformed [RuCl₂(p-cymene)(NHC)] family of complexes.

Full text of DOI:10.1021/om2006529

Article
pubs.acs.org/Organometallics
Synthesis and Catalytic Evaluation of Ruthenium−Arene
Complexes Bearing Imidazol(in)ium-2-thiocarboxylate Ligands
Morgan Hans,† Quentin Willem,† Johan Wouters,‡ Albert Demonceau,† and Lionel Delaude*,†
†Laboratory of Macromolecular Chemistry and Organic Catalysis, Institut de Chimie (B6a), Universite
Sart-Tilman par 4000 Liege, Belgium
‡Department of Chemistry, Facultes
Universitaires N.-D. de la Paix, 61 Rue de Bruxelles, 5000 Namur, Belgium
́ ̀
de Liege,
̀
́
S
* Supporting Information
ABSTRACT: Five new complexes with the generic formula
[RuCl2(p-cymene)(SOC·NHC)] (2−6) were isolated in high
yields by reacting the [RuCl2(p-cymene)]2 dimer with a range
of imidazol(in)ium-2-thiocarboxylate zwitterions bearing
cyclohexyl, 2,4,6-trimethylphenyl (mesityl), or 2,6-diisopropyl-
phenyl groups on their nitrogen atoms in CH2Cl2 at −20 °C.
All the products were fully characterized by IR and NMR
spectroscopy, and the molecular structures of [RuCl2(p-cymene)(SOC·IMes)] (3) and [RuCl2(p-cymene)(SOC·SIMes)] (5)
were determined by X-ray diffraction analysis. Coordination of the NHC·COS ligands took place via the sulfur atom. A
1
remarkable shielding of the methine proton on the p-cymene isopropyl group was observed by H NMR spectroscopy for
complexes 3−6. It is most likely caused by the aromatic ring current of a neighboring mesityl or 2,6-diisopropylphenyl
substituent. The catalytic activity of compounds 2−6 was probed in the ring-opening metathesis polymerization (ROMP) of
cyclooctene, in the atom transfer radical polymerization (ATRP) of methyl methacrylate, and in the synthesis of enol esters from
1-hexyne and 4-acetoxybenzoic acid. In all these reactions, the [RuCl2(p-cymene)(SOC·NHC)] complexes displayed per-
formances slightly inferior to those exhibited by [RuCl2(p-cymene)(NHC)] species that result from the reaction of [RuCl2-
(p-cymene)]2 with NHC·CO2 inner salts. However, they were significantly better catalyst precursors than the much more robust
chelates of the [RuCl(p-cymene)(S2C·NHC)]PF6 type obtained by coordination of NHC·CS2 betaines to the ruthenium dimer.
These results suggest that the Ru−(SOC·NHC) motif undergoes a dethiocarboxylation under the experimental conditions
adopted for the catalytic tests and leads to the same elusive Ru−NHC active species as the preformed [RuCl2(p-cymene)-
(NHC)] family of complexes.
the corresponding [RuCl2(p-cymene)(NHC)] complexes in
INTRODUCTION
■
high yields (Scheme 1).10 This methodology was successfully
Since they were first isolated by Arduengo in 1991,1 stable N-
heterocyclic carbenes (NHCs) based on the imidazole ring
system have found countless applications in organocatalysis and
organometallic chemistry.2 Due to their high nucleophilicity,
they readily add to a wide range of organic compounds,
including allenes, ketenes, and heteroallenes of the XCY
type.3 Upon reaction with carbon dioxide, they form inner salts,
which can be stored and handled with no particular
precautions.4 Such NHC·CO2 zwitterions readily lose their
CO2 moiety upon heating or dissolution. Hence, they act as
convenient surrogates to air- and moisture-sensitive free
carbenes for organometallic synthesis5 and organocatalytic
applications.6 In 2006, we first took advantage of their lability to
generate active species in palladium-catalyzed Suzuki−Miyaura
cross-coupling reactions7 and ruthenium-promoted olefin
metathesis and cyclopropanation processes.8 The next year,
Beller and co-workers adopted a similar strategy to generate
palladium−NHC catalysts in situ for the telomerization of 1,3-
butadiene with amines.9 In 2009, we further investigated the
reactivity of imidazolium-2-carboxylates toward the [RuCl2(p-
cymene)]2 dimer and we confirmed that these betaines cleanly
underwent a decarboxylative cleavage upon heating to afford
Scheme 1. Synthesis of Ruthenium−Arene Complexes
Bearing NHC Ligands from [RuCl2(p-cymene)]2 and
Imidazolium-2-carboxylates
Received: July 20, 2011
Published: October 26, 2011
© 2011 American Chemical Society
6133
dx.doi.org/10.1021/om2006529|Organometallics 2011, 30, 6133−6142

Organometallics
Article
extended to the preparation of various second-generation
ruthenium−alkylidene catalysts for olefin metathesis.11,12
Zwitterionic products are also obtained when carbon
disulfide is reacted with NHCs or precursors thereof.13 The
NHC·CS2 betaines differ from the NHC·CO2 series in that the
dithiocarboxylate moiety shows no significant lability upon
reaction with metals.3 The coordination chemistry of these
stable crystalline compounds remained almost unexplored until
2009, when a report based on ruthenium−arene complexes
provided strong experimental evidence for the formation of
cationic species of the [RuCl(p-cymene)(S2C·NHC)]+ type, in
which the dithiocarboxylate group acted as a κ2S,S′ chelating
ligand (Scheme 2).10 Further investigations on ruthenium or
toward the [RuCl2(p-cymene)]2 dimer. More specifically, we
describe the synthesis of five ruthenium−arene complexes
based on these zwitterionic ligands and we probe their ability to
serve as catalyst precursors for olefin metathesis, atom transfer
radical polymerization, and enol ester synthesis.
RESULTS AND DISCUSSION
■
Synthesis of Ruthenium−Arene Complexes. In order
to prepare [RuCl2(p-cymene)(NHC)] complexes, the
[RuCl2(p-cymene)]2 dimer was refluxed for 2 h in THF with
2 equiv of NHC·CO2 zwitterions (cf. Scheme 1). In these
experiments, a slow stream of argon was applied on top of the
condenser to help displace carbon dioxide.10 When NHC·CS2
betaines were reacted under the same conditions, elimination of
carbon disulfide did not occur, even after prolonged heating.
The coordination of these stable inner salts to a ruthenium
center was accomplished by heating stoichiometric amounts of
[RuCl2(p-cymene)]2 (1 equiv), an imidazol(in)ium-2-dithio-
carboxylate, and KPF6 (2 equiv each) for 1 h at 60 °C in
ethanol (cf. Scheme 2). This simple aerobic protocol afforded a
series of complexes with the generic formula [RuCl(p-
cymene)(S2C·NHC)]PF6 in high yields and purities.10 Because
thermogravimetric analyses had shown that NHC·COS
betaines displayed a thermal stability intermediate between
those of NHC·CO2 and NHC·CS2 adducts,17 we decided to
probe their reactivity toward the ruthenium dimer using both
experimental procedures. Zwitterions derived from five
representative saturated or unsaturated NHCs bearing mesityl
(Mes), 2,6-diisopropylphenyl (Dip), or cyclohexyl (Cy)
substituents on their nitrogen atoms were elected as starting
materials for these investigations (Scheme 3). Unfortunately, all
Scheme 2. Synthesis of Ruthenium−Arene Complexes
Bearing Imidazol(in)ium-2-dithiocarboxylate Ligands
Scheme 3. Imidazol(in)ium-2-thiocarboxylates Used in This
Work
osmium−alkenyl compounds confirmed this binding mode,
although the most bulky imidazolium-2-dithiocarboxylates
under scrutiny displayed a noninnocent behavior in some
instances, thereby allowing an unexpected migration of the
alkenyl fragment to take place.14 In 2010, the inorganic
chemistry of NHC·CS2 zwitterions was extended to the
synthesis of gold(I) complexes and gold nanoparticles.15 On
the catalysis front, the activity of five [RuCl(p-cymene)-
(S2C·NHC)]PF6 complexes was probed in various ring-closing
metathesis (RCM) and ring-opening metathesis polymerization
(ROMP) reactions but, as expected, these robust, 18-electron
species were almost inactive.10 Likewise, they were poorly
active and selective catalyst precursors for the addition of 4-
acetoxybenzoic acid to 1-hexyne. Their efficiency in the
synthesis of enol esters markedly increased, however, when
reactions were carried out at 160 °C under microwave or
conventional heating in a sealed tube.16
In view of the rich chemistry exhibited by NHC·CO2 and
NHC·CS2 betaines and the marked differences existing
between these two types of compounds, we launched a
program to investigate the structural properties and the
reactivity of the related NHC·COS zwitterions. In our first
article, we reported on the synthesis and characterization of five
imidazol(in)ium-2-thiocarboxylates bearing alkyl or aryl groups
on their nitrogen atoms and their use as organocatalysts for
acylation/transesterification reactions and for the benzoin
condensation.17 Another publication focused on the use of
these oxygen−sulfur mixed-donor ligands for the preparation of
cationic gold(I) complexes.18 In this contribution, we further
investigate the coordination chemistry of NHC·COS betaines
our attempts to isolate pure compounds resulting from a full
dethiocarboxylation or chelation of these imidazol(in)ium-2-
thiocarboxylates failed. In all these exploratory trials, complex
mixtures of unidentified products were always obtained. Among
other things, we noted that decreasing the temperature usually
led to cleaner reaction mixtures and that KPF6 did not seem to
have any influence on the reaction course. Replacement of this
salt with AgBF4 to ease the abstraction of a chloro ligand and the
formation of a cationic κ2O,S chelate was equally ineffective.
On the basis of the observations described above, we devised
a new strategy to favor the coordination of the thiocarboxylate
group onto ruthenium as a monodentate ligand. To achieve this
6134
dx.doi.org/10.1021/om2006529|Organometallics 2011, 30, 6133−6142

Organometallics
Article
goal, reactions were performed in dichloromethane at low
temperature under a static inert atmosphere. Initial experiments
were carried out at room temperature or at 0 °C. Subsequently,
we found that the best results were obtained when [RuCl2-
(p-cymene)]2 (1) and a NHC·COS adduct in 1:2 molar
proportions were stirred for 1 h in an ice/salt bath maintained
at −20 °C (Scheme 4). Under these conditions, five new
remarkable shielding effect, due to a favorable spatial
disposition between the methine proton and the aromatic
ring of a neighboring mesityl or 2,6-diisopropylphenyl group
(vide infra).
In addition to the various C−C and C−H vibration bands,
three major absorptions were clearly visible in the FT-IR
spectra of the five ruthenium−arene complexes under
investigation (Table 1). They were assigned to the stretching
Scheme 4. Synthesis of Ruthenium−Arene Complexes
Bearing Imidazol(in)ium-2-thiocarboxylate Ligands
Table 1. IR Stretching Vibration Bands of [RuCl(p-cymene)-
(SOC·NHC)] Complexes (2−6) and NHC·COS
a
Zwitterions
[RuCl2(p-cymene)
(SOC·NHC)]
NHC·COS
νCO
νCN
νCS
νCO
νCN
νCS
̅
̅
̅
̅
̅
̅
b
b
b
NHC
ICy
(cm−1
1571
)
(cm−1
)
(cm−1
934
)
(cm−1
1524
)
(cm−1
)
(cm−1
954
)
1482,
1451
1484,
1450
1486,
1464
1479
IMes
1568
1488,
1469
910
1532
917
IDip
1572
1558
1474
917
1531
1545
937
SIMes
1482,
1470
1029
1483
1050,
1037
SIDip
1573,
1553
1465,
1444
1043
1559,
1538
1465,
1445
1045
a
b
Spectra were recorded in KBR pellets. Data from ref 17.
ruthenium−arene complexes with the generic formula
[RuCl2(p-cymene)(SOC·NHC)] were isolated in high yields
after a simple workup that involved solvent removal and
washing with n-pentane. Compounds 2−6 were obtained as
orange microcrystalline powders and fully characterized by
various analytical techniques. Except for the SIMes derivative
(5), they were all very soluble in CH2Cl2 and CHCl3. Their
stability in solution was, however, limited, especially in the
presence of air and moisture. 1H NMR monitoring of a CD2Cl2
solution of complex 3 in a capped tube under an inert
atmosphere showed that it reverted to [RuCl2(p-cymene)]2 (1)
and IMes·COS within a few hours at 40 °C. Adding a
stoichiometric amount of triphenylphosphine to complex 5 in
CD2Cl2 at room temperature led to the instantaneous
of the CO, CN, and CS bonds within the imidazol(in)ium-2-
thiocarboxylate ligands by comparison with values previously
reported for miscellaneous imidazol(in)ium20 or thiocarboxy-
late derivatives.21,22 In some instances, these absorptions were
further split into two components, both of which are given in
Table 1. It should be pointed out that all these assignments are
tentative and should be taken with caution. For the sake of
comparison, the corresponding νCO, νCN, and νCS values
̅
̅
̅
recorded for the free NHC·COS zwitterions are also included
in Table 1.17 The νCN wavenumbers did not vary significantly
̅
upon binding to the ruthenium center, which indicates that the
imidazol(in)ium moieties of the ligands were not affected by
complexation. These results are in line with the general
observation that there is no electronic communication between
the anionic and the cationic parts of NHC·CXY adducts
(X, Y  O, S, Se).10,23 Conversely, the formation of complexes
2−6 markedly influenced the IR signatures of the thiocarbox-
ylate group. The CS stretching band was systematically
shifted to lower frequencies upon coordination, whereas the
CO absorption occurred at higher energies. These variations
strongly suggest that coordination of the COS− unit to the
metal center takes place via its sulfur atom only, thereby
weakening the CS bond and strengthening the CO bond.21
To establish with certainty that NHC·COS zwitterions acted
as monodentate S-ligands in ruthenium−arene complexes, we
determined the molecular structures of compounds 3 and 5 by
X-ray diffraction analysis. Crystals of [RuCl2(p-cymene)-
(SOC·SIMes)] (5) were obtained from a concentrated solution
in dichloromethane cooled to −18 °C. In the case of [RuCl2(p-
cymene)(SOC·IMes)] (3), n-pentane was added to further
decrease the polarity of the cold solution. Both solids
crystallized in the monoclinic system and belonged to the
same P21/c group. Thus, replacement of the 1,3-dimesitylimi-
dazolium ring in product 3 with its saturated imidazolinium
counterpart in 5 had only a very limited impact on the crystal
structure. Because the molecular parameters of free SIMes·COS
1
formation of [RuCl2(p-cymene)(PPh3)], as evidenced by H
and 31P NMR analyses. In the solid state, compounds 2−6
could be kept in open-air vials for more than 6 months at room
temperature without showing any sign of decomposition on 1H
NMR spectroscopy.
Structural Analysis. In order to minimize degradation
during the period of time required for 13C acquisition, NMR
spectra of complexes 2−6 were recorded in CD2Cl2 at 263 or
273 K. Most resonances fell within the usual ranges of chemical
shifts expected for coordinated arene19 and imidazol(in)ium-2-
thiocarboxylate ligands.18 For instance, 1H NMR analysis
showed the presence of two doublets located between 5.1
and 5.7 ppm with a common coupling constant of ca. 6 Hz for
the aromatic protons of the η6-(p-cymene) ligand, while 13C
NMR spectra displayed a strongly deshielded signal at ca. 186
ppm, due to the COS group. A puzzling anomaly was, however,
detected for the methine protons of the p-cymene isopropyl
group in complexes 3−6. Indeed, the characteristic septet of the
ArCH(CH3)2 motif was located between 1.6 and 1.9 ppm in
these compounds, more than 1 ppm upfield from its usual
position recorded at 2.83 ppm in [RuCl2(p-cymene)-
(SOC·ICy)] (2). HMBC and HSQC experiments confirmed
these assignments. We attribute this large discrepancy to a
6135
dx.doi.org/10.1021/om2006529|Organometallics 2011, 30, 6133−6142

Organometallics
Article
are available for comparison,17 only data pertaining to complex
5 are discussed here (see the Supporting Information for more
details on the crystal structure of complex 3 and an ORTEP
plot).
As anticipated, the metal center in [RuCl2(p-cymene)-
(SOC·SIMes)] (5) displayed the distinctive three-legged
piano-stool geometry observed in many other ruthenium−
arene species24 and was connected to the zwitterionic ligand via
its sulfur atom (Figure 1). Bond lengths and angles determined
(1.671(24) Å) than for single C−S bonds (1.808(10) Å). Last
but not least, the Ru−S bond length of 2.3897(17) Å was
similar to those recorded previously in related monodentate
ruthenium−thiocarboxylate complexes,21,22 whereas the much
greater distance between the metal center and the terminal
oxygen (3.64 Å) precluded any interaction between these two
atoms.
A careful inspection of the molecular structures of complexes
3 and 5 revealed that, in the solid state, the methine proton of
the p-cymene isopropyl group was pointing toward the center
of the neighboring mesityl ring. Indeed, trigonometric
calculations showed that H17 was located 3.132 Å from the
centroid defined by C5 and C29−C33, with a deviation to
orthogonality of 28.670° for [RuCl2(p-cymene)(SOC·IMes)]
(3) (Figure 2). Similar values of H17−Cg (3.151 Å) and
H17_perp (31.068°) were obtained from the crystal structure
of [RuCl2(p-cymene)(SOC·SIMes)] (5). We believe that this
geometry is responsible for the unusually low chemical shift
1
recorded for H17 in complexes 3−6 in H NMR spectroscopy
(vide supra). Indeed, hydrogen nuclei located over an aromatic
ring are known to experience a substantial shielding caused by
the magnetic field induced by the π-electron circulation.26 This
anisotropic effect is reminiscent of the significant high-field shift
observed for one of the aromatic mesityl protons in cis-
dichlororuthenium benzylidene complexes under the influence
of π−π stacking.27 In the related complex [RuCl(p-cymene)-
(S2C·IMes)]PF6 (7),10 which has an intermediate chelating
CS2− group between the carbenoid ligand and the metal center,
and [RuCl2(p-cymene)(IMes)] (8)28 sporting a direct Ru−
NHC bond, the corresponding methine proton was located
much farther away from the neighboring mesityl ring, thereby
precluding any magnetic interaction (Figure 2).
Figure 1. ORTEP representation of [RuCl2(p-cymene)(SOC·SIMes)]
(5) with thermal ellipsoids drawn at the 50% probability
level (hydrogen atoms are omitted for the sake of clarity).
Selected bond lengths (Å) and angles (deg): C1−C6 = 1.517(9),
C6−O1 = 1.227(13), C6−S1 = 1.688(8), C1−N1 = 1.317(12), C1−
N2 = 1.306(11), C2−C3 = 1.522(13), N1−C4 = 1.442(11), N2−C5 =
1.449(12), Ru1−Cl1 = 2.412(2), Ru1−Cl2 = 2.408(3), Ru1−S1 =
2.391(3); N1−C1−N2 = 113.2(8), O1−C6−S1 = 131.9(7); N1−C1−
C6−S1 = 80.4(1), C1−N1−C4−C20 = 82.2(1), C1−N2−C5−C29 =
−84.5(4).
Catalytic Tests. In order to assess the catalytic efficiency of
complexes 2−6, we first investigated the ROMP of cyclooctene
in chlorobenzene at 60 °C for 2 h using a monomer-to-catalyst
ratio of 250 (Scheme 5). In our laboratory, the polymerization
previously for the free SIMes·COS adduct17 remained largely
unaffected by coordination, except for the torsion angle
between the imidazolinium ring and the thiocarboxylate unit,
which was reduced from 95.0(1) to 80.4(1)°, probably due to
steric reasons. At 1.227(13) Å, the C−O distance was much
closer to values commonly reported for CO double bonds
(1.210(8) Å) than for single bonds (1.423(18) Å).25 The C−S
bond length of 1.682(8) Å in complex 5 is also indicative of
electron delocalization within the COS moiety, as this bond is
much closer to the values reported for CS double bonds
Scheme 5. ROMP of Cyclooctene
of this low-strain cycloolefin is commonly used as a benchmark
to evaluate the metathetical activity of new ruthenium−arene
complexes.12,29 An ordinary neon tube placed 10 cm away from
Figure 2. Molecular structures of [RuCl2(p-cymene)(SOC·IMes)] (3), [RuCl(p-cymene)(S2C·IMes)]PF6 (7), and [RuCl2(p-cymene)(IMes)] (8),
1
showing the orientation of the methine proton on the p-cymene isopropyl group relative to the neighboring mesityl ring and its H NMR chemical
shift.
6136
dx.doi.org/10.1021/om2006529|Organometallics 2011, 30, 6133−6142

Organometallics
Article
the Pyrex reaction flasks ensured a strong, reproducible lighting.
Previous investigations had shown that visible light was
required to generate metathetically active species from
[RuCl2(p-cymene)(NHC)] precursors, most likely via a total
or partial decoordination of the η6-arene ligand.30,31 Despite
such a photochemical activation,32 conversion of cyclooctene
stagnated below 25% with all the [RuCl2(p-cymene)-
(SOC·NHC)] complexes under study (Table 2). White fibers
counterparts under the experimental conditions adopted for
ROMP, although with a much reduced efficiency. This
hypothesis is supported by the fact that complexes 2−6 are
labile in solution and that NHC·COS zwitterions are liable to
release free carbenes upon thermolysis.17 To gain further
evidence in favor of this assumption, we refluxed a solution of
complex 3 in THF for 2 h under a slow stream of argon. NMR
analysis of the residue isolated after solvent evaporation and
purification through a short plug of alumina showed the
presence of [RuCl2(p-cymene)(IMes)] (8), but other un-
identified byproducts were also present in larger proportions.
Next, we focused our attention on the ruthenium-promoted
atom transfer radical polymerization (ATRP) of methyl
methacrylate (MMA) initiated by ethyl 2-bromo-2-methylpro-
panoate (Scheme 6). The monomer and the initiator were
Table 2. ROMP of Cyclooctene Catalyzed by Various
a
Ruthenium−Arene Complexes
monomer
conversn (%)
polymer
yield (%)
b
c
c
d
cat.
Mn(kg mol−1
)
Mw/Mn
σcis
2
3
4
5
6
7
8
21
18
17
19
18
1
traces
9
713
0.3
2.1
2.6
1.8
0.39
0.35
0.35
5
Scheme 6. ATRP of MMA
15
727
traces
95
85
715
2.0
0.25
a
Experimental conditions: Ru cat. (0.03 mmol), cyclooctene (1 mL,
stirred in an oil bath at 85 °C in the presence of complexes 2−6
until a highly viscous solution was obtained and the magnetic
stirring bar was immobilized. The initial monomer/initiator/
ruthenium molar proportions were 800:2:1. Under these
conditions, complexes 4 and 6 bearing 2,6-diisopropylphenyl-
substituted ligands did not afford a controlled polymerization.
Instead, they led to high-molecular-weight PMMA with a rather
broad polydispersity, thereby causing a rapid gelation of the
reaction mixtures (Table 3). Catalyst precursors 3 and 5
7.5 mmol), PhCl (5 mL), 2 h at 60 °C under visible light illumination.
b
Determined by GC using cyclooctane as internal standard.
c
d
Determined by SEC in THF with polystyrene calibration. Fraction
of cis double bonds in the polymer, determined by 13C NMR
spectroscopy.
of high-molecular-weight polyoctenamer containing mostly
trans double bonds were isolated with catalyst precursors 3
and 5 containing respectively the IMes·COS and SIMes·COS
ligands, while [RuCl2(p-cymene)(SOC·IDip)] (4) afforded a
small amount of ruthenium-tainted macromolecular product
with a low molecular weight and a broad polydispersity. In the
case of complexes 2 and 6, only traces of polymer precipitated
from the reaction mixtures. Soluble, uncharacterized oligomers
probably accounted for the mass balance.
Table 3. ATRP of MMA Catalyzed by Various Ruthenium−
a
Arene Complexes
b
b
cat.
reacn time (h) polymer yield (%) Mn (kg mol−1
)
Mw/Mn
2
3
4
5
6
7
48
80
59
49
59
50
42
49
97
1.56
1.59
1.62
1.59
1.63
1.68
1.35
100
10
41
For the sake of comparison, we also carried out the ROMP of
cyclooctene with [RuCl(p-cymene)(S2C·IMes)]PF6 (7), as a
representative complex bearing a NHC·CS2 ligand, and
[RuCl2(p-cymene)(IMes)] (8), which results from the reaction
of [RuCl2(p-cymene)]2 (1) with IMes·CO2 (Table 2). In line
with previous observations,10 the former cationic chelate was
completely inefficient, whereas the latter well-known initiator
afforded an almost quantitative conversion of monomer into a
high-molecular-weight polymer. It is worth emphasizing that
complex 3, based on the 1,3-dimesitylimidazolium-2-thiocar-
boxylate ligand, ranked between these two extremes.
Furthermore, when examining the influence of the NHC
moiety on the activity of complexes 2−6, we observed the same
trends as those recorded for preformed [RuCl2(p-cymene)-
(NHC)] complexes or, alternatively, for active species
generated in situ from [RuCl2(p-cymene)]2 (1), an imidazol-
(in)ium salt, and a base. Indeed, structure−activity relationships
derived from these catalytic systems had shown that (i) NHCs
with N-aryl substituents performed much better than those
with N-alkyl groups such as ICy, (ii) the CC double bond in
the imidazole ring of the NHC was not crucial to achieve high
catalytic efficiencies, and (iii) the IMes and SIMes ligands proved
superior to the more bulky IDip and SIDip for polymeriz-
ing cyclooctene.30,33 Hence, we suspect that [RuCl2(p-cymene)-
(SOC·NHC)] complexes are converted into the same elusive
Ru−NHC active species as their [RuCl2(p-cymene)(NHC)]
208
68
100
16
224
52
100
16
c
8
28
a
Experimental conditions: Ru cat. (0.0117 mmol), 0.1 M ethyl 2-
bromo-2-methylpropanoate in toluene (0.25 mL, 0.025 mmol), MMA
b
(1 mL, 9.35 mmol), 85 °C. Determined by SEC in THF with PMMA
c
calibration. Data from ref 34.
bearing respectively the IMes·COS and SIMes·COS ligands
displayed a higher initiation efficiency and gave polymers whose
molecular weights matched more closely those expected for a
well-behaved system (40 kg mol−1). At 1.59, the polydispersity
index obtained with these two complexes denoted, however,
that control was not optimal. With [RuCl2(p-cymene)-
(SOC·ICy)] (2) a slightly narrower molecular weight
distribution and a better yield were obtained, but the molecular
weight exceeded the target value by far.
Control experiments carried out with [RuCl(p-cymene)-
(S2C·IMes)]PF6 (7) and [RuCl2(p-cymene)(IMes)] (8)
helped us to better assess the influence of imidazol(in)ium-2-
thiocarboxylate ligands on the catalytic activity of ruthenium−
arene complexes (Table 3). As was observed for the ROMP of
cyclooctene, complex 3 was a much better catalyst precursor for
the ATRP of MMA than the corresponding dithiocarboxylate
6137
dx.doi.org/10.1021/om2006529|Organometallics 2011, 30, 6133−6142

Organometallics
Article
Scheme 7. Synthesis of Enol Esters from 4-Acetoxybenzoic Acid and 1-Hexyne
chelate (7), but displayed inferior performances compared to
the imidazolylidene-based complex 8.
Table 4. Microwave-Assisted Synthesis of Enol Esters
Catalyzed by [RuCl2(p-cymene)(SOC·NHC)] Complexes
a
(2−6)
To further appraise the potentials of complexes 2−6 in
homogeneous catalysis, we probed their activity in the synthesis
of enol esters. Indeed, various types of ruthenium−arene
complexes are known to catalyze the addition of carboxylic
acids to terminal alkynes, thereby affording vinyl esters in high
yields and with excellent selectivities.35 The hydrooxycarbony-
lation of 1-hexyne with 4-acetoxybenzoic acid was chosen as a
model reaction for these investigations (Scheme 7). Exper-
imental protocols that proved successful for enhancing the
catalytic activity of [RuCl(p-cymene)(S2C·NHC)]PF6 and
[RuCl2(p-cymene)(NHC)] complexes in earlier studies were
employed again.16 Thus, water-saturated toluene was used as a
solvent and sodium carbonate was added as an activator.36
Separation and quantification of the three possible products,
viz., 1-hexen-2-yl 4-acetoxybenzoate (Markovnikov addition
product, M) and the E and Z isomers of 1-hexen-1-yl 4-
acetoxybenzoate (anti-Markovnikov addition products, aME
and aMZ) were achieved by gas chromatography in the
presence of n-dodecane as an internal standard. Because 1-
hexyne was introduced in excess compared to the carboxylic
acid (1.5 equiv), dimerization products of this terminal alkyne
were also detected in the reaction media. In all cases, however,
they represented less than 1% of the total conversion.
In a first series of experiments, we performed a rapid catalytic
screening of the five ruthenium−thiocarboxylate complexes at
our disposal using pressure vials and a monomodal microwave
reactor. As demonstrated in 2009, such a device was very
convenient to quickly heat the reaction mixtures at temper-
atures well above the boiling point of 1-hexyne (71−72 °C)
and to speed up the synthesis of enol esters.37 As a matter of
fact, the microwave-assisted addition of 4-acetoxybenzoic acid
to 1-hexyne catalyzed by complexes 2−5 afforded high yields of
products after 10 min at 140 °C (Table 4). Only complex 6
stood below the 50% threshold under these conditions. In
terms of selectivity, there was a marked dichotomy between
complexes 2−4 prepared from imidazolium-2-thiocarboxylate
zwitterions, which strongly favored the formation of the
Markovnikov adduct (M), and complexes 5 and 6 bearing
respectively the SIMes·COS and SIDip·COS ligands that
afforded rather similar proportions of the three possible
products. Furthermore, monitoring the evolution of the
product distribution over a 30 min period under microwave
irradiation at 140 °C showed that the selectivity significantly
changed with time in the case of complexes 2−4, whereas it
remained almost invariant with catalyst precursor 6 and
underwent only slight modifications with 5 (see the Supporting
Information for kinetic plots).
b
product distribution (%)
b
cat.
yield (%)
M
aMZ
aME
2
3
4
5
6
84
84
67
61
35
68
23.5
19
8.5
4
77
58
34
8
35.5
18.5
35
29.5
38
43.5
a
Experimental conditions: Ru cat. (0.004 mmol), Na2CO3 (0.008
mmol), 4-acetoxybenzoic acid (0.5 mmol), 1-hexyne (0.75 mmol),
water-saturated toluene (2.6 mL), 10 min at 140 °C. Determined by
b
GC using n-dodecane as internal standard.
To better apprehend the influence of NHC·COS ligands on
the outcome of the reaction, we performed a second series of
more thorough catalytic tests on the addition of 4-
acetoxybenzoic acid to 1-hexyne (Scheme 7). These experi-
ments were carried out on a larger scale using Schlenk tubes
placed in an oil bath at 60 °C. With this revised setup, the
formation of enol esters was slowed down and it was possible to
monitor more carefully and more conveniently the reaction
course than in the microwave reactor. Time needed to reach
completion was considerably longer, and differences between
the various catalysts under investigation were more pro-
nounced. Under these conditions, [RuCl2(p-cymene)-
(SOC·ICy)] (2) emerged as the most efficient catalyst
precursor among the five ruthenium−thiocarboxylate com-
plexes under study (Table 5). Indeed, this compound afforded
Table 5. Synthesis of Enol Esters Catalyzed by Various
a
Ruthenium−Arene Complexes
b
product distribution (%)
b
cat.
yield (%)
M
aMZ
aME
2
3
4
5
6
100
66
44
23
21
11
99
83.5
79
13.5
16.5
20
3
4.5
9
71
20
36
44
42
34
3
20
38
c
9
24
42
c
10
83
14
a
Experimental conditions: Ru cat. (0.0092 mmol), Na2CO3 (0.0184
mmol), 4-acetoxybenzoic acid (1.15 mmol), 1-hexyne (1.725 mmol),
water-saturated toluene (6 mL), 50 h at 60 °C. Determined by GC
b
c
using n-dodecane as internal standard. Data from ref 16.
6138
dx.doi.org/10.1021/om2006529|Organometallics 2011, 30, 6133−6142

Organometallics
Article
a quantitative reaction within 50 h, whereas its most serious
contender, [RuCl2(p-cymene)(SOC·IMes)] (3), led to a 66%
yield within the same period of time. Altogether, the duration
needed to reach completion followed the sequence 2 (50 h) <
3 (100 h) < 4 (150 h) ≪ 5 (535 h) < 6 (650 h) (see the
Supporting Information for kinetic plots). Complex 2 was also
the most selective catalyst precursor and strongly favored the
formation of the Markovnikov adduct. However, it should be
pointed out that the product distribution did not remain
constant during the whole run. As noticed previously in the
microwave-assisted reactions, the proportion of 1-hexen-2-yl 4-
acetoxybenzoate (M) progressively increased over time to reach
a final value of ca. 84% (Figure 3).
ROMP of cyclooctene at 60 °C (vide supra). The poor results
obtained with precursors 5 and 6 in the present catalytic system
further support the intermediacy of [RuCl2(p-cymene)(NHC)]
complexes or active species derived thereofbecause SIMes
and SIDip are known to coordinate much more reluctantly to a
ruthenium−arene scaffold than their aromatic counterparts
IMes and IDip.38
In a third and final series of experiments, we examined the
possibility of generating active species in situ from the
[RuCl2(p-cymene)]2 dimer (1) and an imidazol(in)ium-2-
(thio)carboxylate. The addition of 4-acetoxybenzoic acid to 1-
hexyne was chosen once again as a case study for these catalytic
tests (Scheme 7). Results gathered in Table 6 showed that a 1:2
Table 6. Synthesis of Enol Esters Catalyzed by Various
Ruthenium−Arene Complexes Generated in Situ from
[RuCl2(p-cymene)]2 (1) and 1,3-Dimesitylimidazol(in)ium
a
Zwitterions
b
product distribution (%)
b
cat.
yield (%)
M
aMZ
aME
1
26
48
48
22
18
37.5
17
44.5
5.5
1 + IMes·COS
1 + IMes·CO2
1 + SIMes·COS
77.5
76
17.5
34.5
6.5
23
42.5
a
Experimental conditions: [RuCl2(p-cymene)]2 (0.0046 mmol),
NHC·CXY (0.0092 mmol), Na2CO3 (0.0184 mmol), 4-acetoxyben-
zoic acid (1.15 mmol), 1-hexyne (1.725 mmol), water-saturated
b
toluene (6 mL), 50 h at 60 °C. Determined by GC using n-dodecane
as internal standard.
Figure 3. Time course of the addition of 4-acetoxybenzoic acid to 1-
hexyne catalyzed by [RuCl2(p-cymene)(SOC·ICy)] (2) and
[RuCl2(p-cymene)(ICy)] (10) in an oil bath at 60 °C.
mixture of the ruthenium dimer and IMes·COS was less active
than [RuCl2(p-cymene)(SOC·IMes)] (3), as the yield dropped
from 66% to 48% after 50 h at 60 °C (cf. Table 5). Accordingly,
it took ca. 130 h to reach a complete conversion with the two-
component mixture vs 100 h with the preformed complex
(see the Supporting Information). Eventually, both systems
afforded almost identical product distributions, suggesting that
the same active species were at play. It is noteworthy that the
replacement of 1,3-dimesitylimidazolium-2-thiocarboxylate with
the corresponding carboxylate inner salt did not affect the
reaction rate but led to the same differences in the evolution of
selectivity that were already observed when comparing
preformed [RuCl2(p-cymene)(NHC)] and [RuCl2(p-cymene)-
(SOC·NHC)] complexes (cf. Figure 3). Two supplementary
trials were performed with [RuCl2(p-cymene)]2 (1) alone or in
the presence of SIMes·COS (Table 6). Results from these
control experiments unambiguously proved that the imidazo-
linium betaine did not have any significant influence on the
reaction outcome, as the activity of the ruthenium−arene dimer
(1) matched those displayed by [RuCl2(p-cymene)-
(SOC·SIMes)] (5) either preformed or generated in situ.
The same reaction profile had also been recorded with
[RuCl2(p-cymene)(SOC·SIDip)] (6) (cf. Table 5). These
data confirm the crucial importance of choosing a NHC ligand
based on the imidazole ring system rather than the imidazoline
cycle to achieve high catalytic efficiencies in the synthesis of
enol esters.
Comparison of the results obtained with [RuCl2(p-cymene)-
(SOC·ICy)] (2) and those reported previously for [RuCl(p-
cymene)(S2C·ICy)]PF6 (9) and [RuCl2(p-cymene)(ICy)]
(10) showed that the Ru−thiocarboxylate and Ru−NHC
complexes afforded almost identical yields and selectivities after
50 h (Table 5). Conversely, the dithiocarboxylate chelate was
largely inefficient at promoting the hydrooxycarbonylation of
1-hexyne with 4-acetoxybenzoic acid at 60 °C and required
microwave irradiation at 160 °C to become active.16 A close
examination of the reaction profiles of complexes 2 and 10
nevertheless revealed the existence of some minor, albeit
significant, differences between these two catalyst precursors. In
particular, monitoring the product distribution over time clearly
showed that [RuCl2(p-cymene)(ICy)] (10) led to invariant
proportions of the three possible products from the early stage
of the reaction to its completion, whereas the same final
distribution was attained only after ca. 25 h, with complex 2
sporting an intermediate COS group between the metal center
and the carbenoid moiety (Figure 3). Moreover, the reaction
occurred slightly more quickly with complex 10 than with
catalyst precursor 2. The same trends were observed when
comparing complexes 3 and 4 with the corresponding
[RuCl2(p-cymene)(NHC)] species (see the Supporting In-
formation). These data suggest that complexes 2−4 containing
an imidazolium-2-thiocarboxylate zwitterion might undergo a
progressive dethiocarboxylation leading to active species
bearing an imidazol-2-ylidene ligand under the experimental
conditions adopted for enol ester synthesis. A similar
transformation was already assumed to take place during the
CONCLUSION
■
By reacting the [RuCl2(p-cymene)]2 dimer (1) with 2 equiv of
NHC·COS betaines in dichloromethane at −20 °C, we were
able to synthesize five new ruthenium−arene complexes bearing
6139
dx.doi.org/10.1021/om2006529|Organometallics 2011, 30, 6133−6142

Organometallics
Article
solution was stirred for 1 h at −20 °C. The solvent was then
evaporated under high vacuum at −20 °C. The remaining solid was
washed with n-pentane (3 × 10 mL) and dried under high vacuum.
[RuCl2(p-cymene)(SOC·ICy)] (2). Orange powder (0.1940 g, 81%
imidazol(in)ium-2-thiocarboxylate ligands coordinated to the
metal center via their sulfur atom. Thus, complexes 2−6 with
the generic formula [RuCl2(p-cymene)(SOC·NHC)] were
isolated in high yields and fully characterized by various
analytical techniques. These compounds remained stable in
the solid state for more than 6 months in the open air but were
rather labile in solution. A remarkable shielding of the methine
1
yield). H NMR (400 MHz, CD2Cl2, 263 K): δ 7.20 (s, 2H, Im-C4,5),
3
3
5.71 (d, JHH = 5.9 Hz, 2H, p-cym CHar), 5.49 (d, JHH = 5.9 Hz, 2H,
3
p-cym CHar), 4.50 (m, 2H, CHN), 2.83 (sept, JHH = 6.9 Hz, 1H, p-
cym CH(CH3)2), 2.21 (s, 3H, p-cym CH3), 2.15−2.13 (m, 4H, Cy),
1
1.86−1.83 (m, 4H, Cy), 1.71−1.68 (m, 2H, Cy), 1.65−1.52 (m, 4H,
proton on the p-cymene isopropyl group was observed by H
3
Cy), 1.46−1.36 (m, 4H, Cy), 1.26 (d, JHH = 7.0 Hz, 6H, p-cym
NMR spectroscopy for complexes 3−6. It is most likely caused
by the aromatic ring current of a neighboring mesityl or 2,6-
diisopropylphenyl substituent, as evidenced by a detailed
analysis of the molecular structure of complexes 3 and 5.
The catalytic activity of compounds 2−6 was probed in the
ring-opening metathesis polymerization of cyclooctene, in the
atom transfer radical polymerization of methyl methacrylate,
and in the synthesis of enol esters from 1-hexyne and 4-acet-
oxybenzoic acid. In all these reactions, the [RuCl2(p-cymene)-
(SOC·NHC)] complexes displayed performances slightly inferior
to those exhibited by [RuCl2(p-cymene)(NHC)] species that
result from the reaction of [RuCl2(p-cymene)]2 with
NHC·CO2 inner salts. However, they were significantly better
catalyst precursors than the much more robust chelates of the
[RuCl(p-cymene)(S2C·NHC)]PF6 type obtained by coordina-
tion of NHC·CS2 zwitterions to the ruthenium dimer. These
results suggest that the Ru−(SOC·NHC) motif undergoes a
dethiocarboxylation under the experimental conditions adopted
for the catalytic tests and leads to the same elusive Ru−NHC
active species as the preformed [RuCl2(p-cymene)(NHC)]
family of complexes.
CH(CH3)2), 1.23−1.20 (m, 2H, Cy) ppm. 13C NMR (100 MHz,
CD2Cl2, 263 K): δ 188.2 (COS), 142.9 (Im-C2), 117.4 (Im-C4,5),
101.3 (p-cym), 100.6 (p-cym), 83.1 (p-cym), 80.9 (p-cym), 58.7
(CHN), 33.4 (Cy), 31.5 (p-cym), 25.3 (Cy), 25.0 (Cy), 22.5 (p-cym),
19.0 (p-cym) ppm. IR (KBr): ν 3086 (m), 2932 (s), 2857 (s), 1641
̅
(m), 1571 (s), 1482 (s), 1451 (s), 1385 (m), 1271 (w), 1200 (m),
1190 (m), 1171 (m), 1135 (m), 1057 (w), 934 (s), 896 (m), 855 (m),
804 (w) cm−1. HR-MS (ESI+): m/z calcd for C26H38ClN2ORuS
[(M − Cl)+], 563.143 69; found, 563.146 92. Anal. Calcd for
C26H38Cl2N2ORuS (598.63): C, 52.17; H, 6.40; N, 4.68; S, 5.36.
Found: C, 51.86; H, 6.27; N, 4.97; S, 5.29.
[RuCl2(p-cymene)(SOC·IMes)] (3). Orange powder (0.2495 g, 93%
1
yield). H NMR (400 MHz, CD2Cl2, 263 K): δ 7.26 (s, 2H, Im-C4,5),
7.08 (s, 4H, m-CH), 5.19 (d, 3JHH = 5.9 Hz, 2H, p-cym CHar), 5.13 (d,
3JHH = 5.9 Hz, 2H, p-cym CHar), 2.37 (s, 6H, p-CH3), 2.21 (s, 12H,
3
o-CH3), 1.97 (s, 3H, p-cym CH3), 1.93 (m, JHH = 6.9 Hz, 1H, p-cym
CH(CH3)2), 0.92 (d, 3JHH = 6.9 Hz, 6H, p-cym CH(CH3)2) ppm. 13
C
NMR (100 MHz, CD2Cl2, 263 K): δ 185.7 (COS), 144.6 (Im-C2),
141.7 (p-C), 135.4 (o-C), 131.2 (ipso-C), 129.8 (m-CH), 122.8 (Im-
C4,5), 102.1 (p-cym), 100.4 (p-cym), 83.4 (p-cym), 80.8 (p-cym), 31.1
(p-cym), 22.3 (p-cym), 21.4 (p-CH3), 18.6 (p-cym), 18.2 (o-CH3)
ppm. IR (KBr): ν 3160 (m), 3116 (m), 3091 (m), 2951 (m), 2921
̅
(m), 2863 (m), 1607 (m), 1568 (s), 1488 (s), 1469 (m), 1380 (m),
1229 (m), 1159 (m), 1038 (m), 1006 (m), 910 (s), 855 (m) 806 (w)
cm−1. HR-MS (ESI+): m/z calcd for C32H38ClN2ORuS [(M − Cl)+],
635.143 69; found, 635.142 43. Anal. Calcd for C32H38Cl2N2ORuS
(670.70): C, 57.30; H, 5.71; N, 4.18; S, 4.78. Found: C, 56.78; H, 5.61;
N, 4.24; S, 4.73.
EXPERIMENTAL SECTION
■
General Information. Unless otherwise specified, all the
syntheses were carried out under a dry argon atmosphere using
standard Schlenk techniques. Solvents were distilled from appropriate
drying agents and deoxygenated prior to use. Imidazol(in)ium-2-
thiocarboxylates were prepared according to the literature.17 The
[RuCl2(p-cymene)]2 dimer 1 was purchased from Strem. All the other
[RuCl2(p-cymene)(SOC·IDip)] (4). Orange powder (0.2597 g, 86%
yield). 1H NMR (400 MHz, CD2Cl2, 263 K): δ 7.59 (t, 3JHH = 7.8 Hz,
2H, p-CH), 7.36 (d, 3JHH = 7.8 Hz, 4H, m-CH), 7.29 (s, 2H, Im-C4,5),
3
3
5.25 (d, JHH = 5.8 Hz, 2H, p-cym CHar), 5.15 (d, JHH = 5.7 Hz, 2H,
p-cym CHar), 2.53 (sept, 3JHH = 6.7 Hz, 4H, CH(CH3)2), 1.91 (s, 3H,
1
chemicals were obtained from Aldrich. Unless otherwise specified, H
3
and 13C NMR spectra were recorded at 298 K with a Bruker DRX 400
spectrometer operating at 400.13 and 100.62 MHz, respectively.
Chemical shifts are listed in parts per million downfield from TMS and
are referenced from the solvent peaks or TMS. Infrared spectra were
recorded with a Perkin−Elmer Spectrum One FT-IR spectrometer.
Mass spectral analyses were performed on a Bruker Daltonics SolariX
FT-ICR spectrometer operating at 9.4 T in the Laboratory of Mass
p-cym CH3), 1.63 (sept, JHH = 6.7 Hz, 1H, p-cym CH(CH3)2), 1.34
(d, 3JHH = 6.7 Hz, 12H, CH3), 1.15 (d, 3JHH = 6.8 Hz, 12H, CH3), 0.81
(d, 3JHH = 6.8 Hz, 6H, p-cym CH(CH3)2) ppm. 13C NMR (100 MHz,
CD2Cl2, 263 K): δ 184.2 (COS), 145.9 (o-C), 145.4 (Im-C2), 131.9
(p-CH), 130.8 (ipso-C), 125.0 (m-CH), 123.5 (Im-C4,5), 103.4 (p-
cym), 99.0 (p-cym), 84.0 (p-cym), 79.5 (p-cym), 30.6 (p-cym), 29.6
(CH(CH3)2), 25.6 (CH3), 22.9 (CH3), 22.3 (p-cym), 18.6 (p-cym)
ppm. IR (KBr): ν 3062 (m), 2963 (s), 2929 (m), 2868 (m), 1641 (w),
Spectroscopy of the University of Lieg
carried out in the Laboratory of Pharmaceutical Chemistry at the
University of Liege. Size-exclusion chromatography was performed in
̀
e. Elemental analyses were
̅
1572 (s), 1474 (s), 1386 (m), 1365 (m), 1330 (m), 1275 (w), 1210
(m), 1162 (m), 1122 (w), 1060 (m), 1011 (w), 917 (s), 853 (w), 803
(m) cm−1. HR-MS (ESI+): m/z calcd for C38H50ClN2ORuS [(M −
Cl)+], 719.237 59; found, 719.236 43. Anal. Calcd for
C38H50Cl2N2ORuS (754.86): C, 60.46; H, 6.68; N, 3.71; S, 4.25.
Found: C, 59.69; H, 6.37; N, 3.76; S, 3.86.
̀
THF at 45 °C with a SFD S5200 autosampler liquid chromatograph
equipped with a SFD 2000 refractive index detector and a battery of
4 PL gel columns fitted in series (particle size 5 μm; pore sizes
105, 104, 103, and 102 Å; flow rate 1 mL/min). The molecular wei
ghts (not corrected) are reported versus monodisperse polystyrene or
poly(methyl methacrylate) standards used to calibrate the instrument.
Gas chromatography was carried out with a Varian 3900 instrument
equipped with a flame ionization detector and a WCOT fused silica
column (stationary phase CP-Sil 5CB, column length 15 m, inside
diameter 0.25 mm, outside diameter 0.39 mm, film thickness 0.25 μm).
Preparation of Ruthenium−Arene Complexes with Thio-
carboxylate Ligands. A 50 mL round-bottom flask equipped with a
magnetic stirring bar and capped with a three-way stopcock was
charged with [RuCl2(p-cymene)]2 (1; 0.1225 g, 0.2 mmol) and an
imidazol(in)ium-2-thiocarboxylate (0.4 mmol). The mixture was
cooled in a 3/1 w/w ice/salt bath at −20 °C before dry and cold
CH2Cl2 (10 mL) was added with a syringe. The resulting orange-red
[RuCl2(p-cymene)(SOC·SIMes)] (5). Orange powder (0.2341 g,
1
87% yield). H NMR (400 MHz, CD2Cl2, 273 K): δ 7.00 (s, 4H,
3
3
m-CH), 5.11 (d, JHH = 5.8 Hz, 2H, p-cym CHar), 5.07 (d, JHH = 5.9
Hz, 2H, p-cym CHar), 4.26 (s, 4H, Im-C4,5), 2.45 (s, 12H, o-CH3),
2.30 (s, 6H, p-CH3), 1.91 (s, 3H, p-cym CH3), 1.76 (m, 3JHH = 6.6 Hz,
3
1H, p-cym CH(CH3)2), 0.92 (d, JHH = 6.9 Hz, 6H, p-cym
CH(CH3)2) ppm. 13C NMR (100 MHz, CD2Cl2, 273 K): δ 187.2
(COS), 163.6 (Im-C2), 141.1 (p-C), 136.6 (o-C), 130.7 (ipso-C),
130.4 (m-CH), 101.5 (p-cym), 100.7 (p-cym), 83.1 (p-cym), 81.2 (p-
cym), 50.8 (Im-C4,5), 31.2 (p-cym), 22.5 (p-cym), 21.4 (p-CH3), 18.6
(p-cym), 18.5 (o-CH3) ppm. IR (KBr): ν 2985 (m), 2952 (m), 2921
̅
(m), 2864 (w), 1665 (m), 1609 (m), 1590 (s), 1558 (s), 1515 (m),
1482 (m), 1470 (m), 1390 (m), 1378 (m), 1275 (s), 1213 (w), 1199
6140
dx.doi.org/10.1021/om2006529|Organometallics 2011, 30, 6133−6142

Organometallics
Article
(w), 1029 (s), 885 (m), 855 (m), 805 (m) cm−1. HR-MS (ESI+): m/z
calcd for C32H40ClN2ORuS [(M − Cl)+], 637.159 34; found, 637.158
28. Anal. Calcd for C32H40Cl2N2ORuS (672.71): C, 57.13; H, 5.99; N,
4.16; S, 4.77. Found: C, 57.14; H, 5.98; N, 4.37; S, 4.53.
reaction mixture was heated for 16−100 h in an oil bath at 85 °C. It
was cooled to room temperature, diluted with THF (5 mL) and poured
into n-heptane (600 mL) with vigorous stirring. The precipitated
polymer was filtered with suction, dried overnight under dynamic
vacuum, and characterized by SEC.
Microwave-Assisted Enol Ester Synthesis. A 10 mL glass vial
containing a magnetic stirring bar was charged with a ruthenium−arene
complex (0.004 mmol), sodium carbonate (0.85 mg, 0.008 mmol),
and 4-acetoxybenzoic acid (90 mg, 0.5 mmol). Air was expelled by
applying three vacuum−argon cycles before a stock solution
containing 1-hexyne (0.75 mmol) and n-dodecane (internal standard)
in water-saturated toluene (2.6 mL) was added. The vial was capped
under nitrogen and heated to 140 °C (monitored by IR sensor) for
10 min in a CEM Discover instrument using a dynamic heating rate
program (150 W maximum microwave power). No simultaneous
cooling was applied. After rapid air cooling by the unit, the reaction
mixture was analyzed by gas chromatography (GC).
Enol Ester Synthesis with Conventional Heating. A 25 mL
Schlenk tube equipped with a magnetic stirring bar and capped with a
three-way stopcock was charged with a ruthenium−arene complex
(0.0092 mmol), sodium carbonate (1.95 mg, 0.0184 mmol), and 4-
acetoxybenzoic acid (207 mg, 1.15 mmol). Air was expelled by
applying three vacuum−argon cycles before a stock solution
containing 1-hexyne (1.725 mmol) and n-dodecane (internal stand-
ard) in water-saturated toluene (6 mL) was added with a dried syringe
under nitrogen. The reaction mixture was stirred in an oil bath at
60 °C until GC analysis of samples withdrawn at regulat time intervals
showed that a full conversion had occurred.
[RuCl2(p-cymene)(SOC·SIDip)] (6). Orange powder (0.2876 g, 95%
yield). 1H NMR (400 MHz, CD2Cl2, 263 K): δ 7.49 (t, 3JHH = 7.6 Hz,
3
3
2H, p-CH), 7.30 (d, JHH = 7.8 Hz, 4H, m-CH), 5.23 (d, JHH = 5.7
3
Hz, 2H, p-cym CHar), 5.12 (d, JHH = 5.7 Hz, 2H, p-cym CHar), 4.27
(s, 4H, Im-C4,5), 3.14 (sept, JHH = 6.7 Hz, 4H, CH(CH3)2), 1.81 (s,
3
3
3H, p-cym CH3), 1.61 (sept, JHH = 6.9 Hz, 1H, p-cym CH(CH3)2),
1.41 (d, 3JHH = 6.6 Hz, 12H, CH3), 1.29 (d, 3JHH = 6.8 Hz, 12H, CH3),
0.82 (d, JHH = 6.8 Hz, 6H, p-cym CH(CH3)2) ppm. 13C NMR (100
3
MHz, CD2Cl2, 263 K): δ 185.6 (COS), 163.0 (Im-C2), 147 (o-C),
131.2 (ipso-C), 130.1 (p-CH), 125.4 (m-CH), 101.9 (p-cym), 99.8
(p-cym), 83.5 (p-cym), 79.5 (p-cym), 52.8 (Im-C4,5), 30.4 (p-cym),
29.5 (CH(CH3)2), 26.3 (CH3), 23.6 (CH3), 22.3 (p-cym), 18.5
(p-cym) ppm. IR (KBr): ν 3064 (w), 2961 (s), 2928 (m), 2868 (m),
̅
1632 (m), 1593 (s), 1573 (s), 1553 (s), 1465 (m), 1444 (m), 1386
(m), 1365 (w), 1325 (m), 1276 (s), 1181 (m), 1057 (m), 1043 (m),
935 (w), 890 (m), 854 (w), 804 (m) cm−1. HR-MS (ESI+): m/z calcd
for C38H52ClN2ORuS [(M − Cl)+], 721.253 24; found, 721.252 67.
Anal. Calcd for C38H52Cl2N2ORuS (756.87): C, 60.30; H, 6.92; N,
3.70; S, 4.24. Found: C, 59.76; H, 6.82; N, 4.03; S, 4.25.
X-ray Crystal Structure Determination. Data were collected at
293 K on a Gemini diffractometer (Oxford Diffraction Ltd.) equipped
with a Ruby CCD detector using an Enhance (Mo) X-ray source: data
collection program, CrysAlis CCD (Oxford Diffraction Ltd.); data
reduction, CrysAlis RED (Oxford Diffraction Ltd.); structure solution,
SHELXS; structure refinement (on F2), SHELXL-97;39 data analysis,
PLATON.40 A multiscan procedure was applied to correct for
absorption effects. Hydrogen atom positions were calculated and
refined isotropically using a riding model.
ASSOCIATED CONTENT
* Supporting Information
■
S
Crystal Data for [RuCl2(p-cymene)(SOC·IMes)] (3): orange-red
crystals (obtained by slow diffusion of n-pentane into a CH2Cl2 solution
at −18 °C), a = 12.205(4) Å, b = 16.708(4) Å, c = 17.690(5) Å, β =
119.22(2)°, V = 3148.3(16) Å3, monoclinic, P21/c, Z = 4, Mo Kα, μ =
Figures giving an ORTEP representation of [RuCl2(p-cymene)-
(SOC·IMes)] (3) with selected bond lengths and angles and
kinetic plots of all the catalytic runs for enol ester synthesis and
CIF files giving crystallographic data for complexes 3 and 5.
This material is available free of charge via the Internet at
http://pubs.acs.org.
0.761 mm−1, ρ = 1.415 g cm−3, F000 = 1384, 10 455 reflections (Rint
=
0.0731), of which 5100 were unique and 4347 observed (I > 2σ(I)), R1
(all reflections) = 0.1649, R1 (observed reflections) = 0.0516, wR2
(observed reflections) = 0.0938, S = 0.756, residual density (ΔF) 1.166
and −0.626 e Å−3.
AUTHOR INFORMATION
Corresponding Author
*E-mail: l.delaude@ulg.ac.be.
■
Crystal Data for [RuCl2(p-cymene)(SOC·SIMes)] (5): orange-red
crystals (obtained from CH2Cl2 at −18 °C), a = 12.098(2) Å, b =
17.026(4) Å, c = 17.536(5) Å, β = 119.97(2)°, V = 3129.0(8) Å3,
monoclinic, P21/c, Z = 4, Mo Kα, μ = 0.766 mm−1, ρ = 1.428 g cm−3,
F000 = 1392, 12 423 reflections (Rint = 0.1166), of which 5515 were
unique and 3731 observed (I > 2σ(I)), R1 (all reflections) = 0.1211,
R1 (observed reflections) = 0.0786, wR2 (observed reflections) =
0.2117, S = 1.125, residual density (ΔF) 2.271 and −1.956 e Å−3.
ROMP of Cyclooctene. A 25 mL round-bottom flask equipped
with a magnetic stirring bar and capped with a three-way stopcock was
charged with a ruthenium−arene complex (0.03 mmol). Air was
expelled by applying three vacuum−argon cycles before dry
chlorobenzene (5 mL) and cyclooctene (1 mL, 7.5 mmol) were
added with dried syringes under argon. The reaction mixture was
stirred for 2 h in an oil bath at 60 °C. It was irradiated with a 40 W
“cold white” fluorescent tube placed 10 cm away from the Pyrex
reaction flask. The conversion was monitored by gas chromatography
using the cyclooctane impurity of cyclooctene as internal standard.
The resulting gel was diluted with chloroform (20 mL) and slowly
poured into methanol (300 mL) with vigorous stirring. The
precipitated polyoctenamer was filtered with suction, dried overnight
under dynamic vacuum, and characterized by size-exclusion
chromatography and NMR.
ACKNOWLEDGMENTS
■
The financial support of the “Fonds de la Recherche
Scientifique-FNRS”, Brussels, Belgium, is gratefully acknowle-
dged for the purchase of major instrumentation. We thank
Dr. Nicolas Smargiasso for the HR-MS analyses, Mrs. Bernadette
Norberg for the XRD analyses, and Ms. Nelly Tshibalonza
Ntumba and Mr. Dario Bicchielli for their help with the
catalytic tests.
REFERENCES
■
(1) (a) Arduengo, A. J. III; Harlow, R. L.; Kline, M. J. Am. Chem. Soc.
1991, 113, 361−363. (b) Arduengo, A. J. III Acc. Chem. Res. 1999, 32,
913−921.
(2) For monographs, see: (a) N-Heterocyclic Carbenes in Synthesis;
Nolan, S. P., Ed.; Wiley-VCH: Weinheim, Germany, 2006. (b) N-Hetero-
cyclic Carbenes in Transition Metal Catalysis; Glorius, F., Ed.;
Springer: Berlin, Germany, 2007; Topics in Organometallic Chemistry
Vol. 21. (c) N-Heterocyclic Carbenes: From Laboratory Curiosities to
ATRP of MMA. A 25 mL glass tube containing a magnetic stirring
bar and capped with a three-way stopcock was charged with a
ruthenium−arene complex (0.0117 mmol). Air was expelled by
applying three vacuum−nitrogen cycles before methyl methacrylate
(1 mL, 9.35 mmol) and ethyl 2-bromo-2-methylpropanoate (0.1 M in
toluene, 0.25 mL) were added with dried syringes under nitrogen. The
́
́
Efficient Synthetic Tools; Diez-Gonzalez, S., Ed.; Royal Society of
Chemistry: Cambridge, U.K., 2010; RSC Catalysis Series Vol. 6. (d) N-
Heterocyclic Carbenes in Transition Metal Catalysis and Organocatalysis; Cazin,
C. S. J., Ed.; Springer: Dordrecht, The Netherlands, 2011; Catalysis by
Metal Complexes Vol. 32.
6141
dx.doi.org/10.1021/om2006529|Organometallics 2011, 30, 6133−6142

Organometallics
Article
(3) Delaude, L. Eur. J. Inorg. Chem. 2009, 1681−1699.
(4) (a) Kuhn, N.; Steimann, M.; Weyers, G. Z. Naturforsch., B 1999,
54, 427−433. (b) Duong, H. A.; Tekavec, T. N.; Arif, A. M.; Louie, J.
Chem. Commun. 2004, 112−113. (c) Van Ausdall, B. R.; Glass, J. L.;
Wiggins, K. M.; Aarif, A. M.; Louie, J. J. Org. Chem. 2009, 74, 7935−
7942.
(5) (a) Voutchkova, A. M.; Appelhans, L. N.; Chianese, A. R.;
Crabtree, R. H. J. Am. Chem. Soc. 2005, 127, 17624−17625.
(b) Voutchkova, A. M.; Feliz, M.; Clot, E.; Eisenstein, O.; Crabtree,
R. H. J. Am. Chem. Soc. 2007, 129, 12834−12846.
(6) (a) Duong, H. A.; Cross, M. J.; Louie, J. Org. Lett. 2004, 6,
4679−4681. (b) Bantu, B.; Pawar, G. M.; Decker, U.; Wurst, K.;
Schmidt, A. M.; Buchmeiser, M. R. Chem. Eur. J. 2009, 15, 3103−3109.
(c) Naik, P. U.; Petitjean, L.; Refes, K.; Picquet, M.; Plasseraud, L. Adv.
Synth. Catal. 2009, 351, 1753−1756. (d) Riduan, S. N.; Zhang, Y.;
Ying, J. Y. Angew. Chem., Int. Ed. 2009, 48, 3322−3325. (e) Kayaki, Y.;
Yamamoto, M.; Ikariya, T. Angew. Chem., Int. Ed. 2009, 48, 4194−
4197.
(24) (a) Le Bozec, H.; Touchard, D.; Dixneuf, P. H. Adv. Organomet.
Chem. 1989, 29, 163−247. (b) Bennett, M. A. Coord. Chem. Rev. 1997,
166, 225−254. Therrien, B. Coord. Chem. Rev. 2009, 253, 493−519.
(25) Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen,
A. G.; Taylor, R. J. Chem. Soc. Perkin Trans 2 1987, S1−S19.
(26) (a) Martin, N. H.; Allen, N. W. III; Moore, K. D.; Vo, L. J. Mol.
Struct. (THEOCHEM) 1998, 454, 161−166. (b) Martin, N. H.; Allen,
N. W. III; Moore, J. C. J. Mol. Graph. Model. 2000, 18, 242−246−
300−301.
(27) (a) Slugovc, C.; Perner, B.; Stelzer, F.; Mereiter, K.
Organometallics 2004, 23, 3622−3626. (b) Ben-Asuly, A.; Tzur, E.;
Diesendruck, C. E.; Sigalov, M.; Goldberg, I.; Lemcoff, N. G.
Organometallics 2008, 27, 811−813.
(28) Lo, C.; Cariou, R.; Fischmeister, C.; Dixneuf, P. H. Adv. Synth.
Catal. 2007, 349, 546−550.
(29) (a) Sauvage, X.; Borguet, Y.; Noels, A. F.; Delaude, L.;
Demonceau, A. Adv. Synth. Catal. 2007, 349, 255−265. (b) Maj, A.;
Delaude, L.; Demonceau, A.; Noels, A. F. J. Organomet. Chem. 2007,
692, 3048−3056. (c) Sauvage, X.; Borguet, Y.; Zaragoza, G.;
Demonceau, A.; Delaude, L. Adv. Synth. Catal. 2009, 351, 441−455.
(d) Borguet, Y.; Sauvage, X.; Zaragoza, G.; Demonceau, A.; Delaude,
L. Organometallics 2011, 20, 2730−2738.
(7) Tudose, A.; Delaude, L.; Andre,
Lett. 2006, 47, 8529−8533.
́
B.; Demonceau, A. Tetrahedron
(8) Tudose, A.; Demonceau, A.; Delaude, L. J. Organomet. Chem.
2006, 691, 5356−5365.
(30) (a) Delaude, L.; Demonceau, A.; Noels, A. F. Chem. Commun
2001, 986−987. (b) Delaude, L.; Szypa, M.; Demonceau, A.; Noels, A. F.
Adv. Synth. Catal. 2002, 344, 749−756.
(9) Grotevendt, A.; Bartolome, M.; Nielsen, D. J.; Spannenberg, A.;
Jackstell, R.; Cavell, K. J.; Oro, L. A.; Beller, M. Tetrahedron Lett. 2007,
48, 9203−9207.
́
(31) Ahr, M.; Thieuleux, C.; Coperet, C.; Fenet, B.; Basset, J.-M.
(10) Delaude, L.; Sauvage, X.; Demonceau, A.; Wouters, J.
Organometallics 2009, 28, 4056−4064.
Adv. Synth. Catal. 2007, 349, 1587−1591.
(32) For a recent review on light-induced olefin metathesis, see:
Vidavsky, Y.; Lemcoff, N. G. Beilstein J. Org. Chem. 2010, 6, 1099−
1105.
(11) Sauvage, X.; Demonceau, A.; Delaude, L. Adv. Synth. Catal.
2009, 351, 2031−2038.
(12) Sauvage, X.; Zaragoza, G.; Demonceau, A.; Delaude, L. Adv.
Synth. Catal. 2010, 352, 1934−1948.
(33) Delaude, L.; Demonceau, A.; Noels, A. F. Curr. Org. Chem. 2006,
10, 203−215.
(13) (a) Sheldrick, W. S.; Schonberg, A.; Singer, E.; Eckert, P. Chem.
̈
(34) Delaude, L.; Delfosse, S.; Richel, A.; Demonceau, A.; Noels, A. F.
Chem. Commun 2003, 1526−1527.
Ber 1980, 113, 3605−3609. (b) Krasuski, W.; Nikolaus, D.; Regitz, M.
Liebigs Ann. Chem. 1982, 1451−1465. (c) Kuhn, N.; Bohnen, H.;
Henkel, G. Z. Naturforsch., B 1994, 49, 1473−1480. (d) Kuhn, N.;
Niquet, E.; Steimann, M.; Walker, I. Z. Naturforsch., B 1999, 54, 1181−
1187. (e) Delaude, L.; Demonceau, A.; Wouters, J. Eur. J. Inorg. Chem.
2009, 1882−1891.
(35) For selected reviews, see: (a) Bruneau, C.; Dixneuf, P. H.
Chem. Commun 1997, 507−512. (b) Alonso, F.; Beletskaya, I. P.; Yus,
M. Chem. Rev. 2004, 104, 3079−3160. (c) Dragutan, V.; Dragutan, I.;
Delaude, L.; Demonceau, A. Coord. Chem. Rev. 2007, 251, 765−794.
(36) For a discussion of the catalytic importance of H2O and
Na2CO3, see: Nicks, F.; Aznar, R.; Sainz, D.; Muller, G.; Demonceau,
A. Eur. J. Org. Chem. 2009, 5020−5027.
(14) (a) Naeem, S.; Thompson, A. L.; Delaude, L.; Wilton-Ely, J. D. E. T.
Chem. Eur. J. 2010, 16, 10971−10974. (b) Naeem, S.; Thompson, A. L.;
White, A. J. P.; Delaude, L.; Wilton-Ely, J. D. E. T. Dalton Trans. 2011, 40,
3737−3747.
(37) Nicks, F.; Libert, L.; Delaude, L.; Demonceau, A. Aust. J. Chem.
2009, 62, 227−231.
(15) Naeem, S.; Delaude, L.; White, A. J. P.; Wilton-Ely, J. D. E. T.
Inorg. Chem. 2010, 49, 1784−1793.
(38) Ledoux, N.; Allaert, B.; Verpoort, F. Eur. J. Org. Chem. 2007,
5578−5583.
(16) Willem, Q.; Nicks, F.; Sauvage, X.; Delaude, L.; Demonceau, A.
J. Organomet. Chem. 2009, 694, 4049−4055.
(39) Sheldrick, G. M. SHELXL-97, Program for the Refinement of
Crystal Structures; University of Gottingen, Gottingen, Germany, 1997.
̈
̈
(17) Hans, M.; Wouters, J.; Demonceau, A.; Delaude, L. Eur. J. Org.
Chem. 2011, (doi: 10.1002/ejoc.201101286).
(40) Spek, A. L. PLATON, A Multipurpose Crystallographic Tool;
Utrecht University: Utrecht, The Netherlands, 2005.
(18) Chia, E. Y.; Naeem, S.; Delaude, L.; White, A. J. P.; Wilton-Ely,
J. D. E. T. Dalton Trans. 2011, 40, 6645−6658.
(19) See for example: (a) Bennett, M. A.; Smith, A. K. J. Chem. Soc.,
Dalton Trans. 1974, 233−241. (b) Jan, D.; Delaude, L.; Simal, F.;
Demonceau, A.; Noels, A. F. J. Organomet. Chem. 2000, 606, 55−64.
(c) Solari, E.; Gauthier, S.; Scopelliti, R.; Severin, K. Organometallics
2009, 28, 4519−4526. (d) Borguet, Y.; Sauvage, X.; Zaragoza, G.;
Demonceau, A.; Delaude, L. Organometallics 2010, 29, 6675−6686.
̈
(20) Ozdemir, I.; Yasa̧ r, S.; Demir, S.; Çetinkaya, B. Heteroat. Chem.
2005, 16, 557−561.
(21) Savant, V. V.; Gopalakrishnan, J.; Patel, C. C. Inorg. Chem. 1970,
9, 748−751.
(22) (a) Gould, R. O.; Stephenson, T. A.; Thomson, M. A. J. Chem. Soc.,
Dalton Trans. 1978, 769−775. (b) El-Hinnawi, M. A.; Sumadi, M. L.;
Esmadi, F. T.; Jibril, I.; Imhof, W.; Huttner, G. J. Organomet. Chem. 1989,
377, 373−381. (c) Jibril, I.; Esmadi, F. T.; Al-Masri, H.; Zsolnai, L.;
Huttner, G. J. Organomet. Chem. 1996, 510, 109−116. (d) El-khateeb, M.;
Jazzazi, T. M. A.; Gorls, H.; Al-Shboul, T. M. A.; Westerhausen, M.
̈
Transition Met. Chem. 2011, 36, 29−33.
(23) Nakayama, J.; Kitahara, T.; Sugihara, Y.; Sakamoto, A.; Ishii, A.
J. Am. Chem. Soc. 2000, 122, 9120−9126.
6142
dx.doi.org/10.1021/om2006529|Organometallics 2011, 30, 6133−6142