Stability and activity of cis-dichloro ruthenium olefin metathesis precatalysts bearing chelating sulfur alkylidenes

Article abstract of DOI:10.1016/j.jorganchem.2014.06.027

S-Chelated Grubbs type complexes are usually found in their stable cis-dichloro conformation. These precatalysts are usually latent and necessitate external stimuli to promote olefin metathesis reactions. Herein we report the synthesis of new S-chelated ruthenium complexes, by a simple exchange of the benzylidene to an alkylidene chelating ligand. The structure of the complexes was studied by NMR spectroscopy, single crystal X-ray diffraction and DFT calculations. The new complexes were tested in a series of olefin metathesis reactions and were found to be the first cis-dichloro ruthenium precatalysts that are active at room temperature. The observed differences in reactivity and structure between the alkylidene complexes and their benzylidene counterparts highlight the important influence aromaticity may have on the stability and activity of chelated Grubbs type complexes. The findings may also have implications to develop alternative strategies to stabilize, or destabilize, other organometallic complexes bearing relevant chelating ligands.

Full text of DOI:10.1016/j.jorganchem.2014.06.027

Journal of Organometallic Chemistry 769 (2014) 24e28
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Stability and activity of cis-dichloro ruthenium olefin metathesis
precatalysts bearing chelating sulfur alkylidenes
,
Eyal Tzur a *, Elisa Ivry b, Charles E. Diesendruck b, Yuval Vidavsky b, Israel Goldberg c,
N. Gabriel Lemcoff b
a Department of Chemical Engineering, Shamoon College of Engineering, Ashdod 77245, Israel
b Chemistry Department, Ben-Gurion University of the Negev, Beer-Sheva 84105, Israel
c School of Chemistry, Tel-Aviv University, Tel-Aviv 69978, Israel
a r t i c l e i n f o
a b s t r a c t
Article history:
S-Chelated Grubbs type complexes are usually found in their stable cis-dichloro conformation. These
precatalysts are usually latent and necessitate external stimuli to promote olefin metathesis reactions.
Herein we report the synthesis of new S-chelated ruthenium complexes, by a simple exchange of the
benzylidene to an alkylidene chelating ligand. The structure of the complexes was studied by NMR
spectroscopy, single crystal X-ray diffraction and DFT calculations. The new complexes were tested in a
series of olefin metathesis reactions and were found to be the first cis-dichloro ruthenium precatalysts
that are active at room temperature. The observed differences in reactivity and structure between the
alkylidene complexes and their benzylidene counterparts highlight the important influence aromaticity
may have on the stability and activity of chelated Grubbs type complexes. The findings may also have
implications to develop alternative strategies to stabilize, or destabilize, other organometallic complexes
bearing relevant chelating ligands.
Received 23 March 2014
Received in revised form
16 June 2014
Accepted 25 June 2014
Available online 16 July 2014
Keywords:
Ruthenium complexes
cis-Dichloro configuration
Latent catalysts
Olefin metathesis
Sulfur chelation
© 2014 Elsevier B.V. All rights reserved.
Metallacycle aromaticity
Introduction
oxygen [9], halogen [10] and nitrogen [11] chelated complexes,
which are active in room temperature have been reported, to date,
Olefin metathesis [1] has greatly influenced many areas of
organic synthesis, among them, the production of polymers [1f,2],
natural products [1f,3], drugs [4] and chemical commodities from
renewable resources [5]. Much of this progress has been due to the
constant development of more reactive and stable catalysts, a very
active field of research that has certainly afforded impressive
achievements [1e,g,6]. For example, in applications related to the
preparation of thermosets, intimate control over the reaction
initiation and kinetics is essential in order to obtain homogenous
materials featuring the desired mechanical properties [7].With this
in mind, we recently developed a series of sulfur chelated ruthe-
nium olefin metathesis precatalysts which display dormant
behavior [8]. The sulfureRu bond is remarkably strong, thus
allowing for exquisite control over precatalyst initiation [8b,c].
Unexpectedly, these complexes disclosed a cis-dichloro configura-
tion, opposed to the trans-dichloro conformation typically observed
for commercial catalysts (Fig. 1). Even though different cis-dihalo
all sulfur chelated precatalysts were found to be latent in several
olefin metathesis applications at room temperature and could be
activated by the use of heat [8aec,e] or light [8d].
In order to control the initiation and performance of S-chelated
precatalysts both steric and electronic effects may be harnessed. For
example, the addition of a bulky group made compounds 2 and 3c
to be the most active within their series (Fig. 2) [8b,d]. These studies
led to the conclusion that only the trans-dichloro isomer is active,
while the cis-dichloro isomers are latent. In addition, electronic
effects were also used in order to tune the SeRu bond dissociation
and affect catalyst efficiency. Thus, addition of a strongly electron
withdrawing trifluoromethyl group to the sulfur atom formed cis-
dichloro complex 4 (Fig. 2), which turned out to be the fastest S-
chelated catalyst after isomerization (either by light or heat);
competing in reaction rates with the popular O-chelated catalysts
[8e]. Interestingly, when using a chelating sulfoxide instead of a
thioether, the trans-dichloro complex was obtained as the most
stable isomer in CH2Cl2 solution (5, Fig. 2), which was expectedly
active towards ring-closing metathesis (RCM) and enyne metath-
esis reactions at ambient temperatures [12].
* Corresponding author. Tel.: þ972 546813479.
E-mail address: eyalt@sce.ac.il (E. Tzur).
http://dx.doi.org/10.1016/j.jorganchem.2014.06.027
0022-328X/© 2014 Elsevier B.V. All rights reserved.

E. Tzur et al. / Journal of Organometallic Chemistry 769 (2014) 24e28
25
Scheme 1. Synthesis of thioethers 7a and 7b.
the desired products 7 in quantitative yields after isolation.
Consecutively, alkylidene complexes 8a and 8b were obtained by
treatment of Grubbs 3rd generation catalyst (Gru3) [16] with 7a
and 7b respectively in CH2Cl2 (Scheme 2).
Structural characterization of the prepared complexes
Fig. 1. cis-Dichloro ruthenium sulfur complexes 1.
Complexes 8a and 8b were fully characterized by NMR spec-
troscopy and HRMS (ESI) (see Supporting information) analyses.
The solid-state structure of complex 8a was also studied using
single crystal X-ray diffraction (Fig. 3).
Recently, Grela and coworkers studied the effect of the aroma-
ticity of the 5-membered metallacycle ring in several oxygen-
chelated naphthalene-based analogs of Hoveyda type catalysts
[13]. Their results indicated that a stronger metallacyclic aromatic
character increased chelate stability and inhibited precatalyst
initiation. Therefore, in an attempt to further explore the role of
aromaticity in chelated ruthenium catalysts, we decided to reduce
metallacycle aromaticity by the simple approach of using an alky-
lidene instead of a benzylidene ligand. A literature survey reveals
that alkylidene chelated complexes have been prepared in the past;
for example, the 6-membered nitrogen chelated alkylidene com-
plexes studied by Van der Schaaf and Grubbs [14], but the oxygen
chelated alkylidene (5 and 6-membered) analogs were not stable
enough to be isolated [15]. We expected that the stronger sul-
fureruthenium bond would still allow the formation of this binding
form, albeit the reduced stability due to lack of aromaticity.
The NMR spectra obtained were analyzed to understand the
structure of ruthenium complexes 8a and 8b in solution. As typi-
cally observed for the cis-dichloro configuration, every proton and
methyl group in the molecule was clearly resolved. When
comparing the spectra of 8a and 8b to their benzylidene analogs 1d
and 1e [8b,d], a significant difference in the chemical shift of the
corresponding aromatic mesitylene protons can be observed. The
most upfield mesitylene proton appears at 6.06 ppm for both 1d
and 1e compared to 6.70 ppm (for 8a) and 6.65 ppm (for 8b). This
change is associated with a diminished anisotropic effect in 8a and
8b due to the absence of the aromatic moiety in the chelated sulfur
ligand. Moreover, a minor deshielding of the carbene proton can be
observed in 8a and 8b. The proton appears at lower field (17.45 and
17.51 ppm) when compared to 1d and 1e (17.21 and 17.14 ppm)
respectively.
Results and discussion
Table 1 presents a comparison between selected bond lengths
and angles of 8a and its analog 1d [8b]. Notably, despite the fact
that complex 8a is active at room temperature (vide infra), it ex-
hibits a shorter sulfureruthenium bond compared to latent 1d. This
reduced bond length can be explained by the lack of aromaticity in
the chelated ring e double bonds in aromatic rings being longer
than standard double bonds [17]. Another important deviation in
the structure between 1d and 8a can be found in the conformation
of the five membered metallacycle. While in 1d all the atoms
comprising the cycle are found in the same plane, 8a shows a
Synthesis of ruthenium complexes
The synthesis of thioethers 7a and 7b are described in Scheme 1.
4-Bromobutene was added to a mixture of thiol 6, potassium car-
bonate and 18-crown-6 in acetone and refluxed overnight to afford
Fig. 2. Steric and electronic effects in sulfur chelated ruthenium complexes.
Scheme 2. Synthesis of precatalysts 8a and 8b.

26
E. Tzur et al. / Journal of Organometallic Chemistry 769 (2014) 24e28
DFT calculations
In order to understand if the orientation of the S-substituent is a
crystal packing effect or an electronic consequence of the non-
aromaticity of the metallacycle, complexes 1a, 1d, 8a and 8b were
studied by DFT calculations. Minimization and energy calculations
were carried out using the same method previously developed by
us [18], which provided good predictions for configurations of
several chelated Ru complexes. The structures were optimized us-
ing gradient corrected (BP86) DFT calculations [19], at the LACVP*
level of theory [20]. The energies were then determined using
LACV3P**þþ for higher level calculations with additional solvation
energy, which was calculated using the PoissoneBoltzmann (PBF)
Fig. 3. X-ray structure of complex 8a (ellipsoids drawn at 50% probability, hydrogen
atoms and solvent molecules omitted for clarity).
continuum approximation, with CH2Cl2 as
a solvent [21].
Mimicking the experimental results, in the cis-dichloro configura-
tion, the optimized structures for complexes 8a and 8b presented
the phenyl/naphthyl rings parallel to the mesitylene ring, while 1d
and 1e always converged to a structure with a perpendicular aro-
matic system. This calculation supports the observations from the
Table 1
Solid state structure comparison of 1d and 8a.
1dcis
8acis
solid state study, suggesting that the
p-system of the aromatic
Bond
C1eRu1
S1eRu1
Cl1eRu1
Cl2eRu1
S1eC25
S1eC24
2.024
2.030
2.334
2.396
2.397
1.793
1.821
1.517
1.500
1.823
substituent prefers to be perpendicular to an aromatic metallacycle,
yet it may interact with the sulfur atom's available lone-pair when
the metallacycle is not stabilized by aromaticity.
2.347
2.360
2.394
1.788
1.797
1.405
1.447
1.844
Reactivity studies
C
C
C
24eC23
23eC22
22eRu1
To explore the reactivity profiles of the precatalysts, various
olefin substrates were allowed to react in the presence of com-
plexes 8a and 8b. Table 2 summarizes the results for benchmark
RCM, cross-metathesis (CM) and ring-opening metathesis poly-
merization (ROMP) reactions. Surprisingly, good conversions were
obtained for simple RCM reactions after only 4 h at room temper-
ature (without UV irradiation). However, 8a and 8b were not effi-
cient at catalyzing reactions with more challenging tetra-
substituted products (entry 5, Table 2).
Angle
Cl1eRu1eCl2
Ru1eS1eC25
88.21
114.59
100.11
93.88
88.97
120.58
101.98
90.75
C
24eS1eC25
C1eRu1eS1
Torsion angle
Ru1eS1eC25eC26
Ru1eS1eC25eC30
118.22
ꢁ59.51
21.45
ꢁ165.63
Even though possessing the unique cis-dichloro configuration,
room temperature activity was observed for 8a and 8b, in stark
contradiction to their benzylidene analogs 1d and 1e [8b,d]. This
behavior suggests that the Ruesulfur bond in precatalysts 8 can
readily dissociate at room temperature (at r.t. only the cis isomer is
observed in NMR spectroscopy. For a preliminary study on the cis to
trans isomerization of precatalysts 8 under UV irradiation, see the
Supporting information). This phenomenon may seem counterin-
tuitive at first glance, because it conflicts with the simplistic view
that the non-bonding electrons of sulfur would be more available
for coordination as they do not participate in the resonance of the
benzylidene. On the other hand, two effects dictate a more stable
typical puckered half-chair conformation (Fig. 4). This difference
reduces the distance between the sulfur atom in 8a to the mesi-
tylene substituent on the NHC (CNHCeRueS angle is 90.75ꢀ)
compared to the structure observed for 1d (CNHCeRueS angle is
93.88ꢀ). Additionally, a significant structural difference between 1d
and 8a is the orientation of the aromatic phenyl ring bonded to the
sulfur atom (Fig. 4). Whereas the phenyl ring bound to the sulfur
atom in 1d is perpendicular to the mesitylene, the ring in 8a is
parallel to the mesitylene. Preliminary DFT calculations were car-
ried out to further our understanding of the geometrical arrange-
ments in these molecules.
Fig. 4. Comparison of solid state structures of 1d [8b] and 8a. The C1eRu1eS1 angles are presented. The different orientation of the phenyl ring is circled.

E. Tzur et al. / Journal of Organometallic Chemistry 769 (2014) 24e28
27
Table 2
Formation of RCM, CM and ROMP products promoted by 8a and 8b.
Entry
Substrate
Product
Cat.
Conv. [%]
1
8a
8b
75(33)a
86(36)a
2
3
4
5
6
7
8a
8b
81b
95b
8a
8b
76(26)a
97(34)a
8a
8b
66a
68a
8a
8b
0a
0a
8a
8b
4b
35b,c
PolyCOE
8a
8b
57b,d,e
77b,d,e
Reactions were conducted inside a nitrogen glove box. [Substrate] ¼ 0.1 M in CD2Cl2,
[catalyst] ¼ 5 mol%, 4 h, T ¼ 25 ꢀC. In brackets are yields under air.
a
Determined by 1H NMR.
Determined by GCeMS.
Additional products, such as (E) and (Z)-prop-1-ene-1,3-diyldibenzene, were
b
c
observed due to isomerization of the substrate's double bond.
d
With mesitylene as internal standard.
Oligomers were observed.
e
chelating effect for compounds 1. First, in accordance with Grela's
finding [13], the aromaticity of the ruthenacycle in 1 can actually
strengthen the chelate bond and inhibit precatalyst initiation.
Secondly, as shown by our DFT calculations in the case of the
alkylidenes 8a and 8b, the sulfur atom lone-pair electrons are
strongly conjugated with the exocyclic aromatic ring, making the
RueS bond more labile and ultimately leading to a significant in-
crease in precatalyst initiation. As a consequence of this, also the
stability of these new complexes is reduced, as clearly observed by
comparing reaction conversion under air versus nitrogen (entries 1
and 3, Table 2). Precatalyst 8b, with the naphthalene moiety,
exhibited even better activity compared to 8a. A possible expla-
nation could be weakening of the RueS bond by a greater steric
interference of the naphthalene moiety and an increased partici-
Fig. 5. Energy profiles of RueS bond dissociation.
LACV3P**þþ/BP86 were calculated. As shown in Fig. 5, the aro-
matic complex 1d is calculated to require 146.1 kJ/mol for dissoci-
ation, compared to 96.2 kJ/mol for the non-aromatic complex 8a.
This result indicates that in 8a, dissociation of the RueS bond (i.e.
the initiation step for catalyst activation or isomerization to the
trans isomer [8d]), occurs at significantly lower energies, support-
ing the experimental observation of room temperature activities
with 8a and 8b as catalysts. The calculations also indicate that 8a is
15.2 kJ/mol more stable than 8atrans indicating that even though
they are in equilibrium, the amounts of the trans isomer in solution
is very small (for comparison, 1d is 30.3 kJ/mol more stable than
1dtrans).
pation of the sulfur atom lone-pair electrons on the naphthyl
electron system.
p-
Once again we turned to DFT energy calculations to test the
hypothesis of increased activity at room temperature for 8. In
addition to the cis structures of 1d and 8a described above, the
dissociation transition state and dissociated structures, as well as
the trans isomers, were studied at the LACVP*/BP86 level of theory.
For more precise energy results, single point energy at the

28
E. Tzur et al. / Journal of Organometallic Chemistry 769 (2014) 24e28
(c) A. Fürstner, C. Nevado, M. Waser, M. Tremblay, C. Chevrier, F. Teplý,
Conclusions
C. Aïssa, E. Moulin, O. Müller, J. Am. Chem. Soc. 129 (2007) 9150.
[4] (a) J.W. Morzycki, Steroids 76 (10e11) (2011) 949;
We have synthesized model chelating sulfur alkylidene ruthe-
(b) P. Van de Weghe, J. Eustache, Curr. Top. Med. Chem. 5 (15) (2005) 1495.
[5] (a) B.B. Marvey, Int. J. Mol. Sci. 9 (8) (2008) 1393;
nium complexes 8a and 8b bearing a cis-dichloro configuration. In
contrast with their benzylidene analogs, 1a and 1d, these com-
plexes were active at room temperature. This is the first example of
cis-dichloro sulfur chelated ruthenium precatalysts that exhibit
reasonable metathesis activity at room temperature (without UV
irradiation). DFT studies point to a lower transition state for the
RueS dissociation, which allows for room temperature isomeriza-
tion to the active trans form or even direct activation. On the other
hand, the stability of the new complexes was hampered by the
removal of aromatic stabilization of the chelating ring. The
observed differences between the alkylidene complexes 8a and 8b
and their benzylidene counterparts brings to light the important
influence aromaticity may have on the stability and activity of
chelated Grubbs type complexes. The findings may also have im-
plications to develop strategies to stabilize, or destabilize, relevant
organometallic complexes bearing chelating ligands.
(b) J.C. Mol, Green Chem. 4 (1) (2002) 5.
[6] (a) C.E. Diesendruck, E. Tzur, N.G. Lemcoff, Eur. J. Inorg. Chem. 28 (2009) 4185;
(b) A. Szadkowska, K. Grela, Curr. Org. Chem. 12 (18) (2008) 1631.
[7] (a) S. Monsaert, A. Lozano Vila, R. Drozdzak, P. Van Der Voort, F. Verpoort,
Chem. Soc. Rev. 38 (12) (2009) 3360;
(b) R.C. Larock, P.H. Henna, M.R. Kessier, U.S. 2012, US 8318876 B1 20121127.
(c) C.E. Diesendruck, O. Iliashevsky, A. Ben-Asuly, I. Goldberg, N.G. Lemcoff,
Macromol. Symp. 293 (2010) 33.
[8] (a) A. Ben-Asuly, E. Tzur, C.E. Diesendruck, M. Sigalov, I. Goldberg,
N.G. Lemcoff, Organometallics 27 (5) (2008) 811;
(b) T. Kost, M. Sigalov, I. Goldberg, A. Ben-Asuly, N.G. Lemcoff, J. Organomet.
Chem. 693 (12) (2008) 2200;
(c) C.E. Diesendruck, Y. Vidavsky, A. Ben-Asuly, N.G. Lemcoff, J. Polym. Sci. Part
A Polym. Chem. 47 (2009) 4209;
(d) A. Ben-Asuly, A. Aharoni, C.E. Diesendruck, Y. Vidavsky, I. Goldberg,
B.F. Straub, N.G. Lemcoff, Organometallics 28 (16) (2009) 4652;
(e) Y. Ginzburg, A. Anaby, Y. Vidavsky, C.E. Diesendruck, A. Ben-Asuly,
I. Goldberg, N.G. Lemcoff, Organometallics 30 (12) (2011) 3430.
[9] (a) E. Pump, R.C. Fischer, C. Slugovc, Organometallics 31 (2012) 6972;
(b) E. Pump, A. Poater, M. Zirngast, A. Torvisco, R. Fischer, L. Cavallo, C. Slugovc,
Organometallics 33 (2014) 2806.
[10] (a) M. Barbasiewicz, M. Michalak, K. Grela, Chem. Eur. J. 18 (2012) 14237;
Acknowledgments
_
ꢀ
ꢀ
ꢀ
(b) K. Grudzien, K. Zukowska, M. Malinska, K. Wozniak, M. Barbasiewicz,
Chem. Eur. J. 20 (2014) 2819.
[11] M. Barbasiewicz, A. Szadkowska, R. Bujok, K. Grela, Organometallics 25 (2006)
ET acknowledges Shamoon College of Engineering for financial
support. The ISF is also acknowledged for partial funding of this
research.
3599.
[12] A. Szadkowska, A. Makal, K. Wozniak, R. Kadyrov, K. Grela, Organometallics 28
(2009) 2693.
ꢀ
[13] M. Barbasiewicz, A. Szadkowska, A. Makal, K. Jarzembska, K. Wozniak,
K. Grela, Chem. Eur. J. 14 (2008) 9330.
[14] (a) P.A. Van der Schaaf, R. Kolly, H.J. Kirner, F. Rime, A. Muhlebach, A. Hafner,
J. Organomet. Chem. 606 (2000) 65;
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jorganchem.2014.06.027.
(b) T. Ung, A. Hejl, R.H. Grubbs, Y. Schrodi, Organometallics 23 (2004) 5399;
(c) A. Hejl, M.W. Day, R.H. Grubbs, Organometallics 25 (2006) 6149.
[15] (a) I. Czelusniak, J.D. Heywood, A.M. Kenwright, E. Khosravi, J. Mol. Catal. A
Chem. 280 (2008) 29;
References
(b) C. Slugovc, S. Demel, S. Riegler, J. Hobisch, F. Stelzer, Macromol. Rapid
Commun. 25 (2004) 475.
[16] J.A. Love, J.P. Morgan, T.M. Trnka, R.H. Grubbs, Angew. Chem. Int. Ed. 41 (21)
(2002) 4035.
[1] (a) R.H. Grubbs (Ed.), Handbook of Metathesis, Wiley-VCH, Weinheim, Ger-
many, 2003;
[17] 0.139 nm compared to 0.134 nm. See: K.P.C. Vollhardt, N.E. Schore, Organic
Chemistry: Structure and Function, fifth ed., WHF Publication, 2005, p. 670.
[18] C.E. Diesendruck, E. Tzur, A. Ben-Asuly, I. Goldberg, B.F. Straub, N.G. Lemcoff,
Inorg. Chem. 48 (22) (2009) 10819.
(b) R.R. Schrock, Angew. Chem. Int. Ed. 45 (2006) 3748;
(c) R.H. Grubbs, Angew. Chem. Int. Ed. 45 (2006) 3760;
(d) R.H. Grubbs, Tetrahedron 60 (2004) 7117;
(e) G.C. Vougioukalakis, R.H. Grubbs, Chem. Rev. 110 (2010) 1746;
(f) A. Fürstner, Angew. Chem. Int. Ed. 39 (2000) 3012;
(g) C. Samojowicz, M. Bieniek, K. Grela, Chem. Rev. 109 (2009) 3708.
[19] (a) A.D. Becke, Phys. Rev. A 38 (1988) 3098;
(b) J.P. Perdew, Phys. Rev. B 33 (1986) 8822.
€
[20] (a) P.J. Hay, W.R. Wadt, J. Chem. Phys. 82 (1985) 299;
(b) P.C. Hariharan, J.A. Pople, Chem. Phys. Lett. 16 (1972) 217.
[21] ε ¼ 8.93, solvent radius ¼ 2.33 Å.
[2] (a) U.H.F. Bunz, D. Maker, M. Porz, Macromol. Rapid Commun. 33 (2011) 886;
(b) J.C. Mol, J. Mol. Catal. A Chem. 213 (1) (2004) 39.
[3] (a) K.C. Nicolaou, P.G. Bulger, D. Sarlah, Angew. Chem. Int. Ed. 44 (2005) 4490;
(b) A. Fürstner, Chem. Commun. 47 (2011) 6505;