Formation of zirconocene fluoro complexes: No deactivation in the polymerization of olefins by the contact-ion-pair catalysts [Cp′ 2ZrR]+[RB(C6F5)3] -

Article abstract of DOI:10.1002/anie.200600361

(Chemical Equation Presented) Not a catalyst killer: In the reaction of 2 with B(C₆F₅)₃, a nucleophilic aromatic substitution with C-F bond cleavage leads to the unexpected difluoride 1. Although the formation of Zr-F species during olefin polymerization by [Cp′₂ZrR]⁺[RB(C₆F₅) ₃]^- (Cp′ = substituted or unsubstituted η⁵-cyclopentadienyl; R = Me, H) catalysts is often described as a deactivation, the catalyst precursor 2 can be regenerated from 1 through treatment with iBu₂AlH.

Full text of DOI:10.1002/anie.200600361

Angewandte
Chemie
À
C F Activation
DOI: 10.1002/anie.200600361
Formation of Zirconocene Fluoro Complexes:
No Deactivation in the Polymerization of Olefins
by the Contact-Ion-Pair Catalysts
[Cp’2ZrR]+[RB(C6F5)3]À**
Perdita Arndt,* Ulrike Jäger-Fiedler, Marcus Klahn,
Wolfgang Baumann, Anke Spannenberg,
Vladimir V. Burlakov, and Uwe Rosenthal*
In memory of Erhard Kurras
Certain ion pairs [Cp’2MMe]+[MeB(C6F5)3]À (M = Ti, Zr;
Cp’ = substituted or unsubstituted h5-cyclopentadienyl),
formed by the formalabstraction of methylanions from
dimethylcompelxes [Cp ’2MMe2] of the Group 14 transition
metals by the strong Lewis acid B(C6F5)3, are highly active
catalysts for the polymerization of a-olefins.[1] Zirconocene
complexes of olefins, alkynes, and butadiene were also later
activated for catalytic olefin polymerization in a similar way.[2]
Recently, we described the reactions of complexes con-
À
taining two M C s bonds, including the zirconacyclopropenes
[Cp’2Zr(h2-RC2R)] (R = SiMe3)[3a] and severalfive-membered
zirconacycles,[3b,c] with B(C6F5)3. A typicalexampel is [ rac-
(ebthi)Zr(h2-Me3SiC2SiMe3)] (A; ebthi = 1,2-ethylene-1,1’-
bis(h5-tetrahydroindenyl)),[4a] which undergoes electrophilic
substitution at the five-membered ring of the ebthi ligand
upon reaction with B(C6F5)3, forming the alkenyl complex B
[Eq. (1)].[4b] In this complex, the boranate group is substituted
at the 3-position of the five-membered ring, and one of its
ortho fluorine atoms is coordinated to the zirconium center.
The s-bound alkenyl group at the zirconium center partic-
ipates in an additionalagostic interaction. [4b]
Upon thermolysis of B [Eq. (1)], bis(trimethylsilyl)acety-
À
lene is eliminated, and with the cleavage of one B C bond,
[*] Dr. P. Arndt, U. Jäger-Fiedler, M. Klahn, Dr. W. Baumann,
Dr. A. Spannenberg, Prof. Dr. U. Rosenthal
Leibniz-Institut für Katalyse e.V.
an der Universität Rostock
Albert-Einstein-Strasse 29a, 18059 Rostock (Germany)
Fax: (+49)381-1281-51176
E-mail: perdita.arndt@ifok-rostock.de
uwe.rosenthal@ifok-rostock.de
Dr. V. V. Burlakov
Nesmeyanov Institute of Organoelement Compounds
Russian Academy of Sciences
Vavilov Street 28, 117813 Moscow (Russia)
[**] We are grateful to the Deutsche Forschungsgemeinschaft (Research
Training Group 1213 and SPP 1118, project no. RO 1269/6) and the
Russian Foundation for Basic Research (project code 05-03-32515)
for financial support. We thank our technical staff, in particular Petra
Bartels and Regina Jesse, for skilled assistance. Cp’=substituted or
unsubstituted h5-cyclopentadienyl; R=Me, H.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2006, 45, 4195 –4198
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4195

Communications
one C6F5 group is transferred from boron to zirconium. The
resulting complex C contains one bridging m-hydrogen atom
between the boron and zirconium centers.[4b]
Complexes B and C can serve as modelcompounds for the
two most frequently reported side reactions leading to
catalyst poisoning and decreased productivity.[5a] The first
deactivation pathway is the transfer of a C6F5 group to the
cationic metalcentre ( C).[4b,5] The second consists of the
transfer of a fluoride anion to the metal center (via
intermediates such as B). It has been proposed that the
formation of catalytically inactive zirconocene fluorides is an
important deactivation process.[5a]
Nevertheless, we recently found that the difluoride [rac-
(ebthi)ZrF2] (1),[6b,c] in contrast to the analogous dichloride,[6a]
is an efficient catalyst for ethylene polymerization after
activation with iBu3Al(or iBu2AlH).[7a–c] In the literature,
À
À
there are severalempiricalreports of Ti F and Zr F olefin-
polymerization catalysts.[7d–f] To understand these results, we
studied the reaction of 1 with two equivalents of iBu2AlH and
found that hydride anions replace the fluoride ligands of 1 to
give the complex [{rac-(ebthi)ZrH(m-H)}2] (2) (Scheme 1).[8a,b]
Interestingly, the reaction of the dichloride [rac-(ebthi)ZrCl2]
with iBu2AlH under analogous conditions does not lead to a
hydride; rather, only the starting material is isolated.[8a] The
fluoride anion is apparently a more labile ligand than the
chloride anion in this system, which explains the easier
abstraction of a fluoride ligand to give an active catalyst
À
Scheme 1. Elementary steps of C F bond activation in the reaction of
2 with B(C6F5)3.
the product composition or lead to inseparable mixtures of
other complexes, which could not yet be characterized.
Orange crystals of complex 3 form very quickly at 208C in
toluene, n-hexane, or benzene. The insolubility and instability
of this complex prevents a detailed NMR spectroscopic
investigation or recrystallization. Therefore, only low-quality
crystals could be obtained for X-ray crystal-structure analysis.
Nevertheless, it was possible to confirm that 3 consists of an
ion pair of the type [{rac-(ebthi)Zr(m-H)}2]+[HB(C6F5)3]À;[14]
in analogy to the ion pair [Cp*2ZrH]+[HB(C6F5)3]À,[10] the
presence of one terminalhydrogen atom in the cation is
postulated (Scheme 1), but could not be located.
Complex 3 reacts slowly (608C, 5–7 h) with additional
B(C6F5)3 to give the yellow mononuclear complex [rac-
(ebthi)Zr(m-H){o-C6F4B(C6F5)2}] (4) and varying amounts of
complex 5. The molecular structure of 4 contains the typical
{rac-(ebthi)Zr} unit; the zirconium center is bonded to an
ortho-metalated phenyl ring of the boranate anion, as well as
a hydride ligand (H1; Figure 1).[15] The positionaland thermal
parameters of H1 could not be refined. However, in the
1H{11B} NMR spectrum of 4 a signal( d = 1.27 ppm) assigned
to the hydride ligand is detected in the region of the rac-
(ebthi)-ligand signals. Further NMR spectroscopic investiga-
tions were hampered by the instability of 4 in solution.
Nevertheless, high yields of 1 can be obtained by the reaction
of 2 with B(C6F5)3 or by the thermolysis of complex 3.
Only one other example of a similar fluoride transfer from
B(C6F5)3 to zirconium has been described: starting from
À
À
system. The cleavage of the Zr F bonds by Al H groups with
À
À
the formation of Zr H and Al F bonds is accepted as the
reason for this “fluoride effect”.[7,8]
Zirconocene hydrides have frequently been used for the
À
activation of C F bonds. In a series of outstanding papers,
À
Jones and co-workers have described the cleavage of C F
bonds in alkanes, arenes, and olefins using [Cp*2ZrH2] (Cp* =
h5-pentamethylcyclopentadienyl).[9]
In light of the surprising results mentioned above for
zirconocene difluorides in the catalytic polymerization of
À
olefins, we describe herein the effective cleavage of C F
bonds in B(C6F5)3 by complex 2, which is formed by the
À
À
reaction of Zr F and Al H species.
The reaction of 2 with B(C6F5)3 in toluene at 908C for
14 days leads to a nearly quantitative formation of 1
(Scheme 1). A small amount of complex 5 is formed as a
byproduct. To elucidate the reaction pathway, not only was
the intermediate 3 isolated, but NMR spectroscopic experi-
ments were also carried out. In these experiments, compounds
such as 4 and 5, which are difficult to characterize or isolate
because of their instability, were detected. Note that the
system is extremely sensitive with respect to slight changes in
reaction conditions. Variations in the conditions can change
4196 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4195 –4198

Angewandte
Chemie
The formation of the difluoride 1 (by nucleophilic
aromatic substitution on B(C6F5)3), as well as complexes 5
and C (by C6F5-group migration) is relevant for the catalytic
polymerization of olefins. Complexes such as 5 and C are
completely inactive, and their formation clearly represents a
deactivation process. The formation of 1 from 2 and B(C6F5)3
À
by C F bond cleavage, often described as a deactivation
process, is not in fact a deactivation, because 1 can be
reconverted into
(Scheme 1).
2 through treatment with iBu2AlH
These results, obtained through the stepwise reactions of
isolated complexes, have important consequences for cata-
lytic olefin polymerization in general: mixtures of zircono-
cene halides such as [Cp’2ZrCl2], methylalumoxane (MAO; as
an activator and scavenger), and fluorosubstituted arylbor-
anes such as B(C6F5)3 are often used to produce catalytically
active systems. Hydrogen gas is also added to the system to
control the molecular weight of the polymer. Marks and
coworkers have shown that the methylcomplex [Cp* 2ZrMe2]
reacts with hydrogen to produce the hydride [Cp*2ZrH2], with
elimination of methane. The hydride subsequently reacts with
B(C6F5)3 to form the active ion pair [Cp*2ZrH]+[HB(C6F5)3]À
(Scheme 2).[10]
Figure 1. ORTEP representation of 4. Thermal ellipsoids are set at
30% probability. Hydrogen atoms (with the exception of H1) are
À
omitted for clarity. Selected bond lengths [Š] and angles [8]: Zr1 C1
À
À
2.222(4), C1 C2 1.414(6), B1 C2 1.647(7); Zr1-C1-C2 103.4(3),
C1-C2-B1 121.9(4).
[Cp*2ZrMe]+[MeB(C6F5)3]À, Marks and co-workers obtained
[Cp*2(Me)Zr(m-F)Zr(Me)Cp*2]+[MeB(C6F5)3]À,[5c] which can
be viewed as an adduct of [Cp*2ZrMe]+[MeB(C6F5)3]À and
[Cp*2Zr(Me)F]. This product is probably formed through
either a transfer of one C6F5 group to the metalcenter, with
the formation of [Cp*2Zr(Me)C6F5] and MeB(C6F5)2, fol-
lowed by an ortho-fluoride abstraction to give [Cp*2Zr(Me)F]
and organoboron species, or through a direct fluoride transfer
to the zirconium centre.
Complex 5, which is formed in 10 to 30% yield depending
on the reaction temperature and time, was shown by X-ray
crystallography to be an isomer of complex C.[4b,16] In 5, the
rac-(ebthi) ligand is substituted by the boranate group at the
2-position of the cyclopentadienyl ring, whereas in C, the ring
is substituted at the 3-position. Additionally, one ortho
fluorine atom of the zirconium-bound C6F5 group weakly
coordinates the zirconium center in 5 (Figure 2).[4b]
Scheme 2. Mechanisms of catalyst activation and deactivation.
LaL=Cp’2.
In relation to these steps, the formation of fluoro
À
complexes by C F bond cleavage was only discussed as a
deactivation process.[5] Our results clearly show that Zr F
À
À
complexes can be reconverted into Zr H complexes, which
À
are very active catalysts, through interaction with Al H
groups (Scheme 2).
It has been noted that “In many instances organometallic
fluorides have been the unexpected products of fluorine
abstraction in the decomposition of complexes of fluorinated
anions.”[11] This observation, in the case of reactions involving
the fluorinated arylborane B(C6F5)3 and the boranates [RB-
(C6F5)3]À (R = Me, H), was previously considered to be a
marginal problem and interpreted exclusively as the result of
a deactivation process.
There are, in fact, many examples of B(C6F5)3 effecting
electrophilic substitution at cyclopentadienyl ligands (leading
to complexes such as B),[3] but few examples of nucleophilic
aromatic substitution involving B(C6F5)3. Erker and co-work-
ers,[12a,b] and others[12c] have described that the positive
inductive effect of boron is sometimes overcompensated in
B(C6F5)3 by the fluoro substituents, making this special Lewis
acid susceptible to nucleophilic attack at the electron-
deficient arene rings.
Figure 2. ORTEP representation of 5. Thermal ellipsoids are set at
30% probability. Hydrogen atoms (with the exception of H1) are
À
omitted for clarity. Selected bond lengths [Š] and angles [8]: Zr1 C13
The unexpected formation of 1 and its activity as a
precatalyst in olefin polymerization, explained herein, con-
À
À
2.319(8), Zr1···F1 2.604(5), F1 C14 1.389(8), B1 C4 1.593(10);
C13-C14-F1 114.0(7).
Angew. Chem. Int. Ed. 2006, 45, 4195 –4198
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4197

Communications
Vagedes, G. Kehr, D. König, K. Wedeking, R. Fröhlich, G.
Erker, C. Mück-Lichtenfeld, S. Grimme, Eur. J. Inorg. Chem.
2002, 2015 – 2021; c) A. N. Chernega, A. J. Graham, M. L. H.
Green, J. Haggitt, J. Lloyd, C. P. Mehnert, N. Metzler, J. Souter,
J. Chem. Soc. Dalton Trans. 1997, 2293 – 2303.
firm the statement that “early transition metal fluoride
complexes do not preclude catalytic chemistry.”[13] In this
context, our results offer a new perspective on the deactiva-
tion of olefin-polymerization catalysts.
[13] J. L. Kiplinger, T. G. Richmond, J. Am. Chem. Soc. 1996, 118,
1805 – 1806.
Received: January 27, 2006
Published online: May 16, 2006
[14] Crystal-structure analysis of 3: C58H51BF15Zr2, Mr = 1226.24,
0.2 ” 0.2 ” 0.1 mm, orange prisms, space group P21/n, monoclinic,
a = 10.946(2), b = 20.243(4), c = 22.602(5) Š, b = 100.86(3)8, V=
À
Keywords: borate ligands · C F activation · homogeneous
.
4918.5(17) Š3, Z = 4, 1calcd = 1.656 gcmÀ3
, 17564 reflections
catalysis · polymerization · zirconocene fluorides
measured, 5159 independent, 3562 observed (I > 2s(I)), R1 =
0.057, wR2(all data) = 0.141, 696 parameters. The terminal
hydride ligand was not located.
[1] Review: H. H. Brintzinger, D. Fischer, R. Mülhaupt, B. Rieger, R.
Waymouth, Angew. Chem. 1995, 107, 1255–1283; Angew. Chem.
Int. Ed. Engl. 1995, 34, 1143–1171, and references therein.
[2] Reviews: a) W. E. Piers, Chem. Eur. J. 1998, 4, 13–18; b) G.
Erker, Acc. Chem. Res. 2001, 34, 309–317, and references therein.
[3] a) U. Rosenthal, V. V. Burlakov, P. Arndt, W. Baumann, A.
Spannenberg, V. B. Shur, Eur. J. Inorg. Chem. 2004, 4739 – 4749,
and references therein; b) V. V. Burlakov, P. Arndt, W. Bau-
mann, A. Spannenberg, U. Rosenthal, Organometallics 2004, 23,
5188 – 5192; c) M. A. Bach, T. Beweries, V. V. Burlakov, P.
Arndt, W. Baumann, A. Spannenberg, U. Rosenthal, W.
Bonrath, Organometallics 2005, 24, 5916 – 5918.
[15] Crystal-structure analysis of 4: C38H25BF14Zr, Mr = 849.61, 0.4 ”
0.3 ” 0.2 mm, yellow prisms, space group P212121, orthorhombic,
a = 11.562(2), b = 12.493(2), c = 22.206(4) Š, V= 3207.5(10) Š3,
Z = 4, 1calcd = 1.759 gcmÀ3, 17145 reflections measured, 5104
independent, 3749 observed (I > 2s(I)), R1 = 0.034, wR2(all
data) = 0.061, 487 parameters.
[16] Crystal-structure analysis of 5: C39.67H24BF15Zr, Mr = 887.62,
¯
0.3 ” 0.2 ” 0.1 mm, colorless prisms, space group P3, trigonal, a =
23.082(3), c = 12.474(3) Š, V= 5755.5(16) Š3, Z = 6, 1calcd
=
1.537 gcmÀ3, 16976 reflections measured, 5992 independent,
2908 observed (I > 2s(I)), R1 = 0.053, wR2(all data) = 0.153, 507
parameters. CCDC-295535 (3), CCDC-295534 (4), and CCDC-
295536 (5) contain the supplementary 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.
[4] a) C. Lefeber, W. Baumann, A. Tillack, R. Kempe, H. Görls, U.
Rosenthal, Organometallics 1996, 15, 3486 – 3490; b) P. Arndt,
W. Baumann, A. Spannenberg, U. Rosenthal, V. V. Burlakov,
V. B. Shur, Angew. Chem. 2003, 115, 1455 – 1458; Angew. Chem.
Int. Ed. 2003, 42, 1414 – 1418.
[5] Examples: a) T. Wondimagegn, Z. Xu, K. Vanka, T. Ziegler,
Organometallics 2005, 24, 2076 – 2085, and references therein;
b) T. Wondimagegn, Z. Xu, K. Vanka, T. Ziegler, Organo-
metallics 2004, 23, 3847 – 3852; c) X. Yang, C. L. Stern, T. J.
Marks, J. Am. Chem. Soc. 1994, 116, 10015 – 10031.
[6] a) F. R. W. P. Wild, L. Zsolnai, G. Huttner, H. H. Brintzinger, J.
Organomet. Chem. 1982, 232, 233 – 247; b) D. Thomas, W.
Baumann, A. Spannenberg, R. Kempe, U. Rosenthal, Organo-
metallics 1998, 17, 2096 – 2102; c) A. Spannenberg, P. Arndt, W.
Baumann, V. V. Burlakov, U. Rosenthal, S. Becke, T. Weiß,
Organometallics 2004, 23, 3819 – 3825.
[7] a) S. Becke, U. Rosenthal, (Bayer AG), DE 199 32 409A1 (18. 01.
2001); b) S. Becke, U. Rosenthal, (Bayer AG), US 6,303,718 B1
(16. 10. 2001); c) S. Becke, U. Rosenthal, P. Arndt, W. Baumann,
A. Spannenberg, (Bayer AG), DE 101 10 227A1 (05. 09. 2002);
d) J. N. Pedeutour, H. Cramail, A. Deffieux, J. Mol. Catal. A 2001,
174, 81–87; e) C. Schwecke, W. Kaminsky, Macromol. Rapid
Commun. 2001, 22, 508–512; f) Y. Qian, H. Zhang, X. Qian, J.
Huang, C. Shen, J. Mol. Catal. A 2003, 192, 25–33.
[8] a) A. Spannenberg, P. Arndt, W. Baumann, V. V. Burlakov, U.
Rosenthal, S. Becke, T. Weiß, Organometallics 2004, 23, 4792 –
4795; b) R. B. Grossmann, R. A. Doyle, S. L. Buchwald, Orga-
nometallics 1991, 10, 1501 – 1505.
[9] a) B. L. Edelbach, B. M. Kraft, W. D. Jones, J. Am. Chem. Soc.
1999, 121, 10327 – 10331; b) B. M. Kraft, R. J. Lachicotte, W. D.
Jones, J. Am. Chem. Soc. 2000, 122, 8559 – 8560; c) B. M. Kraft,
R. J. Lachicotte, W. D. Jones, J. Am. Chem. Soc. 2001, 123,
10973 – 10979; d) B. M. Kraft, W. D. Jones, J. Organomet. Chem.
2002, 658, 132 – 140; e) B. M. Kraft, W. D. Jones, J. Am. Chem.
Soc. 2002, 124, 8681 – 8689; f) B. M. Kraft, R. J. Lachicotte, W. D.
Jones, Organometallics 2002, 21, 727 – 731; g) W. D. Jones,
Dalton Trans. 2003, 3991 – 3995.
[10] T. J. Marks, X. Yang, C. L. Stern, Angew. Chem. 1992, 104, 1406 –
1408; Angew. Chem. Int. Ed. Engl. 1992, 31, 1375 – 1377.
[11] E. F. Murphy, R. Murugavel, H. W. Roesky, Chem. Rev. 1997, 97,
3425 – 3468.
[12] Examples: a) S. Döring, G. Erker, R. Fröhlich, O. Meyer, K.
Bergander, Organometallics 1998, 17, 2183 – 2187; b) D.
4198
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4195 –4198