A catalyst for the simultaneous ring-opening metathesis polymerization/vinyl insertion polymerization

Article abstract of DOI:10.1002/anie.201004125

Come together: Catalyst 1, which contains a diethylboryl-protected pyridyl unit, copolymerizes ethylene and cyclic olefins such as norborn-2-ene (NBE) and cis-cyclooctene (COE) to give multiblock copolymers. Both vinyl insertion polymerization and ring-opening metathesis polymerization are in effect within a single polymer chain, and the respective NBE and COE blocks can be obtained in different ratios. Copyright

Full text of DOI:10.1002/anie.201004125

Communications
DOI: 10.1002/anie.201004125
Designed Catalysts
A Catalyst for the Simultaneous Ring-Opening Metathesis Polymeri-
zation/Vinyl Insertion Polymerization**
Michael R. Buchmeiser,* Sebnem Camadanli, Dongren Wang, Yuanlin Zou, Ulrich Decker,
Christa Kꢀhnel, and Ingrid Reinhardt
Dedicated to Professor Walter Kaminsky on the occasion of his 70th birthday
Both the vinyl insertion polymerization
(VIP) of olefins[1–7] and the ring-opening
metathesis polymerization (ROMP)[8,9] of
cyclic olefins are well-established polymeri-
zation techniques based on the insertion of a
monomer between a transition metal (ion)
and a growing polymer chain. In fact, they
are related to each other, since a cationic,
VIP-active catalyst may be converted into a
ROMP-active catalyst by a simple a-elimi-
nation process. In contrast, the reverse pro-
cess, that is, the a addition of a proton to a
metal alkylidene to form a cationic VIP-
active species, is much more difficult to
accomplish in a controlled way.
The occurrence of both VIP- (Sche-
me 1a) and ROMP-derived (Scheme 1b)
structures in poly(NBE) (NBE = norborn-2-
ene) was observed already in the early days
of polyolefin chemistry and paved the way
for identifying and understanding metathe-
sis-based processes.[10,11] Nevertheless, at that
time there was no clear evidence for the
Scheme 1. Different types of poly(NBE) and poly(NBE)-co-PE obtained by VIP, ROMP and
presence of both structures, that is, both
ROMP- and VIP-derived repeat units, within
one single polymer chain.[10–14] In fact,
VIP/ROMP as well as the structure of the target copolymer.
because of the high crystallinity of these polymers, it was
argued that two different polymers, that is, one VIP- and one
ROMP-derived, had formed concomitantly. The first
unequivocal evidence for the presence of both ROMP- and
VIP-derived structures within one single polymer chain was
reported by Farina et al. using Mo- and Re-based initia-
tors;[15,16] they unambiguously assigned characteristic signals
for the quaternary carbon and for the methylene group in
VIP/ROMP-derived poly(NBE) (Scheme 1c).
In fact, the structure of this particular polymer can be
explained only by a reversible a-elimination/a-addition
process, which permits VIP and ROMP to take place within
the same polymer chain. Owing to the relationship between
VIP and ROMP, even one single switch between VIP and
ROMP, resulting in the synthesis of AB diblock copolymers,
has attracted attention. Thus, Grubbs et al. reported on a
titanacyclobutane compound active in the ROMP of NBE;
upon addition of an alcohol and activation with Et2AlCl, a
cationic species was created that is active in the VIP of
ethylene (E), resulting in the synthesis of poly(NBE)ROMP-b-
poly(E).[17] Kaminsky et al. reported on a controlled switch
from VIP to ROMP in the preparation of poly(NBE)VIP-b-
poly(NBE)ROMP, poly(NBE)VIP-b-poly(CPE)ROMP, and poly-
(NBE)VIP-b-poly(COE)ROMP (CPE = cyclopentene, COE =
cis-cyclooctene).[18,19]
[*] Prof. Dr. M. R. Buchmeiser, Dr. D. Wang
Lehrstuhl fꢀr Makromolekulare Stoffe und Faserchemie
Institut fꢀr Polymerchemie, Universitꢁt Stuttgart
Pfaffenwaldring 55, 70550 Stuttgart (Germany)
Fax: (+49)711-3940-185
E-mail: michael.buchmeiser@ipoc.uni-stuttgart.de
Dr. S. Camadanli, Dr. Y. Zou, Dr. U. Decker, C. Kꢀhnel, I. Reinhardt
Leibniz-Institut fꢀr Oberflꢁchenmodifizierung e.V. (Germany)
[**] This work was supported by the Wissenschaftsgemeinschaft
Wilhelm Gottfried Leibniz (WGL), the Federal Government of
Germany, and the Freistaat Sachsen.
We aimed at the creation of a Group IV based initiator
system capable of forming both VIP- and ROMP-derived
structures from a cyclic olefin and also capable of copolymer-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201004125.
3566
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 3566 –3571

izing an acyclic olefin to yield high-molecular-weight copoly-
mers with a narrow molecular-weight distribution containing
multiple blocks of both ROMP- and VIP-derived structures
within one single polymer chain. Such unprecedented poly-
meric structures, particularly those derived from COE and E,
would display features of standard cycloolefin copolymers
(COCs)[5] without displaying the standard drawbacks of pure
ROMP-derived polymers, that is, lack of stability in the
presence of oxygen. Additionally, they would offer the
possibility of functionalization through polymer-analogous
reactions, for example, halogenation, epoxidation, and vicinal
dihydroxylation. In this way, the poor derivatization tech-
niques for standard polyolefins can be bypassed and the
copolymerization of functional monomers can be circum-
vented. Particularly the target homogeneous distribution of
the ROMP- and VIP-derived structures along the polymer
backbone, rather than an AB-block-copolymer structure,
would support these features.
Upon activation, for instance with methylalumoxane
(MAO),[26] this highly electrophilic complex was expected to
be capable of accomplishing this task. By design, the free
pyridyl group should be capable of abstracting a proton, that
is, inducing a elimination via a six-membered transition state.
The boryl group competes for the free lone pair of electrons;
thus the a-elimination process should be temperature depen-
dent. We also synthesized the corresponding amine/borane-
free model compound Ti-3 to illustrate and clarify the role of
the amine/borane ligand, respectively.
Upon activation with MAO (Ti-8/MAO = 1:1000) and
addition of 1000 molequiv of NBE, pure poly(NBE)ROMP
(Mn = 4600 gmolÀ1, PDI = 1.51, cis/trans = 40:60) was
obtained at T= 508C. The 1H and 13C NMR spectra (in
=
C6D4Cl2) show the characteristic signals at d = 5.41 (C Ctrans),
=
5.21 (C Ccis), and around 134.0, 134.9 ppm, respectively. As
expected a glass transition was observed (Tg = 738C) and no
reflections for crystalline regions could be observed by wide-
angle X-ray diffraction (WAXD). However, when the NBE
concentration was increased (10000 molequiv), poly(NBE)
containing both VIP- and ROMP-derived repeat units was
Zirconium and titanium are suitable transition metals for
both VIP- and ROMP-active catalysts; in other words,
cationic alkyl complexes as well as methylidene complexes
of ZrIV and TiIV
exist.[20–23] These metal
methylidenes
are
active initiators in the
ROMP of cyclic ole-
fins such as NBE.[24,25]
However, an initiator
capable of reversibly
switching
between
ROMP and VIP must
bear a ligand system
that can reversibly
abstract
from the cationic spe-
cies and re-add
a
proton
a
proton to the metal
alkylidene. Preferably,
one should have con-
trol over this process,
for
example,
by
change of tempera-
ture. These require-
ments can be met
with a basic auxiliary
ligand; however, its
reversible protection
is required to prevent
its irreversible coordi-
nation to the cationic
metal center.
In view of these
requirements,
we
designed an amine/
borane-containing,
TiIV-based, half-sand-
wich precatalyst with
a constrained geome-
try (Ti-8, Scheme 2). Scheme 2. Synthesis of Ti-3 and Ti-8.
Angew. Chem. Int. Ed. 2011, 50, 3566 –3571
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3567

Communications
obtained. Again,
a
fully amorphous polymer (Mn =
to 928C. The role of MAO becomes obvious when one
compares entries 3 and 5 in Table 1. An increase in MAO
concentration results in an increase in activity (better
stabilization of the cationic species), a decrease in PDI
(better stabilization of the cationic species, reduced chain
transfer), and a increase in the poly(NBE)ROMP content. There
is a simple explanation: a elimination results in pyridinium
moieties which react irreversibly with MAO to form methane
and a titanium alkylidene. The latter gives rise to a polymer
with a higher poly(NBE)ROMP content but also a lower Mn
value, since the reaction of the titanium alkylidene with
ethylene cleaves the polymer chain from the metal center.
To support these findings and identify the actual structure
of the polymer, we compared the 13C NMR spectra of
10000 gmolÀ1, PDI = 1.47, Tg = 1368C) was obtained as
evidenced by WAXD. To identify the active species, Ti-8
was mixed with MAO and NBE (1:20:50) in C2D2Cl4. In the
1H NMR spectrum, a doublet at d = 8.01 ppm (J = 7 Hz)
=
À
À
=
corresponding to the [Ti CH c-(1,2-C5H8) CH CHR]
moiety[24,25] was observed, which disappeared upon heating
above 508C.
In the copolymerization of ethylene (4 bar) with NBE, Ti-
8 again displayed NBE-concentration-dependent behavior.
Thus, a pure VIP-derived E-NBE copolymer was obtained at
T= 508C when Ti-8/MAO/NBE was used in a ratio of
1:1000:2000 (Table 1, entry 1). However, at a higher NBE
concentration (Table 1, entry 2), a fully amorphous polymer
poly(NBE)ROMP-co-poly(NBE)VIP
-
co-poly(E) polymers with those of
pure ROMP-derived poly(NBE) as
well as of poly(NBE)ROMP-alt-
poly(E3) formed by the alternating
ring-opening metathesis copolymer-
ization of NBE with COE
(Figure 1).[27,28] Poly(NBE)ROMP-alt-
poly(E3) mimics one ROMP-
derived poly(NBE) repeat unit fol-
lowed by three ethylene units. No
signals were found for poly-
Table 1: Ti-8-catalyzed copolymerization of E with NBE or COE; reaction time: 1 h.
Entry
Ti-8/MAO/NBE
T [8C]
A[a]
Mn [gmolÀ1
]
PDI
Tm [8C]
C[b]
1
2
3
4
5
1:1000:2000
1:1000:10000
1:1000:20000
1:2000:10000
1:2000:20000
50
50
50
70
50
64
35
50
65
115
570000
1.32
1.05
1.29
1.94
1.08
116
122
125
123
92
0
1100000
1540000
480000
16
25
n.a.
71
1180000
Ti-8/MAO/COE
T [8C]
A[a]
Mn [gmolÀ1
]
PDI
Tm [8C]
X[b]
6
7
8
1:1000:2000
1:1000:10000
1:2000:10000
50
50
70
64
12
34
1500000
800000
490000
1.14
1.22
2.5
129
145
142
0
42
8
À
=
(NBE)ROMP-poly(E), that is, CH
[a] Activity in kgmolÀ1 barh. [b] The mol% of poly(NBE)ROMP and poly(COE)ROMP, respectively; n.a.: not
analyzed.
À
À
=
À
CH c-(1,2-C5H8) CH CH
À
(CH2)n sequences. The signals at
d = 47.5 and 41.3 ppm correspond
to the alternating, isotactic VIP-
was obtained containing multiple blocks of both ROMP- and
VIP-derived NBE units (poly(NBE)ROMP/poly(NBE)VIP
poly(E) = 1:1:4). A further increase in NBE concentration
derived E-NBE diads.[29–34] In addition, E-NBE-E-E sequen-
ces become visible. The signals at d = 46.8 and 41.1 ppm
correspond to the alternating, syndiotactic VIP-derived E-
NBE sequences, while the signal at d = 32.6 ppm stems from
alternating E-NBE and isolated NBE sequences. Finally, the
signal at d = 29.4 ppm can be attributed to homo-poly(E)
sequences. The most important signals, however, are found at
d = 133.9, 132.9 42.9, 42.7, 42.5, 41.8, 41.1, 38.5, 38.3, and
33.2 ppm. These can be assigned unambiguously to poly-
(NBE)ROMP-alt-poly(NBE)VIP sequences and provide clear
evidence for the incorporation of the poly(NBE)ROMP units
into the polymer main chain. No signals indicative of a 1,7-
connectivity of poly(NBE)VIP sequences were observed.[35,36]
These findings exclude the presence of the two polymers
poly(NBE)ROMP and poly(NBE)VIP-co-poly(E) as well as the
following alternative pathway: Poly(NBE)ROMP could form by
initial a elimination followed by cleavage of the titanium
alkylidene by E, which would result in a vinyl-terminated
polymer. This macromonomer could then copolymerize with
NBE and E. In fact, our data support the reaction mechanism
shown in Scheme 3: Reaction of a cationic VIP-active species
with NBE, followed by a elimination produces a disubstituted
titanium alkylidene whose formation is favored over the
formation of a monosubstituted alkylidene from the a elimi-
nation after E insertion. High concentrations of NBE
promote both the a elimination and the ROMP of this
monomer. The proton stays, at least for a certain time, at the
pyridine moiety and can be donated back to the titanium
/
resulted in a further increase in the proportion of ROMP-
derived poly(NBE) units (poly(NBE)ROMP/poly(NBE)VIP
/
poly(E) = 2:1:5), and in a further increase in the molecular
weight (Table 1, entry 3). It is important to emphasize that
both the ROMP- and VIP-derived poly(NBE) sequences
occur within the same polymer chain, as suggested by the
narrow polydispersity index of the polymers (PDI ꢀ 1.3,
Table 1), the absence of any additional peaks in the gel
permeation chromatogram that could be assigned to poly-
(NBE)ROMP, the absence of a glass transition attributable to a
poly(NBE)ROMP homopolymer, and the fact that no high Tm
values were found like those usually observed for cyclic olefin
copolymers.
An increase in reaction temperature to 708C resulted in a
decrease in Mn and in an increase in PDI (Table 1, entry 4).
This is attributed to a higher fraction of unprotected pyridyl
moieties which directly results in an increase in a elimination.
Consequently, the concentration of intermediary titanium
alkylidenes is higher; however, these ultimately react with E
to form titanium methylidenes, which decompose at this
temperature (vide supra). This side reaction also accounts for
the observed activities, which are lower than those obtained
with similar Ti(Cp*SiMe2NR) systems.[3] A further increase in
NBE concentration increased the poly(NBE)ROMP fraction
within the polymer to 71 mol%. Consequently, Tg decreased
3568
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 3566 –3571

Figure 1. 13C NMR spectra (in C2D2Cl4) of poly(NBE)ROMP (c), poly(NBE)ROMP-alt-poly(E3) (a), Ti-8-derived poly(NBE)ROMP-co-poly(NBE)VIP-co-poly(E)
(d), and poly(NBE)VIP-co-ploy(E) (b). alt-syndio: alternating, syndiotactic; alt-isot.: alternating, isotactic; V: VIP product; R: ROMP product.
not the most favorable
and
requires,
as
observed, high con-
centrations of NBE
to stabilize the tita-
nium alkylidene. The
proposed structure of
poly(NBE)ROMP-co-
poly(NBE)VIP-co-
poly(E) is shown in
Scheme 1d.
If all that is true,
then the homopoly-
merization of COE as
well as its copolymer-
ization with E should
give similar results. In
fact, the homopolyme-
rization of COE by Ti-
8 activated by MAO
(Ti-8/MAO/COE =
1:1000:1000) at 508C
resulted in the forma-
tion of pure ROMP-
derived
(Mn = 115000 gmolÀ1,
PDI = 1.48, Tg =
poly(COE)
Scheme 3. Proposed mechanism for the Ti-8-catalyzed formation of poly(NBE)ROMP-co-poly(NBE)VIP-co-poly(E).
alkylidene to re-establish the VIP-active species. At this
point, an NBE monomer can insert again, most probably
followed by E insertion. As one might expect, this process is
À548C). Both the 1H and 13C NMR spectra (in C6D4Cl2)
=
=
show the characteristic signals at d = 5.39 (C Ctrans), 5.34 (C
Ccis) and 127.7, 130.8 ppm (cis/trans ꢁ 50:50). Similarly, the
Angew. Chem. Int. Ed. 2011, 50, 3566 –3571
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3569

Communications
copolymerization of E with COE at a
Ti-8/MAO/COE ratio of 1:1000:2000
(Table 1, entry 6) gave crystalline
poly(COE)VIP-co-poly(E)
(Tm =
1298C). An increase in the COE
concentration (Table 1, entry 7)
again resulted in the formation of
poly(COE)ROMP-co-poly(COE)VIP
-
co-poly(E) (Tm = 1458C) in a ratio of
8:1:10, that is, with a very high
content of ROMP-derived poly-
(COE). The unimodal, comparably
narrow molecular-weight distribu-
tion (PDI = 1.22), the different cis/
trans ratios, and the fact that no glass
transition for poly(COE)ROMP was
observed strongly suggest that the
multiple poly(COE)ROMP blocks are
part of the polymer chain and do not
form
a second, separate purely
ROMP-derived polymer.
Figure 2 shows the 13C NMR
spectra of poly(COE)ROMP-co-poly-
(COE)VIP-co-poly(E),
derived poly(COE), and
a
ROMP-
VIP-
a
derived poly(COE)VIP-co-poly(E).
Signals of both the ROMP- (d =
130.1, 129.7, 32.2, 29.4, 28.8, and
27.0 ppm) and VIP-derived polymer
Figure 2. 13C NMR spectra (in C2D2Cl4) of Ti-8-derived poly(COE)ROMP-co-poly(COE)VIP-co-poly(E)
(bottom), poly(COE)VIP-co-poly(E) (middle), and poly(COE)ROMP (top).
at d = 39.0, 32.8, 30.5, 29.8, 29.1, 28.0, 27.9, 25.8 ppm are
visible; however, owing to additional signals of olefins arising
from a elimination, the overall cis/trans ratio of the double
bonds is about 1:1 as compared to a value of 3:7 for pure Ti-8-
derived poly(COE)ROMP. An increase in the Ti-8/MAO ratio
(1:2000, Table 1, entry 8) and in the reaction temperature to
T= 708C resulted in the formation of poly(COE)ROMP-co-
poly(E) (poly(COE)ROMP/poly(E) = 1:16). The absence of any
additional peaks apart from those for linear poly(E) and
poly(COE)ROMP in the 13C NMR spectrum suggests a highly
linear structure for the poly(E)-derived blocks.
The central role of the auxiliary pyridine ligand becomes
apparent when the above-described homo- and copolymer-
izations are carried out with Ti-3 instead of Ti-8. Thus, Ti-3
copolymerizes NBE and E in the same way as other (h5-
cyclopentadienyl)silylamido titanium(IV) catalysts:[5,37] only
poly(NBE)VIP-co-poly(E) (300000 < Mn < 1400000 gmolÀ1
;
1.07 < PDI < 1.56) was obtained (Table 2, entries 1–4).
Under none of the applied reaction conditions, including
those used for Ti-8, were ROMP-derived structures observed.
The same holds for the copolymerization of E with COE
(Table 2, entries 5 and 6).
In summary, we have described a catalytic system capable
of synthesizing copolymers of E with cyclic olefins that
contain both ROMP- and VIP-derived structures within one
polymer chain. We are currently exploring the limits of this
methodology in terms of monomers and the special properties
of the copolymers obtained.
Received: July 6, 2010
Revised: January 23, 2011
Published online: March 15, 2011
Keywords: homogeneous catalysis · insertion polymerization ·
metathesis · polymers · ring-opening polymerization
.
Table 2: Ti-3-catalyzed copolymerization of E with NBE or COE; reaction
time: 1 h, 4 bar, T=508C.
Ti-3/MAO/NBE
cNBE
A[b]
Mn [gmolÀ1
]
PDI
[a]
1:1000:2000
1:1000:5000
1:2000:10000
1:2000:20000
3.3
10
16
19
66
1400000
470000
300000
320000
1.07
1.53
1.38
1.56
300
745
410
[1] L. Resconi, L. Cavallo, A. Fait, F. Piemontesi, Chem. Rev. 2000,
100, 1253 – 1345.
[2] G. W. Coates, Chem. Rev. 2000, 100, 1223 – 1252.
[3] V. C. Gibson, S. K. Spitzmesser, Chem. Rev. 2003, 103, 283 – 315.
[4] G. G. Hlatky, Chem. Rev. 2000, 100, 1347 – 1376.
[5] X. Li, Z. Hou, Coord. Chem. Rev. 2008, 252, 1842 – 1869.
[6] W. Kaminsky, Adv. Catal. 2001, 46, 89 – 159.
A[b]
Mn [gmolÀ1
]
PDI
Tm [8C]
[a]
Ti-3/MAO/COE
cCOE
1:1000:2000
1:2000:10000
1
16
170
670
2500000
1750000
1.07
1.21
122
119
[7] W. Kaminsky, Macromol. Chem. Phys. 2008, 209, 459 – 466.
[8] M. R. Buchmeiser, Chem. Rev. 2000, 100, 1565 – 1604.
[a] Amount of cyclic olefin (mol%) in the copolymer. [b] Activity in
kgmolÀ1 barh.
3570
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 3566 –3571

[9] “Ring-Opening Metathesis Polymerization”: M. R. Buchmeiser
in Handbook of Ring-Opening Polymerization, 1st ed. (Eds.: P.
Dubois, O. Coulembier, J.-M. Raquez), Wiley-VCH, Weinheim,
2009.
[10] G. DallꢀAsta, J. Polym. Sci. A 1968, 6, 2397 – 2404.
[11] G. DallꢀAsta, G. Motroni, J. Polym. Sci. A 1968, 6, 2405 – 2413.
[12] T. Tsujino, T. Saegusa, J. Furukawa, Makromol. Chem. 1965, 85,
71 – 79.
[24] L. R. Gilliom, R. H. Grubbs, J. Am. Chem. Soc. 1986, 108, 733 –
742.
[25] N. A. Petasis, D.- K. Fu, J. Am. Chem. Soc. 1993, 115, 7208 – 7214.
[26] W. Kaminsky, M. Miri, H. Sinn, R. Woldt, Makromol. Chem.
Rapid Commun. 1983, 4, 417 – 421.
[27] K. Vehlow, D. Wang, M. R. Buchmeiser, S. Blechert, Angew.
Chem. 2008, 120, 2655 – 2658; Angew. Chem. Int. Ed. 2008, 47,
2615 – 2618.
[13] G. DallꢀAsta, G. Mazzanti, G. Natta, L. Porri, Macromol. Chem.
1962, 56, 224 – 227.
[14] M. A. Alonso, M. F. Farona, Polym. Bull. 1987, 18, 203 – 207.
[15] M. A. Alonso, K. E. Bower, J. A. Johnston, M. F. Farona, Polym.
Bull. 1988, 19, 211 – 216.
[16] J. A. Johnston, M. Tokles, G. S. Hatvany, P. L. Rinaldi, M. F.
Farona, Macromolecules 1991, 24, 5532 – 5534.
[17] I. Tritto, M. C. Sacchi, R. H. Grubbs, J. Mol. Catal. 1993, 82, 103 –
111.
[28] M. Lichtenheldt, D. Wang, K. Vehlow, I. Reinhardt, C. Kꢁhnel,
U. Decker, S. Blechert, M. R. Buchmeiser, Chem. Eur. J. 2009,
15, 9451 – 9457.
[29] D. Ruchatz, G. Fink, Macromolecules 1998, 31, 4669 – 4673.
[30] D. Ruchatz, G. Fink, Macromolecules 1998, 31, 4674 – 4680.
[31] D. Ruchatz, G. Fink, Macromolecules 1998, 31, 4681 – 4683.
[32] D. Ruchatz, G. Fink, Macromolecules 1998, 31, 4684 – 4686.
[33] I. Tritto, L. Boggioni, M. C. Sacchi, P. Locatelli, J. Mol. Catal. A
1998, 133, 139 – 150.
[18] R. Manivannan, G. Sundararajan, W. Kaminsky, Macromol.
Rapid Commun. 2000, 21, 968 – 972.
[19] R. Manivannan, G. Sundararajan, W. Kaminsky, J. Mol. Catal. A
2000, 160, 85 – 95.
[20] F. M. Hartner, Jr., J. Schwartz, S. M. Clift, J. Am. Chem. Soc.
1983, 105, 640 – 641.
[21] J. D. Meinhart, E. V. Anslyn, R. H. Grubbs, Organometallics
1989, 8, 583 – 589.
[34] R. A. Wendt, G. Fink, J. Mol. Catal. A 2003, 203, 101 – 111.
[35] C. Karafilidis, H. Hermann, A. Rufinska, B. Gabor, R. J. Mynott,
G. Breitenbruch, C. Weidenthaler, J. Rust, W. Joppek, M. S.
Brookhart, W. Thiel, G. Fink, Angew. Chem. 2004, 116, 2498 –
2500; Angew. Chem. Int. Ed. 2004, 43, 2444 – 2446.
[36] C. Karafilidis, K. Angermund, B. Gabor, A. Rufinska, R. J.
Mynott, G. Breitenbruch, W. Thiel, G. Fink, Angew. Chem. 2007,
119, 3819 – 3823; Angew. Chem. Int. Ed. 2007, 46, 3745 – 3749.
[37] G. J. P. Britovsek, V. C. Gibson, D. F. Wass, Angew. Chem. 1999,
111, 448 – 468; Angew. Chem. Int. Ed. 1999, 38, 428 – 447.
[22] J. Schwartz, K. I. Gell, J. Organomet. Chem. 1980, 184, C1-C2.
[23] J. J. Eisch, A. A. Adeosun, Eur. J. Org. Chem. 2005, 993 – 997.
Angew. Chem. Int. Ed. 2011, 50, 3566 –3571
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3571