Stable borate-bridged ansa-zirconocene complexes. Preparation and X-ray crystallographic characterization of [Cp*2Al]+[Me(Ph)B(η5-C 5H4)2ZrCl2]- and [PPN]+[

Article abstract of DOI:10.1021/om990744p

The title complexes are stable, borate-bridged ansa-zirconocene species which have been prepared by reacting [{Ph(SMe₂)B(η⁵-C₅H₄) ₂}ZrCl₂] (1) with Cp*₂-AlMe and [PPN]Cl. Their m

Full text of DOI:10.1021/om990744p

5
432
Organometallics 1999, 18, 5432-5434
Sta ble Bor a te-Br id ged a n sa -Zir con ocen e Com p lexes.
P r ep a r a tion a n d X-r a y Cr ysta llogr a p h ic
+
5
-
Ch a r a cter iza tion of [Cp * Al] [Me(P h )B(η -C H ) Zr Cl ]
2
5
4 2
2
+
5
-
a n d [P P N] [Cl(P h )B(η -C H ) Zr Cl ]
5
4 2
2
Christopher T. Burns, Daniel S. Stelck, Pamela J . Shapiro,* and Ashwani Vij
Department of Chemistry, University of Idaho, Moscow, Idaho 83844-2343
Klaus Kunz and Gerald Kehr
Organisch-Chemisches Institut, Universit a¨ t M u¨ nster, Corrensstrasse 40,
D-48149 M u¨ nster, Germany
Tom Concolino and Arnold L. Rheingold
Department of Chemistry and Biochemistry, University of Delaware, Newark, Delaware 19716
Received September 21, 1999
Summary: The title complexes are stable, borate-bridged
ansa-zirconocene species which have been prepared by
reacting [{Ph(SMe2)B(η -C5H4)2}ZrCl2] (1) with Cp*2-
AlMe and [PPN]Cl. Their molecular structures were
determined by X-ray diffraction. Noteworthy are the
stabilities of the borate-cyclopentadienyl linkages and
the enhanced Lewis acidity of the boron within its
strained ansa position.
One of the drawbacks that we1d,i and others1k have
faced in developing the chemistry of boryl-substituted
metallocenes is the vulnerability of the boryl group to
cleavage from the cyclopentadienyl ring. This typically
occurs in the presence of anionic nucleophiles such as
metal alkyls, which are an essential feature in the
transition-metal chemistry of interest to us. Such
problems with cleavage of the boryl group from the
ligand framework beset our earlier efforts to alkylate
various half-sandwich boryl-cyclopentadienyl titanium
and zirconium complexes and are probably responsible
for the decomposition we have encountered upon trying
to alkylate our boron-bridged ansa-zirconocene com-
5
The covalent attachment of boryl groups to the
cyclopentadienyl or indenyl rings of group 4 metallocene
and half-sandwich complexes is a strategy that has been
1
pursued by ourselves and by others with a variety of
goals in mind, the most popular one being the develop-
ment of well-defined, single-component, zwitterionic
olefin polymerization catalysts as alternatives to the
traditional, two-component systems. Incorporation of
the boryl group in the interannular bridging position
between the cyclopentadienyl rings has been of particu-
lar interest to us, because it offers the additional
prospect of using reversible Lewis base coordination by
the boron bridge to manipulate the geometry of the
metallocene as well as to anchor special functionalities
to the metallocene.
5
plexes. Indeed, whereas treatment of [Ph(SMe2)B(η -
C5H4)2ZrCl2] with alkyllithium, alkyl Grignard, dialkyl-
zinc, or trialkylaluminum reagents leads to decompo-
sition, presumably due to the lability of the dimethyl
sulfide adduct, substitution of the dimethyl sulfide
ligand with a more tightly coordinating trimethylphos-
phine enables alkylation of the zirconium center by
these reagents to cleanly form complexes of the type
5
[
{Ph(PMe3)B(η -C5H4)2}ZrR2] (R ) CH2SiMe3, CH2-
1i
C6H5). Unfortunately, blocking or attenuating the
electrophilicity of the boron bridge is a less than ideal
solution for us, since it defeats some of the anticipated
applications we have for these complexes. We now
report the isolation of stable, borate-bridged ansa-
zirconocene complexes which appear to belie the as-
sumed instability of the borate group to cleavage from
the cyclopentadienyl ring and offer renewed hope for our
developing further the chemistry of these complexes.
(
1) (a) Reetz, M. T.; Br u¨ mmer, H.; Kessler, M.; Kuhnigk, J . Chimia
1
995, 49, 501. (b) Bochmann, M.; Lancaster, S. J .; Robinson, O. B. J .
Chem. Soc., Chem. Commun. 1995, 2081. (c) Rufanov, K. A.; Kotov, V.
V.; Kazennova, N. B.; Lemenovskii, D. A.; Avtomonov, E. V.; Lorberth,
J . J . Organomet. Chem. 1996, 525, 287. (d) Larkin, S. A.; Golden, J .
T.; Shapiro, P. J .; Yap, G. P. A.; Foo, D. M. J .; Rheingold, A. L.
Organometallics 1996, 15, 2393. (e) Ruwwe, J .; Erker, G.; Fr o¨ hlich, R.
Angew. Chem., Int. Ed. Engl. 1996, 35, 80. (f) Sun, Y.; Spence, R. E. v.
H.; Piers, W. E.; Parvez, M.; Yap, G. P. A. J . Am. Chem. Soc. 1997,
1
19, 5132. (g) Duchateau, R.; Lancaster, S. J .; Thornton-Pett, M.;
Bochmann, M. Organometallics 1997, 16, 4995. (h) Rufanov, K.;
Avtomonov, E.; Kazennova, N.; Kotov, V.; Khvorost, A.; Lemenovskii,
D.; Lorberth, J . J . Organomet. Chem. 1997, 536-537, 361. (i) Stelck,
D. S.; Shapiro, P. J .; Basickes, N. Organometallics 1997, 16, 4546. (j)
Rufanov, K.; Avtomonov, E.; Kazennova, N.; Kotov, V.; Khvorost, A.;
Lemenovskii, D.; Lorberth, J . J . Organomet. Chem. 1997, 536-537,
Reaction of the ansa-zirconocene complex [{Ph(PMe3)B-
5
(
η -C5H4)2}ZrCl2] (1) with Cp*2AlMe (Cp* ) C5Me5), a
2
new compound recently developed in our laboratories,
+
5
produced the ionic complex [Cp*2Al] [{Ph(Me)B(η -
3
61. (k) Shafiq, F. A.; Abboud, K. A.; Richardson, D. E.; Boncella, J .
-
3
C5H4)2}ZrCl2] (2), in which a methyl anion from the
M. Organometallics 1998, 17, 982. (l) Lancaster, S. J .; Thornton-Pett,
M.; Dawson, D. M.; Bochmann, M. Organometallics 1998, 17, 3829.
aluminum had been completely transferred to the boron
(m) Braunschweig, H.; von Koblinski, C.; Wang, R. Eur. J . Inorg. Chem.
1
999, 69. (n) Ashe, A. J .; Fang, X.; Kampf, J . W. Organometallics 1999,
8, 2288. (o) Reetz, M. T.; Willuhn, M.; Psiorz, C.; Goddard, R. Chem.
1
(2) Burns, C. T.; Shapiro, P. J . Abstracts of Papers, 216th National
Meeting of the American Chemical Society, Boston, MA; American
Chemical Society: Washington, DC, 1998; INOR 5.
Commun. 1999, 1105. (p) Starzewski, K. A. O.; Kelly, W. M.; Stumpf,
A.; Freitag, D. Angew. Chem., Int. Ed. Engl. 1999, 38, 2439.
1
0.1021/om990744p CCC: $18.00 © 1999 American Chemical Society
Publication on Web 11/30/1999

Communications
Organometallics, Vol. 18, No. 26, 1999 5433
interaction between the aluminocenium cation and a
chlorine ligand on zirconium can also be ruled out since
the closest Al- - -Cl contact is 5.8 Å. Within the ansa-
zirconocene counteranion, the Cp-B-Cp angle of the
borate bridge of 97.2° is considerably narrower than the
corresponding angle in 1 (101.1°) and the trimeth-
5
1i
ylphosphine adduct [Ph(PMe3)B(η -C5H4)2ZrCl2] (100.1°).
The (ring centroid)-Zr-(ring centroid) angle of 120.5°
for 2 is only slightly narrower than that of the other
complexes (121.1 and 121.3°, respectively). The bond
distance between the boron and methyl carbon of 1.626-
(
8) Å is even shorter than the corresponding distance
-
of 1.638(5) Å found for a “free” [MeB(C6F5)3] counter-
anion and is considerably shorter than that of associated
borate counteranions in a series of cationic zirconocene
complexes reported by Marks and co-workers.5 That
complete methyl anion transfer is maintained in solu-
2
7
tion is supported by the Al NMR spectrum of the
F igu r e 1. Molecular structure of 2. Thermal ellipsoids are
drawn at 30% probability. Hydrogen atoms are omitted for
clarity. Selected interatomic distances (Å) and angles
complex, which exhibits a sharp peak at δ -114.5
6
characteristic of the permethylaluminocenium cation.
1
The H NMR spectrum of 2 in CDCl3 exhibits a signal
(
deg): B-C21 ) 1.626(8), B-C32 ) 1.639(8); cent-Zr-
cent ) 120.5, cent-Al-cent ) 178.6, C32-B-C37 ) 97.2-
4).
at δ 0.31 for the borate methyl group, as compared to δ
-
1.58 for the Al-methyl resonance of Cp*2AlMe. There
(
1
3
is a broad signal at δ 10 in the C NMR spectrum of 2
which is partially obscured by the methyl resonance of
the aluminocenium cation. We tentatively assign this
to the methyl carbon of the borate bridge.
bridge (eq 1). Probably due to a competitive equilibrium
The stability of the methylphenylborate bridge in 2
was surprising to us for the reasons mentioned above.
The only decomposition we observe in a room-temper-
ature CDCl3 solution sample of 2 over time involves the
gradual decomposition of the aluminocenium cation via
the elimination of Cp*H, a decomposition which we
2
observe with other decamethylaluminocenium salts. In
contrast, a similar reaction between 1 and Cp2AlMe in
1
CDCl3 monitored by H NMR spectroscopy revealed
with the dimethyl sulfide released in the reaction,
approximately 2 equiv of Cp*2AlMe is required to drive
the reaction to completion. The remaining, unreacted
Cp*2AlMe is then washed away from 2 with petroleum
ether. The molecular structure of the ion pair was
determined by single-crystal X-ray diffraction methods
and is shown in Figure 1.4
complete conversion of the reactants to an ill-defined
mixture of products. Unlike 1, complex 2 is activated
by an excess of methylalumoxane toward the polymer-
ization of ethylene with an activity of 902 kg of PE/((mol
5
of catalyst) bar h) (cf. [{(Ph)(PMe3)B(η -C5H4)2}ZrCl2],
which has an activity of 1664 kg of PE/((mol of catalyst)
7
bar h)).
The structure of the rather exotic aluminocenium
countercation is similar to that determined previously
We attribute the stability of 2 to the bulky, nonin-
trusive nature of the decamethylaluminocenium coun-
tercation. This countercation effect could perhaps also
explain why [Li(THF)4][C5H5B(C6F5)3] and [NEt4][C5H5B-
(C6F5)3] are successfully zirconated to form the corre-
sponding borato-cyclopentadienyl zirconium complexes,1l
whereas attempted zirconation of Li[1,3-Me3SiC5H4B-
6
by Schn o¨ ckel and co-workers for [Cp*2Al][Cp*AlCl3].
The absence of any residual interactions between the
aluminocenium cation and the methyl anion in the solid
state is evidenced by the fact that the nearest Al- - -
C(21) contact is nearly 6 Å. The possibility of any
1
6 5 HH
H J )
B, 3
(
3) H NMR (500.13 MHz, CDCl
3
): δ 7.77 (d, 2H, o-C
B), 7.06 (pseudotriplet, 1H,
B), 6.62-6.64 (two overlapping m, 4H, (C B), 5.84 (m, 2H,
)B), 5.65 (m, 2H, (C )B), 2.19 (s, 30 H, [C (CH Al, 0.31 (s,
B). 13C{ H} NMR: δ 136.4 (br, C
B-ipso), 134.5, 127.0, 126.2
B), 124.2, 123.3, 115.0, 114.8 ((C B), 118.8 (C (CH ), 10.7
)). Al NMR (vs external Al(OH) ): δ -114.5.
): δ -31.6. Anal. Calcd for C37 46AlBCl
(
C6F5)3] led instead to the elimination of Li[MeB-
7
Hz), 7.23 (pseudo-triplet, 2H, m-C
p-C
6 5
H
1
k
5
H
4
5
5
H
5
4
)
2
(C6F5)3]. Guided by this hypothesis, we have since pre-
pared a variety of other stable borate-bridged ansa-
zirconcene species by reacting 1 with salts of other bulky
cations such as tetrabutylammonium and bis(triphen-
ylphosphoranylidene)ammonium ([PPN] ) [(Ph2P)2N] ).
One of these salts, [PPN] [{Ph(Cl)B(η -C5H4)2}ZrCl2]
(C
5
H
5
H
4
3 5 2
) ]
1
3
H, CH
3
6 5
H
(
C
6
H
5
5
H
4
)
2
5
3 5
)
2
7
(
C
5
(CH
3
)
5
, 10 (br, B(CH
3
3
1
1
B NMR (vs external B(OH)
Zr: C, 64.34; H, 6.71. Found: C, 62.39; H, 6.54. The carbon analysis
was consistently low. We attribute this to Cp*
which we were unable to completely eliminate.
3
H
2
-
+
+
+
+
5
-
2
Al decomposition,
(
4) Crystal data: C37
2 1
H46AlBCl Zr; space group P2 /n; monoclinic; a
)
13.961(2) Å, b ) 17.058(3) Å, c ) 14.639(2) Å, â ) 93.158(10)° at
(6) Dohmeier, C.; Schn o¨ ckel, H.; Robl, C.; Schneider, U.; Ahlrichs,
R. Angew. Chem., Int. Ed. Engl. 1993, 32, 1655.
(7) The ethylene polymerizations were performed as described in:
Duda, L.; Erker, G.; Fr o¨ hlich, R.; Zippel, F. Eur. J . Inorg. Chem. 1998,
1153.
3
2
13 K; V ) 3481.0(9) Å ; Z ) 4; final R indices (I > 2σ(I)) R1 ) 0.0669,
wR2 ) 0.1265 for 7852 independent reflections.
5) Yang, X.; Stern, C. L.; Marks, T. J . J . Am. Chem. Soc. 1994, 116,
0015.
(
1

5
434 Organometallics, Vol. 18, No. 26, 1999
Communications
comparable to B(C6F5)3 in its reactivity. To assess the
effective Lewis acidity of the boron bridge in base free
[PhB(C5H4)2ZrCl2], 1 was combined with an equimolar
amount of B(C6F5)3 in CD2Cl2 and the ratio of base-free
B(C6F5)3 to (C6F5)3B(SMe2) was determined by 19F NMR.
Interestingly, the ratio of base-free to coordinated
B(C6F5)3 was found to be 1.7:1 at 273 K, indicating that
the boron bridge of the ansa-zirconocene has a higher
affinity for the dimethyl sulfide ligand than B(C6F5)3.
Similarly, combination of [{Ph(tBuNC)B(C5H4)2}ZrCl2]
with a equimolar amount of B(C6F5)3 produced a 2.1:1
ratio of base-free to coordinated B(C6F5)3 at 263 K.11 The
combined results suggest that the “internal strain”
2
experienced by the sp -hybridized boron atom incorpo-
rated into the bridging position of an ansa-zirconocene
complex raises the Lewis acidity of the boron bridge to
a level comparable to that of B(C6F5)3.
F igu r e 2. Molecular structure of the ansa-zirconocene
anion of one of two independent molecules of 3 in the unit
cell. Thermal ellipsoids are drawn at 30% probability.
Hydrogen atoms are omitted for clarity. Selected inter-
atomic distances (Å) and angles (deg): B(1A)-Cl(3A) )
In summary, the important lessons that we have
learned from the isolation of the title complexes are (a)
that a borate-derivatized cyclopentadienyl ring on a
transition metal can be stable in the presence of a
suitably bulky, noninteracting countercation and (b)
that the Lewis acidity of a boron species can be
enhanced substantially by introducing strain in the
molecule which can be relieved through the coordination
1
.937(5), B(1A)-C(8A) ) 1.634(7); average cent-Zr-cent
)
120.9, C(4A)-B(1A)-C(8A) ) 99.4(4).
8
(
3), has also been crystallographically characterized,
and an ORTEP drawing of one of two unique ansa-
zirconocene anions in the unit cell is shown in Figure
9
2
2
. The Cp-B-Cp angle is 99.4(4)°, between that of
of Lewis bases which rehybridize the boron from sp to
3
compounds 1 and 2. An average (ring centroid)-Zr-
ring centroid) angle of 120.5° for the two unique
sp . We are using these lessons to develop additional
(
examples of borate-bridged ansa-zirconocene complexes
and to aid us in our pursuit of useful catalysis with these
complexes.
molecules is comparable to that of 2. This compound is
stable indefinitely under an N2 atmosphere in the solid
state and in CDCl3 solution. It is also activated by
methylalumoxane toward ethylene polymerization, ex-
hibiting an activity of 574 kg of PE/((mol of catalyst)
bar h).
Ack n ow led gm en t. Support of this work by the NSF
Idaho EPSCoR program (Grant No. EPS 9350539) and
a University of Idaho Seed grant are gratefully acknowl-
edged. P.J .S. is also grateful to the Alexander von
Humboldt Foundation for a fellowship supporting a six-
month sabbatical in the research group of Prof. Dr.
Gerhard Erker, during which some of the results
reported in this paper were obtained.
Another relevant aspect of the formation of 2 is the
ability of the boron bridge of the ansa-zirconocene
complex to abstract a methanide ion from Cp*2AlMe.
On the basis of the CtN IR stretching frequences for
5
the tert-butyl isocyanide adducts [{Ph(t-BuNC)B(η -
-
1
-1
Su p p or tin g In for m a tion Ava ila ble: Text giving experi-
C5H4)2}ZrCl2] (2272 cm ), Ph3B(CN-t-Bu) (2272 cm ),
-
1
10
mental descriptions for the synthesis of Cp*
figure giving the 19F NMR spectrum of a 1:1 mixture of 1 and
B(C in CD Cl , and tables of crystal data, thermal
2
AlMe, 2, and 3, a
and (C6F5)3B(CN-t-Bu) (2310 cm ), one would expect
the reactivity of the phenylboron bridge to be similar
to that of BPh3. However, BPh3 is unreactive toward
Cp*2AlMe, as determined from the absence of either H
or Al NMR spectral evidence for the formation of [Cp*2-
6
F
5
)
3
2
2
parameters, bond distances, bond angles, and atomic coordi-
nates for complexes 2 and 3. This material is available free of
charge via the Internet at http://pubs.acs.org.
1
2
7
+
Al] when Cp*2AlMe is combined with triphenylborane
OM990744P
in CDCl3. In constrast, the phenylboron bridge is
(
11) We cannot say with certainty that these ratios represent true
equilibrium values, since we are, as yet, unable to approach the
equilibrium from the opposite direction by combining either (C B-
(SMe ) or (C B(CN-t-Bu) with base-free [{PhB(C }ZrCl ]. This
is because we have not yet successfully isolated the base-free ansa-
1
(
8) H NMR (200 MHz, CDCl
3
): δ 7.82-7.86 (m, 2H, C
B), 7.21-7.45 (m, 30H, N[P(C ), 6.57-6.64 (m,
B), 6.10-6.14 (m, 2H, (C B), 5.44-5.48 (m, 2H,
): δ 134.2, 132.3, 129.9
B), 125.5, 125.0, 115.7, 113.2
6 5
H B), 7.55-
7
4
.63 (m, 3H, C
H, (C
6
H
5
6
H
5
)
3
]
2
6 5 3
F )
2
5
H
4
)
2
5
H
4
)
3
2
2
6
F
5
)
3
5
H
4
)
2
1
3
1
(
(
(
C
C
5
H
4
)
)
4
2
B). C{ H} NMR (96 MHz, CDCl
P), 127.9, 127.4, 126.5 (C
B). 11B NMR (vs external H
PO ): δ 21.2. Anal. Calcd for C52
H, 4.43; N, 1.33. Found: C, 65.41; H, 4.43; N, 1.33.
9) Crystal data: C52 45BCl NP Zr; space group P 1h ; monoclinic; a
13.6573(3) Å, b ) 15.6301(3) Å, c ) 23.2356(4) Å, R ) 102.4478(6)°,
1
h,i
1
6
(C
H
5
5
3
)
6
H
5
zirconocene species.
pretation of the F spectra, and we see the appearance of a new set of
AB pseudo-triplets at δ 7.13 and 6.6 in the H NMR spectra, which we
attribute to the cyclopentadienyl protons of the base-free ansa-
zirconocene. There is also some evidence of decomposition of the ansa-
zirconocene to form a B-Cp cleavage product, which we believe arises
due to the instability of the base-free ansa-zirconocene complex. Thanks
The H spectra are consistent with our inter-
31
19
H
2
3
BO
3
): δ -22.1. P NMR (vs
1
external H
3
4
H
43BCl NP Zr: C, 65.59;
3
2
(
H
3
2
)
3
â ) 94.3487(7)°, γ ) 105.3451(6)° at 173 K; V ) 4561.74(22) Å ; Z ) 4
with two independent, chemically equivalent molecules in the unit cell;
final R indices (I > 2σ(I)) R1 ) 0.0645, wR2 ) 0.1734 for 19 571
independent reflections.
to a reviewer’s suggestion, we have verified that (C
(C B(CN-t-Bu) do not dissociate to any detectable amounts (<1%)
of free (C B and ligand under similar experimental conditions.
Hence, the ratios of free and coordinated (C B that we observe in
our competition experiments do not simply reflect the dissociation
equilibria of the (C B adducts.
6 5 3 2
F ) B(SMe ) and
6
5 3
F )
6
5 3
F )
(
10) J acobson, H.; Berke, H.; D o¨ ring, S.; Kehr, G.; Erker, G.;
6 5 3
F )
Fr o¨ hlich, R.; Meyer, O. Organometallics 1999, 18, 1724. This reference
also reports a ν(CtN) band for free tert-butyl isocyanide of 2140 cm-
1
.
6 5 3
F )