Homodinuclear ruthenium catalysts for dimer ring-closing metathesis

Article abstract of DOI:10.1002/anie.200801626

(Chemical Equation Presented) Two ring or not to ring: Novel diruthenium olefin metathesis catalysts show a tendency to avoid oligomerization and favor cyclic dimerization when the distances between the ruthenium centers and between the diene extremities match (see scheme).

Full text of DOI:10.1002/anie.200801626

Communications
DOI: 10.1002/anie.200801626
Ring-Closing Metathesis
Homodinuclear Ruthenium Catalysts for Dimer Ring-Closing
Metathesis**
Eyal Tzur, Amos Ben-Asuly, Charles E. Diesendruck, Israel Goldberg, and N. Gabriel Lemcoff*
The development of practical olefin metathesis catalysts has
led to significant advances in several areas of chemistry
during the last decade.[1] For example, the olefin metathesis
reaction has found utility in areas ranging from natural
product synthesis[2] to polymer[3] and supramolecular chemis-
try.[4] Olefin metathesis is also now widely used in the
pharmaceutical industry[5] and in materials sciences,[6]
among other areas. The exceptional reactivity and func-
tional-group tolerance of well-defined catalysts, such as the
ruthenium catalysts 1–3 (Figure 1), have resulted in a plethora
ization (ROMP), and cross metathesis or acyclic diene
metathesis (CM and ADMET). Whereas ROMP utilizes
strained cyclic alkenes as a feedstock to produce polymers,
RCM and ADMET reactions make use of dienes as starting
materials. The product ratio between the intramolecular
(RCM) and the intermolecular (ADMET) reaction is depen-
dent on both concentration and effective molarity parame-
ters. Thus, if the concentration of the diene is equal to the
effective molarity then, by definition, the rate of the
intermolecular reaction will equal the rate of the intra-
molecular reaction. This means that cyclizations are preferred
at low concentrations and that the use of dienes with very low
effective molarities will prevent ring closure.[10]
Cyclic dimers are appealing synthetic targets. For exam-
ple, steroid cyclodimers have been studied as model biological
systems and for their molecular recognition in enzymatic
processes.[11a] In addition, many cyclodimers display remark-
able properties in supramolecular and self-assembly sys-
tems.[11] We hypothesized that the development of a double
centered olefin metathesis catalyst could combine the power
of effective molarity and olefin metathesis to produce an
alternative mode for the olefin metathesis reaction—a
directed type of metathesis that would prefer to produce
cyclodimers rather than oligomers or cyclic monomers.
Scheme 1 highlights the main intermediates involved in
the process of generating dimeric rings from dienes in a
process we have called “dimer ring-closing metathesis”
Figure 1. Grubbs first- (1) and second-generation (2) catalysts and the
second-generation Hoveyda–Grubbs catalyst (3).
of novel synthetic pathways and strategies that were pre-
viously beyond the reach of organic chemists. Although not
effortless by any means, the development of modified
ruthenium olefin metathesis catalysts is fairly common and
variations on the original Grubbs catalyst are many and
varied:[7] the introduction of Arduengo-type N-heterocyclic
carbene ligands[8] led to improved stability and increased
reactivity, for example, while exchanging the phosphine
ligand for an isopropoxy ligand (bound directly to the
aromatic benzylidene) made the catalysts even more robust.[9]
There are three main modes for olefin metathesis: ring-
closing metathesis (RCM), ring-opening metathesis polymer-
[*] E. Tzur, Dr. A. Ben-Asuly, C. E. Diesendruck, Dr. N. G. Lemcoff
Chemistry Department, Ben-Gurion University of the Negev
Beer-Sheva 84105 (Israel)
Fax: (+972)8647-1740
E-mail: lemcoff@bgu.ac.il
Scheme 1. Possible DRCM mechanism (first ruthenacyclobutane inter-
mediates not shown for clarity).
Dr. A. Ben-Asuly
Achva Academic College, Shikmim (Israel)
Prof. I. Goldberg
School of Chemistry, Tel-Aviv University (Israel)
(DRCM). For this strategy to be successful, the rate-
determining step must be the disruption of the metallacyclo-
butane intermediate, in other words there must be enough
time for the second double bond to bind to the adjacent
ruthenium center before the first metallacyclobutane breaks
[**] This research was supported by a Young Scientist Grant from the
German-Israeli Foundation for Scientific Researchand Develop-
ment (GIF).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200801626.
6422
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 6422 –6425

Angewandte
Chemie
up; if this does not occur each side of the bimetallic catalyst
would carry out the reaction independently. Many theoretical
studies support the disruption of the ruthenacyclobutane as
the rate-determining step,[12] although most authors have
emphasized that simpler models may not simulate the
complicated intermediates obtained under real metathesis
conditions correctly. Nonetheless, for the reaction to be
viable, we must assure that the highest effective molarity
possible is achieved for the critical dimer ring-closing step.
We therefore set out to synthesize homo-bimetallic
ruthenium-type catalysts to test our proposal and study
their properties. The first step in the production of these
catalysts involves the synthesis of bis-N-heterocyclic carbene
(NHC) precursors. Although a few bis-NHC compounds are
known in the literature,[13] most of them are used to ligate a
single metal atom. We are also unaware of any attempts to
obtain tailored distances between the NHC atoms by a
modular spacer approach. A recent communication has
reported the use of a dimeric linker for the production of
oligomeric self-supported catalysts.[14]
monomeric counterparts, the addition of potassium hexa-
methyldisilazide to compound 7d, followed by addition of
catalyst 1 in situ, produced the desired diruthenium catalyst 8
in fair yields (Scheme 3). Significantly, the shorter spacers in
7a and 7b did not allow the pure diruthenium catalyst to be
The bis-NHC ligand precursors 7 were readily synthe-
sized, as shown in Scheme 2. The length of the bis-aniline
spacer is critical since it determines the final distance between
the ruthenium atoms in the catalyst. The second step in this
synthesis involves deprotonation of 7 to generate bis-carbenes
and the exchange of a labile tricyclohexylphosphine ligand in
two molecules of 1. After several unsuccessful attempts to
generate the bis-carbenes by typical procedures used for their
Scheme 3. Synthesis of bis-NHC catalyst 8.
obtained by this methodology. We believe that significant
steric constraints due to the large tricyclohexylphosphine
ligand prevent the formation of the bis-metallic catalyst and
lead to mainly mono-insertion.
Catalyst 8 was characterized by NMR spectroscopy and
FAB mass spectrometry but proved to be relatively unstable
and decomposed after some hours in solution (both in CD2Cl2
and [D8]toluene), in contrast to monomeric catalysts of this
type. The detection of benzaldehyde as one of the decom-
position compounds indicates the involvement of water in this
process.[15] The addition of isopropoxystyrene and copper(I)
chloride to improve the stability of the catalyst led to the
formation of the isopropoxy chelated bis-catalyst 9, albeit in
low yields due to the instability of 8 (Scheme 4).
The attachment of the chelating isopropoxybenzylidene
moiety greatly improved the catalyst stability (no discernible
decomposition after standing in air for one week in deuter-
ated dichloromethane solution). Diruthenium catalysts 8 and
9 may appear in the NMR spectrum as a mixture of rotamers
À
due to restricted rotation around the Ru NHC carbon
bond.[16] Indeed, catalyst 8 shows three benzylidene signals
due to “out-out” and “out-in” rotamers, while catalyst 9 also
displays three benzylidene signals at low temperatures that
coalesce into a single signal when the solution is warmed to
room temperature. As observed for the monomeric Grubbs
catalyst, the introduction of isopropoxybenzylidene ligands
À
allows for faster rotation around the Ru NHC carbon bond.
We will report further details of the NMR dynamics of these,
and similar, systems in a future paper.[17]
To our great satisfaction compound 9 also afforded single
crystals that allowed its structure to be solved by X-ray
diffraction (Figure 2).[18]
To probe the preliminary metathesis activity of the
catalysts we realized typical RCM reactions with diethyl
Scheme 2. Synthesis of bis-NHC precursors.
Angew. Chem. Int. Ed. 2008, 47, 6422 –6425
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6423

Communications
Figure 3. Metathesis of 10 by catalysts 2, 3, 8, and 9. Average of two
measurements. [10]=10 mm; catalyst loading: 5 mol%. DRCM con-
version is calculated as the sum of cyclic dimers.[19] ADMET conversion
is calculated by difference.
Scheme 4. Synthesis of isopropoxy chelated diruthenium catalyst 9.
Furthermore, the use of diruthenium catalysts prevented the
formation of ADMET products almost completely. Thus, a
larger ratio of cyclic dimer to oligomer product is observed
when the distance between the sp2 centers is appropriate (the
distance between the two Ru centers in 9 in the solid state is
10.2 Š).[20]
To support the assumption that both ruthenium atoms are
metathetically active simultaneously, we attached diolefin 14
to catalyst 8. Compound 14 was synthesized, and compound
15 was isolated after a typical benzylidene exchange proce-
dure involving 8 (Scheme 6) and characterized by NMR
spectroscopy and FAB mass spectrometry. The proposed
structure for 15 confirms our hypothesis that both ruthenium
centers can participate in the DRCM reaction.
Figure 2. X-ray crystal structure of catalyst 9.
diallylmalonate and observed that 8 and 9 show similar
reactivity to the commercial Grubbs catalysts 1–3. We then
decided to test the metathesis reaction with 1,12-tridecadiene
(10) to see whether our bis-catalyst could “catch” both ends of
the diene (Scheme 5). Two main factors influenced the choice
In summary, we have demonstrated that by spacing two
reactive catalytic centers at the correct distance, effective
molarities may help dictate the course of the reaction and
Scheme 5. Metathesis reactions of 10.
of a 13-membered diene as the preferred substrate to test the
DRCM protocol: first, molecular modeling predicted this
diene to have the right length to attach to both metal centers,
and second, the formation of an unfavorable 11-membered
ring should hinder ring-closing metathesis. The reaction
progress was followed by GC-MS (using decaline as internal
standard); the results are summarized in Figure 3.
As predicted, in this case the diruthenium catalysts 8 and 9
showed a propensity to preferentially afford DRCM prod-
ucts[19] in comparison to the monomeric catalysts 2 and 3.
Scheme 6. Synthesis of compound 15.
6424 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 6422 –6425

Angewandte
Chemie
see: V. G. H. Lafitte, A. E. Aliev, P. N. Horton, M. B. Hurst-
produce specific compounds. By using the DRCM protocol,
we have successfully inhibited oligomerization and promoted
cyclodimerization when the diene starting materials are of the
appropriate length. We are currently continuing our work on
using these bis-NHC spacers with other metals to try to
promote diverse dimer ring-closing reactions, as well as
changing the length and rigidity of the spacer to improve the
selectivity and scope of DRCM reactions.
house, H. C. Hailes, Chem. Commun. 2006, 2173 – 2175; d) K.
Tashiro, T. Aida, J.-Y. Zheng, K. Kinbara, K. Saigo, S. Sakamoto,
K. Yamaguchi, J. Am. Chem. Soc. 1999, 121, 9477 – 9478;
e) M. R. Ghadiri, J. R. Granja, L. K. Buehler, Nature 1994,
369, 301 – 304; f) P.-H. Huang, J. T. Lin, M.-C. P. Yeh, J. Organo-
met. Chem. 2006, 691, 975 – 982.
[12] a) M. Jordaan, P. van Helden, C. G. C. E. van Sittert, H. C. M.
Vosloo, J. Mol. Catal. A 2006, 254, 145 – 154; b) B. F. Straub,
Angew. Chem. 2005, 117, 6129 – 6132; Angew. Chem. Int. Ed.
2005, 44, 5974 – 5978; c) A. C. Tsipis, A. G. Orpen, J. N. Harvey,
Dalton Trans. 2005, 2849 – 2858.
Received: April 7, 2008
Published online: July 10, 2008
[13] a) D. M. Khramov, A. J. Boydston, C. W. Bielawski, Angew.
Chem. 2006, 118, 6332 – 6335; Angew. Chem. Int. Ed. 2006, 45,
6186 – 6189; b) W. A. Herrmann, J. Schwarz, M. G. Gardiner,
Organometallics 1999, 18, 4082 – 4089; c) W. A. Herrmann, M.
Elison, J. Fischer, C. Köcher, G. R. J. Artus, Chem. Eur. J. 1996,
2, 772 – 780; d) K. Öfele, W. A. Herrmann, D. Mihalios, M.
Elison, E. Herdtweck, W. Scherer, J. Mink, J. Organomet. Chem.
1993, 459, 177 – 184; e) M. V. Baker, D. H. Brown, P. V. Simpson,
B. W. Skelton, A. H. White, C. C. Williams, J. Organomet. Chem.
2006, 691, 5845 – 5855; f) C. Marshall, M. F. Ward, T. A. Harri-
son, J. Organomet. Chem. 2005, 690, 3970 – 3975; g) J. U. Scheele,
S. Dechert, F. Meyer, Inorg. Chim. Acta 2006, 359, 4891 – 4900.
[14] S. Chen, J. H. Kim, C. E. Song, S. Lee, Org. Lett. 2007, 9, 3845 –
3848.
Keywords: alkenes · carbene ligands · cyclization · metathesis ·
ruthenium
.
[1] a) A. H. Hoveyda, A. R. Zhugralin, Nature 2007, 450, 243 – 251;
b) T. M. Trnka, R. H. Grubbs, Acc. Chem. Res. 2001, 34, 18 – 29;
c) D. Astruc, New J. Chem. 2005, 29, 42 – 56; d) A. Fürstner,
Angew. Chem. 2000, 112, 3140 – 3172; Angew. Chem. Int. Ed.
2000, 39, 3012 – 3043; e) R. R. Schrock, C. Czekelius, Adv. Synth.
Catal. 2007, 349, 55 – 77; f) P. H. Deshmukh, S. Blechert, Dalton
Trans. 2007, 2479 – 2491.
[2] K. C. Nicolaou, P. G. Bulger, D. Sarlah, Angew. Chem. 2005, 117,
4564 – 4601; Angew. Chem. Int. Ed. 2005, 44, 4490 – 4527.
[3] a) C. W. Bielawski, R. H. Grubbs, Prog. Polym. Sci. 2007, 32, 1 –
29; b) T. E. Hopkins, K. B. Wagener, Macromolecules 2004, 37,
1180 – 1189.
[4] a) O. A. Scherman, G. B. W. L. Ligthart, H. Ohkawa, R. P.
Sijbesma, E. W. Meijer, Proc. Natl. Acad. Sci. USA 2006, 103,
11850 – 11855; b) E. N. Guidry, S. J. Cantrill, J. F. Stoddart, R. H.
Grubbs, Org. Lett. 2005, 7, 2129 – 2132.
[5] W. H. C. Martin, S. Blechert, Curr. Top. Med. Chem. 2005, 5,
1521 – 1540.
[6] X. Liu, A. Basu, J. Organomet. Chem. 2006, 691, 5148 – 5154.
[7] a) A. Michrowska, K. Grela, Pure Appl. Chem. 2008, 80, 31 – 43;
b) B. F. Straub, Adv. Synth. Catal. 2007, 349, 204 – 214.
[8] a) A. J. Arduengo, R. L. Harlow, M. Kline, J. Am. Chem. Soc.
1991, 113, 361 – 363; b) M. Scholl, S. Ding, C. W. Lee, R. H.
Grubbs, Org. Lett. 1999, 1, 953 – 956.
[9] a) S. B. Garber, J. S. Kingsbury, B. L. Gray, A. H. Hoveyda, J.
Am. Chem. Soc. 2000, 122, 8168 – 8169; b) S. Gessler, S. Randl, S.
Blechert, Tetrahedron Lett. 2000, 41, 9973 – 9976.
[10] For equilibrium studies between RCM and ADMETreactions of
different dienes see: J. C. Conrad, M. D. Eelman, J. A. Duarte
Silva, S. Monfette, H. H. Parnas, J. L. Snelgrove, D. E. Fogg, J.
Am. Chem. Soc. 2007, 129, 1024 – 1025. For a study on RCM
selectivity for differently sized rings see: J. B. Beil, N. G.
Lemcoff, S. C. Zimmerman, J. Am. Chem. Soc. 2004, 126,
13576 – 13577.
[11] a) A. D. Averin, E. R. Ranyuk, N. V. Lukashev, I. P. Beletskaya,
Chem. Eur. J. 2005, 11, 7030 – 7039; b) S. Abramson, D.
Berkovich-Berger, S. Dagan, I. Goldberg, L. Golender, M.
Grabarnik, N. G. Lemcoff, S. Weinman, B. Fuchs, Eur. J. Org.
Chem. 2007, 1957 – 1975. c) For a hydrogen-bonded cyclodimer
[15] M. Kim, M. S. Eum, M. Y. Jin, K. W. Jun, C. W. Lee, K. A. Kuen,
C. H. Kim, C. S. Chin, J. Organomet. Chem. 2004, 689, 3535 –
3540.
À
[16] A recent NMR study showed no rotation around the Ru NHC
bonds in monomeric Grubbs catalysts: M. M. Gallagher, A. D.
Rooney, J. J. Rooney, J. Organomet. Chem. 2008, 693, 1252 –
1260.
[17] For an NMR experimental measurement of the rotation barrier
in a ruthenium olefin metathesis catalyst, see: A. Ben-Asuly, E.
Tzur, C. E. Diesendruck, M. Sigalov, I. Goldberg, N. G. Lemcoff,
Organometallics 2008, 27, 811 – 813.
[18] C60H70Cl4N4O2Ru2 (excluding disordered CH2Cl2 solvent), mon-
oclinic, space group C2/c, a = 31.0017(4), b = 17.4982(3), c =
24.4562(5) Š, b = 91.0662(5)8, V= 13264.6(4) Š3, T= 110 K,
1x = 1.225 gcmÀ3. After subtracting the contribution of the
disordered solvent from the diffraction data, R1 = 0.086 for
6897 reflections with I > 2s(I), and R1 = 0.13 and wR = 0.23 for
all 11561 unique data. CCDC 679217 contains the supplemen-
tary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[19] Rings of several sizes were observed as a consequence of
isomerization (see the Supporting Information). For more
information on isomerization of long-chain olefins during meta-
thesis, see: F. C. Courchay, J. C. Sworen, I. Ghiviriga, K. A.
Abboud, K. B. Wagener, Organometallics 2006, 25, 6074 – 6086.
[20] Similar results were obtained with other substrates, although the
largest difference with the monomeric catalysts was observed
with substrate 10 (see the Supporting Information).
Angew. Chem. Int. Ed. 2008, 47, 6422 –6425
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6425