Chemical transformations of cis-W(CO)4(C5H 5N)2 in the ring-opening metathesis polymerization of norbornene

Article abstract of DOI:10.1139/v04-007

The six-coordinate W⁰ complex cis-W(CO)₄(C ₅H₅N)₂ has been found to be active in the in situ formation of a carbene species from norbornene, which generates a typical ring-opening metathesis product (ROMP). A proposed mechanism of initiation suggests that the reaction involves a 2,3-hydrogen shift in the coordinated norbornene (η² → η¹). The initiating carbenoid group is identified from the products of the spontaneous carbene-CO coupling and Wittig reactions test. Formation of W(CO)₃(η ⁶-C₆H₅CH₃) when toluene is the solvent, followed by reaction with the carbene, is blamed for catalyst deactivation.

Full text of DOI:10.1139/v04-007

499
Chemical transformations of cis-W(CO)4(C5H5N)2 in
the ring-opening metathesis polymerization of
norbornene
L. Bencze, N. Bíró, B. Szabó-Ravasz, and L. Mihichuk
Abstract: The six-coordinate W0 complex cis-W(CO)4(C5H5N)2 has been found to be active in the in situ formation of
a carbene species from norbornene, which generates a typical ring-opening metathesis product (ROMP). A proposed
mechanism of initiation suggests that the reaction involves a 2,3-hydrogen shift in the coordinated norbornene (η2 →
η1). The initiating carbenoid group is identified from the products of the spontaneous carbene–CO coupling and Wittig
reactions test. Formation of W(CO)3(η6-C6H5CH3) when toluene is the solvent, followed by reaction with the carbene,
is blamed for catalyst deactivation.
Key words: bis-pyridine-tetracarbonyl-tungsten, ROMP, olefin metathesis, norbornene.
Résumé : On a trouvé que le complexe hexacoordiné du W0, cis-W(CO)4(C5H5N)2, est actif dans la formation in situ
d’une espèce carbénique dérivée du norbornène qui conduit à un produit résultant d’une ouverture typique de cycle
avec métathèse. Un mécanisme d’initiation proposé suggère que la réaction implique un déplacement 2,3 d’hydrogène
dans le norbornène pentacoordiné (η2 → η1). Le groupe carbénoïde qui provoque l’initiation est identifié à partir
des produits qui résultent d’un couplage spontané carbène–CO et de tests de réactions de Wittig. La formation de
W(CO)3(η6-C6H5CH3), quand on utilize du toluène comme solvant et qu’on effectue ensuite la réaction avec le carbène,
est attribuée à une désactivation du catalyseur.
Mots clés : bis-pyridine-tétracarbonyl-tungstène, produit d’ouverture de cycle avec métathèse, métathèse d’oléfine, nor-
bornène.
[Traduit par la Rédaction] Bencze et al. 503
Introduction
We wish to report here the catalytic properties of cis-
W(CO)4(C5H5N)2 (1) in the ring-opening metathesis poly-
merization (ROMP) of norbornene (NBE). This is the first
six-coordinate W0 noncarbenoid unimolecular catalyst that
does not require a co-catalyst, a carrier, or light for activa-
tion. The aim of this work is to study the initiation reactions
of a system in which the catalyst precursor contains the min-
imum number of ligands whose reactions might interfere
with an olefin → carbene transformation and which involves
no side reactions other than metathesis.
Olefin metathesis is of great value in organic and polymer
synthesis. The reactions do not occur spontaneously, but all
require the presence of a transition-metal carbene complex
(well defined catalyst) or a catalyst system containing a tran-
sition-metal complex (precursor) in conjunction with a
second compound, generally an organo main-group metal
compound (co-catalyst), and sometimes a third compound
(promoter) (1).
The transition-metal carbenes formed in these systems
trigger the transalkylidenation (i.e., metathesis) of olefins (2,
3). Catalyst systems having neither a preformed metallo-
carbene nor an organometallic hydrocarbon group in any
component represent a relatively minor class of metathesis
initiators (4, 5).
Experimental
cis-W(CO)4(C5H5N)2 (1) was prepared according to litera-
ture methods (6), with CO evolution being monitored until
complete conversion of W(CO)6 and formation of only the
tetracarbonyl complex, as seen by infrared analysis. All
experiments were conducted using standard Schlenk tech-
niques under deoxygenated argon. Toluene was distilled
from sodium and benzophenone, and NBE (Aldrich) was
distilled from sodium before use. Tungsten hexacarbonyl
(Aldrich) was used without further purification.
ROMP experiments were carried out in a reaction mixture
composed of toluene (10 mL), olefin (0.52 mol/dm3), and
cis-W(CO)4(C5H5N)2 (4.4 mmol/dm3). The reaction mixture
was immersed in an oil bath heated to 116 °C and was
stirred by a magnetic stirrer.
Received 10 July 2003. Published on the NRC Research
Press Web site at http://canjchem.nrc.ca on 10 March 2004.
L. Bencze, N. Bíró, B. Szabó-Ravasz, and L. Mihichuk.1,2
Department of Organic Chemistry, Müller Laboratory,
University of Veszprém, H-8201 Veszprém, POB 158,
Hungary.
1Permanent address: Department of Chemistry and
Biochemistry, University of Regina, Regina, SK S4S 0A2,
Canada.
2Corresponding author (e-mail: Lynn.Mihichuk@uregina.ca).
Can. J. Chem. 82: 499–503 (2004)
doi: 10.1139/V04-007
© 2004 NRC Canada

500
Can. J. Chem. Vol. 82, 2004
Fig. 1. Norbornene polymer yield vs. reaction time. Each point represents the average of three experimental runs.
The NBE reaction mixture changed in time, from yellow
to orange, to brown-orange, and then to a transparent golden
brown color, with a gradual increase of viscosity. The reac-
tion was quenched by the addition of ethanol (20 mL). The
precipitated polymer was washed with ethanol, dried, and
analyzed by 13C{1H} NMR spectroscopy. The spectra of the
polymers were recorded in CDCl3 solution at 125 MHz, us-
ing a GE GN Omega 500 spectrometer. The assignments of
13C{1H} NMR shifts were based on previous analysis, and
the microstructural details of the polymers were calculated
from the NMR integrals (7–10). GC–MS analyses were car-
ried out on a Shimadzu QP5000, 70 eV instrument equipped
with a 10 m × 0.20 mm HPI column at a flow rate of
2 mL/min. The identification of products from GC–MS anal-
yses was carried out using an existing mass spectral library
database. IR spectra were obtained using a SPECORD M80
spectrophotometer in CH2Cl2 solution, using KBr (0.065 mm)
cells.
inhibits the ROMP and leads to 2-benzylidenenorbornane
(m/z (%): 184 (25, [M]+), 155 (27), 130 (100), 115 (38), 77
(25)), the product of a Wittig-type carbene – aldehyde reac-
tion, as illustrated in Scheme 1.
To elucidate the mechanism of carbene initiation, a GC–
MS analysis was carried out on the low-molecular-weight
fraction of the ethanol-quenched solution. The major prod-
ucts identified and mass spectral data are listed in Table 1.
The first two entries (5 and 6 in Scheme 2) are also consis-
tent with a norbornylidene–carbene–ketene pathway, as found
earlier for an NBE–W(CO)3Cl2(AsPh3)2 system, with the
ketone being the major product formed (4c).
IR analysis of the quenched solution shows
W(CO)5(C5H5N) (2) and W(CO)3 (η6-C6H5CH3) (3) (besides
traces of 1) as the only complex-carbonyl-containing prod-
ucts identified by comparison of the νCO bands with litera-
ture values (6, 12) (Fig. 3). The GC–MS analysis of the
quenched solution also provides evidence for the presence of
the W(CO)3 (η6-C6H5CH3) complex by a comparison of its
mass spectrum with that of an authentic sample. The pres-
ence of the toluene-derived product, which was the second
major product, seemed to be perplexing in a metathesis cata-
lyst system; however, a separate experiment showed that it
also forms from cis-W(CO)4(C5H5N)2 in refluxing toluene
after a few hours, the result of a transformation pathway.
The transformation of the catalyst precursor is com-
menced by the loss of a pyridine ligand followed by the
competition of CO, NBE, and toluene for the vacant coordi-
nation site (Scheme 2). The basis for this deduction is pro-
vided by the fact that W(CO)5(C5H5N) is the only product
obtained when the above reaction is carried out in the pres-
ence of CO. The presence, in increasing quantity, of the
penta- (2) and tricarbonyl (3) species at the beginning and
by the end of the reaction helps to rule out their involvement
as precursors in the formation of the active carbene species.
Their presence does not maintain the catalytic activity. Ow-
ing to the rapid disappearance of cis-W(CO)4(C5H5N)2 and
the quantity of the trapped primary alkylidene groups, it is
Results and discussion
The rate of polymer formation is rather slow, but nearly
the same throughout the active period. The catalyst is poi-
soned or inactivated in some fashion within approximately
220 min. The maximum yield for the polymer is 25%, at-
tainable after 220 min of reaction, as seen in Fig. 1.
The solubility of the polymer in CDCl3 is satisfactory for
NMR examination, but it is not free from some swollen gel.
The 13C{1H} NMR analysis indicates a typical ROMP prod-
uct of NBE. The assignments of 13C{1H} NMR shifts, based
on previous analysis (7–10), were made as shown in Fig. 2,
where a comparison of line intensity illustrates the high cis
content (σc = 0.75–0.80) of the polymer. The distribution of
the cis and trans double bonds in all cases is blocky (rc·rt ~
9–14), consistent with the high cis- content of the polymers,
as previously found (9–11).
The addition of benzaldehyde to the NBE–
cisW(CO)4(C5H5N)2 system, followed by heating to 116 °C,
© 2004 NRC Canada

Bencze et al.
501
Fig. 2. Assignments of the 13C NMR chemical shifts, illustrating the high cis content of the NBE polymer. Carbon signals are labeled
as follows:
Table 1. Low-molecular-weight products from the NBE–cis-
Scheme 1.
W(CO)4(C5H5N)2 reaction.
Product
Mass spectral data (m/z, relative intensity (%))
168 (2, [M]+),139 (2), 123 (4), 95 (100)
easily envisaged that the tetracarbonyl species is the precur-
sor of the initiating carbene, and the latter forms soon after
the reaction commences.
218 (23, [M]+), 151 (69), 123 (70), 95 (100),
67 (88)
Thus, it appears that cis-W(CO)4(C5H5N)2 initiates the re-
action at 116 °C through loss of one pyridine ligand fol-
lowed by formation of a π-complex with NBE and
rearrangement by a 2,3-hydrogen shift (4) to give the initiat-
ing carbene. Loss of the other pyridine ligand is followed by
coordination of another NBE ligand in a π fashion, generat-
ing a metallacyclobutane intermediate (3), which then forms
the coordinatively unsaturated ROMP propagator. The slow
accumulation and the high molecular weight of the polymer
product indicates that the primary carbene complex is rela-
tively stable (13) and that the initiation is slow. Having no
low-molecular-weight polymer fraction formed, living poly-
merization is assumed with little termination. The 13C{1H}
NMR analysis of the polymer is indicative of only the re-
peating unit, as illustrated in the caption for Fig. 2, with no
spectral lines associated with end groups, thereby suggesting
no low-molecular-weight fraction.
112 (2, [M]+), 94 (60), 79 (100), 67 (70)
190 (18, [M]+), 161 (18), 95 (96), 67 (100)
186 (32, [M]+), 95 (100), 91 (36), 77 (8), 67
(28)
The catalyst deactivation, leading to an approximate 25%
polymer yield, is intriguing. The last entry in Table 1 is not a
product derived from a norbornylidene–carbene–ketene path-
way, as illustrated in Scheme 2, but may be indicative of an
additional pathway involving the initiating carbene complex
and the W(CO)3(η6-C6H5CH3) complex, leading to deactiva-
tion of the catalyst and the 25% maximum polymer yield. To
gain further insight concerning this aspect, we are actively
pursuing a detailed computational study of this reaction at
the DFT level using the Gaussian 98 series of programs (14).
It is interesting that although the olefin metathesis reaction
is one of the most studied in organometallic chemistry, there
have been few reports dealing with the modeling of this
reaction and the ROMP reaction using quantum chemistry
tools (15–17). A future publication will deal with these com-
putational results, as well as attempts to isolate the initiating
carbene complex and an extension of these studies with the
related norbornadiene substrate.
Conclusions
cis-W(CO)4(C5H5N)2 was found to be active in the in situ
formation of a metathesis active carbene species from NBE
substrate. The reactions follow the course of a 2,3-hydrogen
shift in the coordinated olefin, and the initiating carbene
could be identified from the products of the carbene–CO
© 2004 NRC Canada

502
Can. J. Chem. Vol. 82, 2004
Scheme 2.
Fig. 3. Infrared spectra in the CO stretching region for W(CO)4(C5H5N)2 (upper trace) and a mixture of W(CO)4(C5H5N)2 (A),
W(CO)5(C5H5N) (B), and W(CO)3(η6-C6H5CH3) (C) (bottom trace).
© 2004 NRC Canada

Bencze et al.
503
Szymanska-Buzar, T. Golowiak, and I. Czelusniak. J.
Organometal. Chem. 640, 72 (2001).
6. C.S. Kraihanzel and F.A. Cotton. Inorg. Chem. 2, 533 (1963).
7. K.J. Ivin, D.T. Laverty, and T.J. Katz. Makromol. Chem. 178,
1545 (1977).
coupling reactions and Wittig reactions, as also seen in sev-
eral seven-coordinate initiator systems.
This precatalyst (1) makes a unique single-component cat-
alytic system, involving a six-coordinate W0 species with no
organophosphine or halide ligands involved.
8. K.J. Ivin, D.T. Laverty, and J.J. Rooney. Makromol. Chem.
179, 253 (1978).
9. K.J. Ivin, D.T. Laverty, J.H. O’Donnel, J.J. Rooney, and C.D.
Stewart. Makromol. Chem. 180, 1989 (1979).
Acknowledgments
10. R.M.E. Greene, J.G. Hamilton, K.J. Ivin, and J.J. Rooney.
This work was supported by the Hungarian Science Foun-
dation under grants No. T35221 and T016326. L.M. thanks
the Department of Chemistry and Biochemistry, University
of Regina (Saskatchewan) for the granting of a sabbatical
leave and the Department of Organic Chemistry, University
of Veszprém (Hungary) for the invitation.
Makromol. Chem. 187, 619 (1986).
11. L. Bencze, G. Szalai, J.G. Hamilton, and J.J. Rooney. J. Mol.
Catal. A Chem. 115, 193 (1997).
12. V. Zanotti, V. Rutar, and R.J. Angelici. J. Organomet. Chem.
414, 177 (1991).
13. L. Bencze and R.K. Szilágyi. J. Organometal. Chem. 505, 81
(1995).
14. M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A.
Robb, J.R. Cheeseman, V.G. Zakrzewski, J.A. Montgomery,
Jr., R.E. Stratmann, J.C. Burant, S. Dapprich, J.M. Millam,
A.D. Daniels, K.N. Kudin, M.C. Strain, O. Farkas, J. Tomasi,
V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C.
Adamo, S. Clifford, J. Ochterski, G.A. Petersson, P.Y. Ayala,
Q. Cui, K. Morokuma, D.K. Malick, A.D. Rabuck, K.
Raghavachari, J.B. Foresman, J. Cioslowski, J.V. Ortiz, A.G.
Baboul, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I.
Komaromi, R. Gomperts, R.L. Martin, D.J. Fox, T. Keith,
M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe,
P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, J.L. Andres,
C. Gonzalez, M. Head-Gordon, E.S. Replogle, and J.A. Pople.
Gaussian 98 [computer program]. Revision A.9. Gaussian,
Inc., Pittsburgh, Pa. 1998.
References
1. K.J. Ivin and J.C. Mol. Olefin metathesis and metathesis poly-
merization. Academic Press, London. 1997.
2. (a) J.L. Harrison and Y. Chauvin. Makromol. Chem. 141, 161
(1970); (b) T.J. Katz, J. McGinnis, and C.J. Altus. J. Am.
Chem. Soc. 98, 606 (1976); (c) T.J. Katz and W.H. Hersh. Tet-
rahedron Lett. 585 (1977); (d) T.J. Katz, S.J. Lee, and M.A.J.
Shippey. J. Mol. Catal. 8, 219 (1980); (e) J. Kress, M.
Wesolek, and J.A. Osborn. J. Chem. Soc. Chem. Commun.
514 (1982); (f) R.R. Schrock, S. Rocklage, J. Wengrovius, G.
Rupprecht, and J. Fellman. J. Mol. Catal. 8, 73 (1980).
3. R.H. Grubbs and S. Chang. Tetrahedron, 54, 4413 (1998).
4. (a) D.T. Laverty, M.A. McKervey, J.J. Rooney, and A. Stewart.
J. Chem. Soc. Chem. Commun. 193 (1976); (b) L. Bencze,
K.J. Ivin, and J.J. Rooney. J. Chem. Soc. Chem. Commun. 834
(1980); (c) L. Bencze, A. Kraut-Vass, and L. Prokai. J. Chem.
Soc. Chem Commun. 911 (1985).
15. L.J. Cavallo. J. Am. Chem. Soc. 124, 8965 (2002).
16. S. Fomine, S.M. Vargas, and M.A. Tlenkopatchev.
Organometallics, 22, 93 (2003).
17. F. Bernardi, A. Bottoni, and G.P. Miscione. Organometallics,
5. (a) T. Szymanska-Buzar and T. Golowiak. J. Organometal.
Chem. 489, 207 (1995); (b) T. Szymanska-Buzar, T. Golowiak,
and I. Czelusniak. Inorg. Chem. 3, 102 (2000); (c) T.
22, 940 (2003).
© 2004 NRC Canada