Bifunctional Lewis acids. Synthesis and olefin polymerization chemistry of the 1,1-Di[bis(perfluorophenyl)boryl]alkenes RCH=C[B(C6F5)2]2 (R = t-Bu, C6H5, C6F5)

Article abstract of DOI:10.1021/om980259e

The synthesis and characterization, including crystallographic analysis, of the bifunctional boranes RCH=C[B(C₆F₅)₂]₂ (R = t-C₄H₉, 1a; C₆H₅, 1b; C₆F₅, 1c) by regioselective hydroboration of the corresponding 1-boraalkynes using HB(C₆F₅)₂ are reported herein. Compounds 1a and 1b have been screened as cocatalysts for ethylene polymerization in the presence of Cp₂ZrMe₂ (3) under a variety of conditions. NMR spectroscopic studies indicate that Cp₂-Zr[η²-Bu^tC=CB(C₆F ₅)₂] (4a), Cp₂ZrMe(C₆F₅), the organoborane Me₂BC₆F₅, and methane gas are the final products formed from reaction of 1a with 3 in toluene solution at room temperature. The stoichiometric mechanism for this transformation has been elucidated through variable-temperature NMR studies. Complex 4a and MeB(C₆F₅)₂ (7) were prepared independently and screened as ethylene polymerization catalysts and cocatalysts, respectively. Compound 4a is inactive for ethylene polymerization, either alone or in the presence of additional 1a. However, the combination of Cp₂ZrMe₂ and 7 gives rise to the species [Cp₂ZrMe]⁺[Me₂B(C₆F ₅)₂]^- 8), which although unstable at room temperature in solution (decomposing over a period of 60 min to give Cp₂ZrMe(C₆F₅) and the organoborane Me₂-BC₆F₅), is active for ethylene polymerization. From a comparison of activity and MW data, it is concluded that the putative ion pairs formed from 1a (or 1b) and 3 lack sufficient thermal stability at conventional polymerization temperatures and that the polymerization activity observed can be interpreted as arising from species 8.

Full text of DOI:10.1021/om980259e

Organometallics 1998, 17, 3557-3566
3557
Bifu n ction a l Lew is Acid s. Syn th esis a n d Olefin
P olym er iza tion Ch em istr y of th e
1,1-Di[bis(p er flu or op h en yl)bor yl]a lk en es
RCHdC[B(C6F 5)2]2 (R ) t-Bu , C6H5, C6F 5)
Katrin Ko¨hler,† Warren E. Piers,*,† Adam P. J arvis,‡ S. Xin,‡ Y. Feng,‡
A. M. Bravakis,‡ Scott Collins,*,‡ William Clegg,§ Glenn P. A. Yap,| and
Todd B. Marder
Department of Chemistry, University of Calgary, 2500 University Drive Northwest,
Calgary, Alberta, Canada T2N 1N4, Department of Chemistry, University of Waterloo,
Waterloo, Ontario, Canada N2L 3G1, Department of Chemistry, University of Newcastle,
Newcastle upon Tyne NE1 7RU, U.K., Windsor Molecular Structure Center, Department of
Chemistry and Biochemistry, University of Windsor, Windsor, Ontario, Canada N9B 3P4, and
Department of Chemistry, South Road, University of Durham, Durham DH1 3LE, U.K.
Received April 6, 1998
The synthesis and characterization, including crystallographic analysis, of the bifunctional
boranes RCHdC[B(C6F5)2]2 (R ) t-C4H9, 1a ; C6H5, 1b; C6F5, 1c) by regioselective hydrobo-
ration of the corresponding 1-boraalkynes using HB(C6F5)2 are reported herein. Compounds
1a and 1b have been screened as cocatalysts for ethylene polymerization in the presence of
Cp2ZrMe2 (3) under a variety of conditions. NMR spectroscopic studies indicate that Cp2-
Zr[η2-ButCtCB(C6F5)2] (4a ), Cp2ZrMe(C6F5), the organoborane Me2BC6F5, and methane gas
are the final products formed from reaction of 1a with 3 in toluene solution at room
temperature. The stoichiometric mechanism for this transformation has been elucidated
through variable-temperature NMR studies. Complex 4a and MeB(C6F5)2 (7) were prepared
independently and screened as ethylene polymerization catalysts and cocatalysts, respec-
tively. Compound 4a is inactive for ethylene polymerization, either alone or in the presence
of additional 1a . However, the combination of Cp2ZrMe2 and 7 gives rise to the species
[Cp2ZrMe]+[Me2B(C6F5)2]- (8), which although unstable at room temperature in solution
(decomposing over a period of 60 min to give Cp2ZrMe(C6F5) and the organoborane Me2-
BC6F5), is active for ethylene polymerization. From a comparison of activity and MW data,
it is concluded that the putative ion pairs formed from 1a (or 1b) and 3 lack sufficient thermal
stability at conventional polymerization temperatures and that the polymerization activity
observed can be interpreted as arising from species 8.
In tr od u ction
employed and perhaps least well-understood.3 There
is general agreement that MAO serves as an alkylating
agent (in the case of metallocene dichlorides) and also
serves to ionize the metallocene dialkyl complex to form
the active, cationic alkylzirconocene complex.3b,c,4
Conceptually related, Lewis-acidic, single-component
cocatalysts have been developed and are used in com-
bination with metallocene dialkyl complexes (either
preformed or generated in situ); prototypical examples
There has been extensive study into the development
of highly active homogeneous catalytic systems for olefin
polymerization over the last 15 years, with group 4
metallocene complexes figuring prominently. While,
understandably, considerable attention has focused on
the structure of the metallocene complex and its influ-
ence on polymerization activity and polymer properties,1
it is now becoming increasingly evident that the nature
of the cocatalyst employed in olefin polymerization can
have equally dramatic effects on these features.2
Of the various cocatalysts used in olefin polymeriza-
tion, methylaluminoxane (MAO) is the most widely
(2) (a) Deck, P. A.; Beswick, C. L.; Marks, T. J . J . Am. Chem. Soc.
1998, 120, 1772. (b) Chen, Y.-X.; Stern, C. L.; Marks, T. J . J . Am. Chem.
Soc. 1997, 119, 2582. (c) Chen, Y.-X.; Stern, C. L.; Yang, S.; Marks, T.
J . J . Am. Chem. Soc. 1996, 118, 12451. (d) Giardello, M. A.; Eisen, M.
S.; Stern, C. L.; Marks, T. J . J . Am. Chem. Soc. 1995, 117, 12114. (e)
Hlatky, G. G.; Eckman, R. R.; Turner, H. W. Organometallics 1992,
11, 1413. (f) Yang, X.; Stern, C. L.; Marks, T. J . Organometallics 1991,
10, 840. (g) Hlatky, G. G.; Turner, H. W.; Eckman, R. R. J . Am. Chem.
Soc. 1989, 111, 2728.
(3) (a) For a recent review see: Reddy, S. S.; Sivaram, S. Prog.
Polym. Sci. 1995, 20, 309. See also: (b) Harlan, C. J .; Bott, S. G.;
Barron, A. R. J . Am. Chem. Soc. 1995, 117, 6465. (c) Harlan, C. J .;
Mason, M. R.; Barron, A. R. Organometallics 1994, 13, 2957. (d)
Pasynkiewicz, S. Polyhedron 1990, 9, 429.
† University of Calgary.
‡ University of Waterloo.
§ University of Newcastle.
| University of Windsor.
University of Durham.
(1) For recent reviews see (a) Kaminsky, W.; Arndt, M. Adv. Polym.
Sci. 1997, 127, 144. (b) Bochmann, M. J . Chem. Soc., Dalton Trans.
1996, 255. (c) Brintzinger, H. H.; Fischer, D.; Mu¨lhaupt, R.; Rieger,
B.; Waymouth, R. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1143. (c)
Mo¨hring, P. C.; Coville, N. J . J . Organomet. Chem. 1994, 479, 1.
(4) (a) Sishta, C.; Hathorn, R. M.; Marks, T. J . J . Am. Chem. Soc.
1992, 114, 1112. (b) Gassman, P. G.; Callstrom, M. R. J . Am. Chem.
Soc. 1987, 109, 7875.
S0276-7333(98)00259-3 CCC: $15.00 © 1998 American Chemical Society
Publication on Web 07/15/1998

3558 Organometallics, Vol. 17, No. 16, 1998
Ko¨hler et al.
include [Ph3C]+[B(C6F5)4]- 5 and B(C6F5)3.6,7 It is widely
accepted that the role of these cocatalysts is to generate
the highly active, coordinatively unsaturated, 14-
electron species [Cp2MR]+ through alkide abstraction.
However, it is clear that even noncoordinating counte-
rions8 can still interact with the metal center in a
manner which modulates polymerization activity and
possibly the rates of chain transfer in olefin polymeri-
zation.
Sch em e 1
Marks and co-workers have studied the relationship
between cation-anion ion-pairing and catalytic activity
in zirconocene complexes activated by B(C6F5)3.6 In
these cases, the strength of the Cp2Zr+Me(µ-Me)B--
(C6F5)3 interaction, as revealed through X-ray diffraction
studies of these ion pairs, has a decisive influence on
polymerization activity and polymer molecular weight
(MW). Even less-coordinating counterions such as
[B(C6F5)4] can still interact with the metal center
through formation of a Zr-F bridge, thus affecting
catalytic activity and stability.9
Thus, the search for even less-coordinating counteri-
ons seems worthwhile from the point of view of tuning
polymerization activity, catalyst stability, and other
properties. In this connection, Marks and co-workers
reported a number of years ago that the borohydride
anion [ButCH2CH{B(C6F5)2}2(µ-H)] as its tributylam-
monium salt was an active cocatalyst for ethylene
polymerization.10 Despite this report, much less atten-
tion has been given to bis-boryl compounds incorporat-
ing B(C6F5)2 units (I) which, by analogy to B(C6F5)3,
might also be useful cocatalysts for ethylene polymer-
ization.11 In the resulting counteranion, the negative
Hydroboration of terminal alkynes with the highly
electrophilic borane HB(C6F5)2 offers a convenient route
to a variety of fully saturated bidentate boranes12 with
a one-carbon linker, i.e., RCH2CH[B(C6F5)2]2. Unfor-
tunately, these compounds tend to undergo facile ret-
rohydroboration and their reaction chemistry can be
dominated by the small amounts of HB(C6F5)2 present
as a result.13 To be effective cocatalysts for olefin
polymerization, their conversion to a hydridoborate
anion is necessary;10 an alternative approach would be
to eliminate the possibility of retrohydroboration. Herein,
we describe the synthesis and characterization of the
unsaturated diboranes RCHdC[B(C6F5)2]2 (R ) t-C4H9,
1a ; C6H5, 1b; C6F5, 1c) which do not undergo â-elimina-
tion. Their reactions with Cp2ZrMe2 and efficacy as
catalyst activators are also discussed.
Resu lts a n d Discu ssion
The diboranes 1 were synthesized by the common
route shown in Scheme 1. Bis(alkynyl)dimethyltin
compounds were prepared from in situ generated lithium
acetylide and dimethyltin dichloride and purified prior
to use in a transmetalation reaction with ClB(C6F5)2.
These reactions, carried out at -78 °C, yielded the
alkynyl boranes 2, which could be isolated (2a , 85%;14
2b, 75%) or used in situ in the next step of the sequence.
As pure solids, the alkynyl boranes are stable for several
weeks if stored at -35 °C but decompose slowly (days)
in solution. Many alkynyl boranes with less electron-
withdrawing substituents have been prepared,15 but
charge is potentially delocalized over more atoms,
reducing the Coulombic interactions between the cation
and anion in a catalytically active ion pair and giving a
more electron-deficient, more reactive cation. Variation
in the nature of the backbone linking the two Lewis-
acid centers offers the opportunity to tune the chelating
properties of the activator as well as the extent of
delocalization of charge.
(5) (a) Bochmann, M.; Lancaster, S. J . Organometallics 1993, 12,
633. (b) Bochmann, M.; Lancaster, S. J . J . Organomet. Chem. 1992,
434, C1. (c) Chien, J . C. W.; Tsai, W.-M.; Rausch, M. D J . Am. Chem.
Soc. 1991, 113, 8570. (d) Turner, H. W. Eur. Pat. Appl. EP 0277004,
1988.
(6) (a) Yang, X.; Stern, C. L.; Marks, T. J . J . Am. Chem. Soc. 1994,
116, 10015. (b) Yang, X.; Stern, C. L.; Marks, T. J . J . Am. Chem. Soc.
1991, 113, 3623. For work on related boranes, see ref 2c.
(7) Piers, W. E.; Chivers, T. Chem. Soc. Rev. 1997, 345.
(8) For a review, see: Strauss, S. H. Chem. Rev. 1993, 93, 927.
(9) (a) J ia, L.; Yang, X.; Stern, C. L.; Marks, T. J . Organometallics
1997, 16, 842. (b) Temme, B.; Erker, G.; Karl, J .; Luftmann, H.;
Fro¨hlich, R.; Kotila, S. Angew. Chem., Int. Ed. Engl. 1995, 34, 1755.
We have also noted binding of [B(C6F5)4]- to cationic methylzirconocene
complexes, even in polar solvents (e.g., toluene:chlorobenzene mixtures)
at low temperatures (i.e., Zr-Me signals coupled to F), although
solvent-separated, ion-pairs predominate under these conditions. Xin,
S.; Collins, S. manuscript in preparation.
(10) (a) J ia, L.; Yang, X.; Stern, C.; Marks, T. J . Organometallics
1994, 13, 3755. (b) Marks, T. J .; J ia, L.; Yang, X. U.S. Patent 5,447,895,
1995 (Northwestern University). (c) Galsworthy, J . R.; Green, M. L.
H.; Williams, V. C.; Chernega, A. N. Polyhedron 1998, 17, 119.
(11) Marks and co-workers have also noted that the organodiborane,
ButCH2CH[B(C6F5)2]2, was an effective cocatalyst for ethylene polym-
erization but indicated that this material was unstable with respect
to formation of B(C6F5)3.9a
(12) Parks, D. J .; Spence, R. E. v H.; Piers, W. E. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 809.
(13) For example, HB(C6F5)2 is highly reactive toward neutral bis-
(alkyl)zirconocenes, see: (a) Spence, R. E. v H.; Parks, D. J .; Piers, W.
E.; MacDonald, M.; Zaworotko, M. J .; Rettig, S. J . Angew. Chem., Int.
Ed. Engl. 1995, 34, 1230. (b) Sun, Y.; Piers, W. E.; Rettig, S. J .
Organometallics 1996, 15, 4110. (c) Sun, Y.; Spence R. E. v H.; Piers,
W. E.; Parvez, M.; Yap, G. P. A. J . Am. Chem. Soc. 1997, 119, 5132.
(d) Piers, W. E. Chem. Eur. J . 1998, 4, 13.
(14) An X-ray structural analysis of 2a established the connectivity
and general features of the compound’s structure; however, severe
disorder precluded adequate refinement of the details associated with
the structure.
(15) For examples, see: (a) Wrackmeyer, B.; No¨th, H. Chem. Ber.
1977, 110, 1086. (b) Wrackmeyer, B. Z. Naturforsch. 1982, 37b, 788.
(c) Maderna, A.; Pritzkow, H.; Siebert, W. Angew. Chem., Int. Ed. Engl.
1996, 35, 1501. (d) Schulz, H.; Gabbert, G.; Pritzkow, H.; Siebert, W.
Chem. Ber. 1993, 126, 1593. (e) Curtis, M. A.; Mu¨ller, T.; Beez, V.;
Pritzkow, H.; Siebert, W.; Grimes, R. N. Inorg. Chem. 1997, 36, 3602.
(f) Wrackmeyer, B.; Schanz, H.-J .; Milius, W. Angew. Chem. 1997, 109,
1145. (g) Meller, A.; Maringgele, W.; Elter, G.; Bromm, D.; Noltemeyer,
M.; Sheldrick, G. M. Chem. Ber. 1987, 120, 1437. (h) Ko¨ster, R.; Boese,
R.; Wrackmeyer, B.; Schanz, H.-J . J . Chem. Soc., Chem. Commun.
1995, 1691.

Bifunctional Lewis Acids
Organometallics, Vol. 17, No. 16, 1998 3559
those with more electrophilic boron centers tend not to
be isolable.16 Indeed, compound 2c was less stable than
the other two derivatives and was best used in situ for
the next step of the synthesis.
Hydroboration of compounds 2 with 1 equiv of HB-
(C6F5)2 yielded the 1,1-alkenyl diboranes 1 in good to
excellent yields. On electronic grounds, this hydrobo-
ration was predicted to give the geminal-substituted
products 1 as shown; indeed, the reactions proceed with
>95% regioselectivity. The diboranes 1 were initially
formulated as the 1,1-substituted alkenyl derivatives on
the basis of their NMR spectra; this was subsequently
confirmed crystallographically (vide infra). Resonances
for the dCH moiety in both the 1H and 13C NMR spectra
are shifted downfield from values typical of olefinic
carbons and hydrogens due to the electron-withdrawing
B(C6F5)2 groups. For 1a , these signals appear at 7.16
and 183.6 ppm, respectively. Both signals are slightly
broadened by the quadrupolar boron nuclei, an effect
which precluded detection of the olefinic carbon directly
attached to the boron atoms. The most convincing
evidence for geminal substitution came in the form of
NOE experiments on 1a , which exhibited strong en-
hancements between the olefinic proton and the tert-
butyl group. Such observations would not be expected
if these groups were trans disposed about the carbon-
carbon double bond in a 1,2-substituted product.
F igu r e 1. ORTEP diagram of 1a (thermal ellipsoids are
at the 50% probability level).
Quadrupolar broadening also had a significant impact
on the 11B NMR data; spectra for each diborane 1 were
extremely broad, and the inequivalent boron nuclei
could not be distinguished at the field employed. Al-
though no fine structure was observed in these reso-
nances, scalar coupling between the chemically inequiv-
alent boron nuclei also likely contributes to the broadness
of the signals. The resonances observed for 1a (39.9 (
1.0), 1b (62.7 ( 1.0), and 1c (61.8 ( 1.0) are all, however,
in the region expected for neutral, three-coordinate
boranes.17 Fluorine-19 NMR spectroscopic data is more
useful in distinguishing the two different -B(C6F5)2
moieties in that two sets of ortho, para, and meta
resonances are observed for each compound at room
temperature. For 1c, a third set of resonances for the
olefinic C6F5 group is also observed.
F igu r e 2. ORTEP diagram of 1b (thermal ellipsoids are
at the 50% probability level). Unlabeled carbon atoms have
the same number as the corresponding F atom.
X-r a y Str u ctu r es of 1a -c. The 1,1-diboryl substi-
tution pattern for each of the diboranes 1a -c was
confirmed via X-ray crystallography; the molecular
structures are shown in Figures 1-3, while selected
metrical parameters are given in Table 1. Note that
the numbering scheme for 1a is slightly different than
that for 1b and 1c; Table 1 is constructed so that each
entry refers to the same parameter for each structure.
In all three structures, the boron center trans to the
R substituent lies essentially in conjugation with the
CdC bond, i.e., the boron trigonal plane is roughly
coincident with the olefin plane. This is approximated
by the first pair of torsion angles listed in Table 1. For
phenyl-substituted 1b, these parameters suggest that
the Btrans trigonal plane is rotated ca. 4° out of copla-
narity with the olefin plane while for 1a and 1c with
their bulkier substitutions the deviation from coplanar-
ity is more severe at ca. 15-19°. Nonetheless, the trans
-B(C6F5)2 moiety’s effect on the C(1)-C(2) lengths in
these compounds is apparent in that each exhibits Cd
C bonds elongated by 0.02-0.03 Å over the typical Cd
C distance of 1.335 Å.18 The longest CdC bond is
observed in 1b, where conjugation is most pronounced.
Conversely, the cis boron centers are rotated out of
conjugation with CdC (see the second pair of torsion
angles listed) to avoid severe steric interactions with
the olefinic R group. Another measure of the steric
crowding in these compounds is the ca. 10-12° of
twisting about the CdC double bonds seen in each
compound, as illustrated by the third pair of torsion
angles given.
The congestion about the CdC double bond is also
manifested by the angles about C(1) and C(2) which
stray from ideal values of 120°. This is particularly true
for 1a , in which the C(2)-C(1)-Bcis and C(2)-C(1)-
(16) Leung, S.-W.; Singleton, D. A. J . Org. Chem. 1997, 62, 1955
and references therein.
(17) Kidd, R. G. In NMR of Newly Accessible Nuclei; Laszlo, P., Ed.;
Academic Press: New York, 1983; Vol. 2.
(18) Gordon, A. J .; Ford, R. A. The Chemist’s Companion; Wiley and
Sons: New York, 1972; p 108.

3560 Organometallics, Vol. 17, No. 16, 1998
Ko¨hler et al.
and allowed to react under gradually warming condi-
tions, clean formation of the product mixture shown is
observed.
Methane was identified by its characteristic proton
1
chemical shift in C6D6 of 0.16 ppm, while H, 11B, and
19F NMR data clearly indicate the presence of dimeth-
ylpentafluorophenylborane (see Experimental Section).
Of the two zirconium-containing products, Cp2Zr(C6F5)-
CH3 is implicated by the characteristic triplet at 0.32
ppm (5J HF ) 3.9 Hz) for the methyl group along with
19F NMR data consistent with the presence of a Zr-
C6F5 moiety. Similar data was observed for a related
derivative, [1,3-(SiMe3)2C5H3]2Zr(C6F5)CH3, reported by
Marks et al.6a The other organometallic species present
in the same molar ratio as Cp2Zr(C6F5)CH3 we formu-
late as the intriguing zirconocene alkyne complex Cp2-
Zr[η2-ButCtCB(C6F5)2], 4a . Although the intimate
structural details of this compound have yet to be
determined, the spectroscopic and analytical data ob-
tained are consistent with the gross features depicted.
Furthermore, this compound could be synthesized sepa-
rately and isolated analytically pure by reaction of the
boryl acetylide 2a with [Cp2Zr(H)CH3]n as shown in
Scheme 3. This reaction is accompanied by methane
evolution, which likely occurs from a boravinyl methyl
zirconocene21 formed by regioselective hydrozirconation
of 2a ; this intermediate was not observed.
F igu r e 3. ORTEP diagram of 1c (thermal ellipsoids are
at the 50% probability level). Unlabeled carbon atoms have
the same number as the corresponding F atom.
Btrans angles are 125.8(5)° and 115.4(5)°, respectively,
and C(1)-C(2)-C(3) is 131.4(5)°. In the aryl-substitut-
ed derivatives 1b-c, the interactions between R and
B
cis(C6F5)2 are less severe because of the aryl groups’
The reaction of Cp2ZrMe2 with 1a (in a 2:1 ratio) was
carried out at low temperature in an NMR tube and
ability to rotate out of conjugation with the CdC double
bond by 34.4° and 40.3°, respectively (Table 1). In these
two compounds, the Bcis-C(1)-Btrans angles are larger
than that found in 1a while the C(2)-C(1)-B angles
are closer to normal values.
1
monitored by both H and 19F NMR spectroscopy. The
results of these experiments are consistent with the
chemistry shown in Scheme 4. At - 60 °C, a rapid
reaction between the two reactants occurs, forming the
product 5a , which is stable for several hours under these
conditions. This ion pair, containing a binuclear cation,
is very similar to that formed from various dimethyl
zirconocenes and the sterically bulky fluoroborane tris-
(2,2′,2′′-perfluorobiphenyl)borane as reported by Marks
et al.2c,22 The dimeric cation is implicated by the
observed stoichiometry of the reaction along with signals
in the 1H NMR spectrum at -0.10 and -1.23 ppm
(found in a 2:1 ratio) which are characteristic of the
terminal and bridging methyl groups.22 Dinuclear
cations such as this form when the counterion is a
poorer base than dimethylzirconocene itself and attest
to 1a ’s ability to form a very weakly coordinating
counterion.
The steric tensions in these molecules, coupled with
the fact that the two boron trigonal planes approach
orthogonality, have some implications for how the boron
centers might (or might not) act in concert as Lewis
acids. Although the cis boron center might be expected
to be more Lewis acidic since it is not in conjugation
with CdC, both faces of the trigonal plane are sterically
blocked, one by the cis R substituent and the other by
a C6F5 group on the other B(C6F5)2 moiety. Indeed, the
region of space between the boron centers in these
molecules is quite crowded and, at least by appearances,
inaccessible to Lewis bases. Consequently, 1a does not
appear to coordinate THF even in solution and the
binding of simple ketones such as acetone is reversible.19
The open face of the conjugated Btrans(C6F5)2 group thus
appears, a priori, to be the most likely site of reactivity
for these molecules. This prediction is born out in the
reactions of 1a with dimethylzirconocene, which are
described in detail below.
This ability is not, however, due to chelation of the
abstracted methide. Proton spectroscopic data associ-
ated with the counteranion suggest that the diborane
utilizes only one of the boron centers to abstract the
1
alkyl group. In the H NMR spectrum at -60 °C, the
Rea ction s of 1a w ith Cp 2Zr Me2. Since the 1,1-
alkenyl boranes serve as effective activators in olefin
polymerization experiments (vide infra), the reactions
of 1a with dimethylzirconocene were examined in
detail.20 When reacted in a 1:1 ratio, the product
mixture included one-half of the starting borane, sug-
gesting that a 2:1 ratio of [Zr]:1a would be more
appropriate. This was born out as depicted in Scheme
2; when the reagents are mixed in this ratio at -78 °C
resonance for the BCH3 group appears at -0.39 ppm
as a broad signal; NOE experiments in which this signal
is irradiated show strong enhancement in the resonance
due to the vinylic proton at 6.31 ppm, while a negligible
effect on the resonance for the tert-butyl protons was
observed. A reciprocal enhancement in BCH3 was found
when the irradiation was directed to the 6.31 ppm
resonance. Together, these results imply that the
borate methyl group is cis to the olefinic proton; a
(19) Ko¨hler, K.; Piers, W. E. Can. J . Chem., submitted for publica-
tion.
(20) Reactions of Cp2ZrMe2 and 1b had essentially the same outcome
but were not studied in as much detail.
(21) (a) Buchwald, S. L.; Watson, B. T.; Huffman, J . C. J . Am. Chem.
Soc. 1987, 109, 2544. (b) Neilsen, R. B.; Buchwald, S. L. Chem. Rev.
1988, 88, 1047.
(22) Chen, Y.-X.; Marks, T. J . Organometallics 1997, 16, 3649.

Bifunctional Lewis Acids
Organometallics, Vol. 17, No. 16, 1998 3561
Ta ble 1. Selected Metr ica l P a r a m eter s for Dibor a n es 1a -c
1a
1b
1c
Bond Distances (Å)
C(1)-C(2)
C(1)-C(2)
C(2)-C(3)
B(1)-C(1)
B(2)-C(1)
B(1)-C(16)
B(1)-C(26)
B(2)-C(36)
B(2)-C(46)
1.357(7)
1.504(8)
1.544(8)
1.564(8)
1.601(7)
1.613(7)
1.622(7)
1.618(6)
1.365(2)
1.454(2)
1.540(2)
1.564(2)
1.591(2)
1.578(2)
1.585(2)
1.577(2)
1.357(6)
1.456(6)
1.555(7)
1.561(6)
1.580(7)
1.566(7)
1.569(6)
1.586(6)
C(2)-C(3)
B(2)-C(1)
B(1)-C(1)
B(2)-C(27)
B(2)-C(21)
B(1)-C(9)
B(1)-C(15)
Bond Angles (deg)
C(2)-C(1)-B(2)
C(2)-C(1)-B(1)
B(1)-C(1)-B(2)
C(1)-C(2)-C(3)
C(1)-B(2)-C(27)
C(1)-B(2)-C(21)
C(21)-B(2)-C(27)
C(1)-B(1)-C(15)
C(1)-B(1)-C(9)
C(15)-B(1)-C(9)
C(2)-C(1)-B(1)
C(2)-C(1)-B(2)
B(1)-C(1)-B(2)
C(1)-C(2)-C(3)
C(1)-B(1)-C(16)
C(1)-B(1)-C(26)
C(16)-B(1)-C(26)
C(1)-B(2)-C(36)
C(1)-B(2)-C(46)
C(36)-B(2)-C(46)
115.4(5)
125.8(5)
118.7(4)
131.4(5)
124.2(5)
122.8(4)
112.9(4)
119.9(4)
118.6(4)
121.0(4)
117.48(13)
119.24(13)
123.22(13)
124.04(13)
120.55(13)
122.04(13)
117.40(13)
122.53(12)
119.36(12)
117.94(12)
115.7(4)
122.5(4)
121.7(4)
129.9(4)
120.0(4)
121.9(4)
118.2(4)
118.4(4)
121.7(4)
119.9(4)
Torsion Angles (deg)
C(27)-B(2)-C(1)-C(2)
C(21)-B(2)-C(1)-C(2)
C(15)-B(1)-C(1)-C(2)
C(9)-B(1)-C(1)-C(2)
B(2)-C(1)-C(2)-C(3)
B(1)-C(1)-C(2)-C(3)
C(1)-C(2)-C(3)-C(4)
C(16)-B(1)-C(1)-C(2)
C(26)-B(1)-C(1)-C(2)
C(36)-B(2)-C(1)-C(2)
C(46)-B(2)-C(1)-C(2)
B(1)-C(1)-C(2)-C(3)
B(2)-C(1)-C(2)-C(3)
12.3(8)
4.5(2)
-17.8(6)
161.4(4)
143.5(4)
-36.6(6)
169.3(4)
-10.6(7)
-40.2(7)
-165.8(5)
-72.1(7)
115.7(6)
164.5(6)
-10.3(9)
-176.1(2)
131.6(2)
-53.1(2)
169.7(2)
-12.9(2)
-34.4(2)
Sch em e 2
Sch em e 3
chelated methide anion would not be expected to exhibit
NOE enhancements to either of the olefinic groups. In
the 19F NMR spectrum at -60 °C, 20 different reso-
nances are observed, indicating that all four C6F5 groups
are chemically distinct and that rotation of each of the
aryl rings is restricted. Complete assignment of this
spectrum and conclusions as to the coordination envi-
ronments about the boron centers based on the relative
chemical shifts23 of the meta and para fluorine reso-
nances were not possible.
Upon warming to -40 °C, 5a undergoes decomposi-
tion to various products, a process which begins with
the loss of CH3B(C6F5)2 from the anion. The resulting
vinyl anion, which is likely stabilized to some extent by
the strongly electron-withdrawing -B(C6F5)2 group,24
reacts with the zirconocene cation to form the boravinyl
methyl zirconocene precursor to 4a . Again, this species
is not observed, not surprising given its propensity to
lose methane. The other products of the decomposition
of ion pair 5a , i.e., Cp2ZrMe2 and CH3B(C6F5)2, are also
not spectroscopically observed in this process; rather,
they react together rapidly under these conditions to
give a different ion pair [Cp2ZrMe]+[Me2B(C6F5)2]-, 8,
as shown in Scheme 4. This can be shown to occur in a
separate experiment using Cp2ZrMe2 and CH3B(C6F5)2.
This latter experiment also reveals the tendency of ion
pair 8 to decompose at higher temperatures to the
neutral products Cp2ZrMeC6F5 and Me2BC6F5, which
(24) Krishnamurthy, S.; Brown, H. C. J . Org. Chem. 1980, 45, 849.
(b) Yoshida, T.; Negishi, E. J . Chem. Soc., Chem. Commun. 1974, 763.
(c) Pelter, A.; Smith, K.; Brown, H. C. Borane Reagents; Academic
Press: New York, 1988; p 22-23.
(23) Horton, A. D.; de With, J . Organometallics 1997, 16, 5424.

3562 Organometallics, Vol. 17, No. 16, 1998
Ko¨hler et al.
Sch em e 4
Sch em e 5
Ta ble 2. P olym er iza tion of Eth ylen e Usin g
Cp 2Zr Me2 (3) a n d Bor on Coca ta lystsa
occurs with a half-life of about 20 min at room temper-
ature. This accounts for the presence of these neutral
compounds in the product mixture of the reaction
between 1a and dimethylzirconocene.
[3]
(µM)
A (103 kg of PE/
entry
cocatalystb
T (°C)
(mol of Zr × h))c
1
2
3
4
5
6
7
8
9
70.0
65.0
65.0
70.0
70.0
70.0
70.0
65.0
65.0
B(C6F5)3
B(C6F5)3
B(C6F5)3
1a
1a
1a
1b
1b
1b
50
50
30
70
50
50
70
50
50
3.9
5.2
9.5
3.0
4.9
5.5
3.4
2.0
6.0
A final observation concerning this system involves
the course of the reaction when dimethyl zirconocene
and 1a are mixed in a 1:1 ratio at -60 °C (Scheme 5).
This reaction shows that dinuclear cation formation is
not favored over further methyl abstraction, as was the
case when the bulkier tris(2,2′,2′′-perfluorobiphenyl)-
borane was employed.2c,22 The ion pair 6a was charac-
1
terized by H NMR spectroscopy and underwent con-
a
Polymerization conditions: toluene 500 mL, 75 psig of ethyl-
version to dinuclear 5a and 1a at temperatures where
decomposition of 5a ensued, precluding quantitative
measurements of the equilibrium between these two
species. The final product mixture of this reaction was
the same as that in the 2:1 reaction with the addition
of 0.5 equiv of diborane 1a .
ene, 1000 rpm with 3 and cocatalyst premixed in toluene solution
b
at room temperature. For each cocatalyst, 1.2 equiv with respect
to 3 was employed. c Activity in 106 g PE/(mol of Zr × h), based
on an impurity level of 50 µM.
employed, and these were found to have a considerable
influence on polymerization activity and, therefore, will
be discussed in sequence.
The first procedure involved pre-contacting complex
3 with the appropriate cocatalyst (1.2 equiv) in a small
volume of toluene at room temperature for a period of
10-15 min,25 followed by introduction into the reactor,
presaturated with monomer at the desired pressure and
Eth ylen e P olym er iza tion s w ith Bor on Coca ta -
lysts. Ethylene polymerizations using 1a , 1b, and, for
comparison purposes, B(C6F5)3 as cocatalysts with Cp2-
ZrMe2 (3) were conducted in a toluene slurry under a
variety of conditions, and the results are summarized
in Tables 2 and 3. Two different procedures were

Bifunctional Lewis Acids
Organometallics, Vol. 17, No. 16, 1998 3563
Ta ble 3. Eth ylen e P olym er iza tion Usin g Cp 2Zr Me2
(3) a n d Bor on Coca ta lystsa
Table 3. Instead of premixing the catalyst and cocata-
lyst, the requisite cocatalyst was introduced first, fol-
lowed by metallocene 3. Although this sometimes led
to measurable induction periods prior to monomer
uptake, this procedure led to much more reproducible
results (i.e., temperature control to better than (2 °C
with stable flow profiles) than the protocol involving
premixing of the two catalyst components. The second
procedure also involved the use of TMA as a scrubbing
agent, and the optimal level for this additive was found
to be 30 µM (i.e., 90 µM based on Al-Me bonds) using
the catalyst combination 3/B(C6F5)3 (each at ca. 15 µM);
lower TMA loadings resulted in low/no ethylene con-
sumption, while higher loadings also suppressed activity
(to a lesser extent), possibly via formation of [Cp2Zr(µ-
Me)2AlMe2][X] complexes in situ.26
From the activity data summarized in Table 3, it is
clear that the most active cocatalyst is B(C6F5)3. With
this cocatalyst, the mode of catalyst formation, i.e.,
premixed vs in situ generation, has no effect on polym-
erization activity (entries 1 vs 2). In entry 3, no TMA
was used as a scrubbing agent and 60 µM 3 was initially
present but only 15 µM B(C6F5)3 was introduced; the
fact that the productivity was comparable to the poly-
merizations conducted in the presence of TMA indicates
that, at the levels used, this additive has little effect on
either polymerization activity or polymer properties.
Cocatalysts 1a and 1b were closely equivalent and
gave rise to catalysts which were about 50-60% less
active than 3/B(C6F5)3 under equivalent conditions
(Table 3, entries 4 and 6 vs 1). The MW of the PE
produced was somewhat higher in the case of 1a
compared to 1b or B(C6F5)3, although one should be
cautious in concluding that the MWs are significantly
different given the uncertainties in both MW determi-
nation and polymerization conditions. Interestingly, the
MWD of the polymers produced using either 1a or 1b
were significantly narrower than those using B(C6F5)3.
This is quite surprising given the thermal instability
of the ion pairs derived from 3 or 1a (vide supra).
Given the uncertainty as to which species was pro-
ducing polymer in polymerizations involving 1a or 1b,
ethylene polymerizations were also conducted with the
initially formed decomposition products from 3 and 1a
(i.e., 4a and MeB(C6F5)2, 7). The polymerization of
ethylene was first examined with zirconacyclopropene
4a or with a mixture of 1a and 4a (1:1 mol ratio). The
lack of measurable ethylene uptake during these ex-
periments indicates that complex 4a is inactive in
ethylene polymerization, despite the fact that this
complex does react with ethylene at room temperature.27
entry cocatalystb T (°C)
Ac
Mn (K)
Mw/Mn
1
B(C6F5)3
B(C6F5)3
B(C6F5)3
1a
1a
1b
1b
1b
7
30
30
30
30
65
30
65
65
30
5.2 (0.5)d 195 (5)d
3.9 (0.4)d
4.3 (0.5)f
3.5 (0.2)f
2e
3g
4
5.0 (0.5)f
5.9 (0.4)f
170 (3)f
155 (15)f
3.3 (0.3)d 270 (25)d 1.9 (0.1)d
5
6
3.0 (0.3)f
3.2 (0.2)f
4.6 (0.3)f
1.4
132 (14)f
204 (10)f
100 (3)f
108
2.6 (0.2)f
1.9 (0.1)f
3.1 (0.1)f
3.43
7
8e
9
2.4 (0.2)d 245 (15)d 2.1 (0.1)d
a
Polymerization conditions: toluene 500 mL; 75 psig of C2, 1000
rpm, 20-30 min polymerization time. [TMA] ) 30 µM; Cp2ZrMe2
(15 µM) was injected into a saturated solution of the cocatalyst
b
and TMA, except where noted. The amount of cocatalyst present
was 1.2 equiv with respect to 3. c Activity in 106 g of PE/(mol of
d
Zr × h). Average value with estimated standard deviation in
parentheses (4-6 trials). e Catalyst and cocatalyst were premixed
prior to introduction into the reactor containing TMA. f Average
of two trials with range in parentheses. g Cocatalyst (15 µM) added
to a solution of Cp2ZrMe2 (60 µM) and monomer without TMA
present.
temperature (Table 2). No scrubbing agent was em-
ployed, and as a result, quite high catalyst loadings (65-
70 µM) were required to observe measurable consump-
tion of monomer. Control experiments (at lower catalyst
loadings) revealed that the level of protic impurities in
the solvent/monomer system under these conditions was
between 40 and 60 µM, based on [Cp2ZrMe(µ-Me)B-
(C6F5)3] (i.e., ca. 100-120 µM based on potentially
reactive M-Me bonds). The activity data in Table 2 are
quoted on the basis of an impurity level of 50 µM, based
on similar work conducted in the presence of TMA (vide
infra).
In practice, these values are not very meaningful; in
most cases, rapid and exothermic polymerization was
observed under these conditions followed by an (equally)
rapid decline in the rate of monomer consumption to a
much lower steady-state value. Presumably, the latter
value is indicative of the actual amount of active catalyst
present; the initial phase of these polymerizations can
be viewed as a competition reaction for [Zr*] between
monomer (which leads to polymerization) and adventi-
tious quenching by, e.g., H2O.
The interpretation of these results are complicated
by the instability of the ion pairs derived from 3 and
1a (and presumably 1b) at room temperature in toluene
solution (vide supra). Under the conditions studied (pre-
contacting in toluene), one would have expected maxi-
mal decomposition of the ion pairs into the final
products, as outlined previously, and thus the observed
catalytic activity and polymer properties could have
resulted from any of the species formed. As unimodal
molecular weight distributions (MWDs) with Mw/Mn )
2-4 were observed in all cases, one can at least conclude
that there is either a single active species present and/
or (less likely) the ratio of chain propagation to chain
transfer is similar for the different species present.
With a view to overcoming the thermal instability
problem, a second procedure was developed for screen-
ing these materials and the results are summarized in
In contrast, polymerization studies involving the use
of MeB(C6F5)2 as a cocatalyst with 3 produced high
molecular weight polyethylene with a narrow MWD
(Table 3, entry 9). The catalytic activity was somewhat
lower than that observed for 1a or 1b, but the MW and
MWD were comparable to those observed when using
either cocatalyst 1a or 1b. These results suggest that
the polymerization observed in the presence of diboranes
1 can be interpreted as arising from MeB(C6F5)2 and 3.
(25) Catalysts were introduced into the reactor by using a sample
bomb and blowing the contents of this into the reactor using N2. The
precontacting time reflects the amount of time needed to fill the bomb
in the glovebox followed by removal and attachment to the reactor
system.
(26) Bochmann, M.; Lancaster, S. Angew. Chem., Int. Ed. Engl.
1994, 33, 1634.
(27) Complex 4a reacts with 1 equiv of ethylene to form a compound
of unknown structure: Ko¨hler, K.; Piers, W. E. Unpublished results.

3564 Organometallics, Vol. 17, No. 16, 1998
Ko¨hler et al.
Ta ble 4. Kin etics of Eth ylen e P olym er iza tion
Usin g Cp 2Zr Me2 (3) a n d Bor on Coca ta lysts
fundamentally alter the intrinsic reactivity of the cat-
ionic alkyl. It might be instructive to determine active
site concentrations with simple systems such as these.29
The only unexplained phenomenon is the narrower
MWD observed in the case of cocatalysts 1a , 1b, and
MeB(C6F5)2 compared to 3/B(C6F5)3. We cannot account
for this difference on the basis of the experiments
reported here, but it may reflect stabilization of the
metal center, by strong counterion association, toward
(irreversible) catalyst deactivation in the presence of
monomer.
Con clu sion s. A series of 1,1-di[bis(perfluorophenyl)-
boryl]alkenes were prepared and shown to be less prone
to loss of HB(C6F5)2 than their fully saturated counter-
parts. Although Lewis acidic enough to activate the
simple metallocene 3, the borate anion formed was
unstable at ambient temperatures and decomposed to
form a well-defined product mixture. The spectroscopic
data obtained on this process suggested that only one
of the borane centers is involved in this chemistry, and
the structural data on the diboranes themselves pro-
vides a plausible explanation as to why this is the case.
These observations imply that a wider bite angle (or a
> 1 atom linker) is desirable for multidentate Lewis-
acid activators. Nonetheless, the diboranes 1a and 1b
do serve as cocatalysts for the polymerization of ethylene
using Cp2ZrMe2. On the basis of these data, as well as
that involving reactions of 1a and 1b with 3, it seems
probable that the active species in both cases is
[Cp2ZrMe]+[Me2B(C6F5)2]-, generated in situ from the
reaction of MeB(C6F5)2 with 3.
entry conditionsa cocatalyst
Rp
kp[Zr*]c Xn (K)
Rtr
b
d
1
2
3
4
1
4
6
9
B(C6F5)3 5.9(0.6) 8.2(0.8) 7.0(0.2) 8.4(1.1)
1a
1b
7
2.0(0.2) 2.8(0.3) 9.6(0.9) 2.1(0.4)
2.4(0.2) 3.3(0.3) 7.3(0.4) 3.3(0.4)
2.4(0.2) 3.3(0.3) 8.7(0.5) 2.7(0.4)
a
For polymerization conditions, see the appropriate entries in
b
Table 2. Rate of polymerization in 10-4 M s-1 as measured by
calibrated mass flow meters at steady state. c Value (10-4 s-1
)
obtained by dividing Rp by [C2H4] at the temperature indicated.
Ethylene solubility in toluene at various temperatures and 75 psi
was estimated using methods described in Wilhelm, E.; Battino,
d
R. Chem. Rev. 1973, 73, 1. Value calculated from Rp and Xn in
10-8 M s-1
.
With this in mind, it is instructive to compare the
data in Table 3 on a different basis, namely, the steady-
state rate of polymerization, coupled with differences
in MW. Activity values represent the area under the
curve of monomer consumption vs time and thus are
not really indicative of instrinsic differences in the rate
of polymerization and/or catalyst stability. Qualita-
tively, catalysts derived from either 1a , 1b, or MeB-
(C6F5)2 and 3 were less stable under the polymerization
conditions, as revealed by a decline in polymerization
rates from peak values, whereas the combination
3/B(C6F5)3 appeared to be more stable (little or no
decline from peak values).
Summarized in Table 4 are the steady-state rates of
polymerization for the different cocatalysts, as well as
an estimate for kp[Zr*] at steady state (assuming first-
order kinetics in monomer concentration) at 30 °C. At
this temperature, catalyst decay was minimal using the
different cocatalysts and so the results can be meaning-
fully compared. In addition, we have included estimates
for Rtr, the rate of chain transfer, based on the number
average degree of polymerization observed under the
Exp er im en ta l Section
Gen er a l. All manipulations of air- and moisture-sensitive
materials were undertaken using standard vacuum and
Schlenk techniques30 or in a glovebox under an atmosphere
of nitrogen. All solvents were dried and purified by passing
through suitable drying agents (alumina and Q5).31 NMR
spectra were recorded in C6D6 at room temperature or in C7D8
for low-temperature experiments. Data are given in ppm
28
different conditions.
As can be seen from the results summarized in Table
4, within experimental error, the Rp and Rtr for 1a , 1b,
and MeB(C6F5)2 are closely comparable while those for
3/B(C6F5)3 are significantly different. Also, the magni-
tude of Rtr is significantly higher (ca. 3 times) for
3/B(C6F5)3 compared to the other catalysts, but the
differences in Rp are also of a similar magnitude (ca.
2.5 times). This accounts for the fortuitous formation
of PE with similar MW. It is possible that weaker
counterion interactions in the former case lead to more
facile chain transfer reactions.2
The derived values for kp[Zr*] in Table 4 obviously
correlate strongly with Rp. If kp is characteristic of only
the cationic alkyl (i.e., [Cp2ZrMe]), one interpretation
of counterion effects is that more strongly coordinating
counterions serve to reduce the steady-state concentra-
tion of actively growing chains but otherwise do not
1
relative to solvent signals for H and 13C spectra or relative to
external standards for 11B (BF3‚OEt2, 0.0 ppm) and 19F (CFCl3
at 0.0 ppm) experiments. IR data is reported in wavenumbers
(cm-1). Elemental analyses were performed by Mrs. Dorothy
Fox in the microanalytical laboratory of the Department of
Chemistry at the University of Calgary. The complexes Me2-
Sn(CtCR)2 (R ) But, Ph),32 Cp2ZrMe2,33 [Cp2Zr(H)CH3],34 ClB-
(C6F5)2,35 HB(C6F5)2,12 and MeB(C6F5)2 were prepared using
36
literature methods. B(C6F5)3 was purchased from Boulder
Scientific, dried by treatment with Me3SiCl, and sublimed prior
to use.
P r ep a r a tion of Bu tCtCB(C6F 5)2, 2a . Hexane (20 mL)
was condensed into an evacuated flask containing ClB(C6F5)2
(29) (a) Marques, M. M.; Tait, P. J . T.; Mejzl´ık, J .; Dias, A. R. J .
Polym. Sci., Part A. 1998, 36, 573. (b) Tait, P. J . T. In Transition Metal
Catalyzed Polymerization: Alkenes and Dienes; Quirk, R. P., Ed.;
Hartwood Acadademic Publishers: New York, 1983; p 115.
(30) Shriver, D. F.; Drezdzon, M. A. The Manipulation of Air-
Sensitive Compounds, 2nd ed.; Wiley: New York, 1986.
(31) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J . Organometallics 1996, 15, 1518.
(28) All of these polymerizations were conducted for time periods
between 20 and 30 min following establishment of steady-state
conditions (i.e., constant mass flow vs time). On the basis of the turn-
over numbers calculated from Rp in Table 4 (assuming all of the added
Cp2ZrMe2 is active) and the degree of polymerization, 2-3 polymer
chains were produced per Zr atom over this time period using
cocatalysts 1a , 1b, or MeB(C6F5)2. Thus, the calculated values of Rtr
represent true chain-transfer rates as opposed to rates of termination;
the lack of significant decline in Rp over this time period at 30 °C also
indicates catalyst stability under these conditions. Evidently, the
presence of monomer (or a growing polymer chain) hinders the aryl
transfer reaction observed when Cp2ZrMe2 and MeB(C6F5)2 are mixed
at room temperature.
(32) Wrackmeyer, B.; Kundler, S.; Boese, R. Chem. Ber. 1993, 126,
1361.
(33) Samuel, E.; Rausch, M. D. J . Am. Chem. Soc. 1973, 95, 6263.
(34) (a) J ordan, R. F.; Bajgur, C. S.; Dasher, W. E.; Rheingold, A. L.
Organometallics 1987, 6, 1041. (b) Gell, K. I.; Posin, B.; Schwartz, J .;
Williams, G. M. J . Am. Chem. Soc. 1982, 104, 1846.
(35) Chambers, R. D.; Chivers, T. J . Chem. Soc. 1965, 3933.
(36) Biagini, P.; Lugli, G.; Garbassi, F.; Andreussi, P. Eur. Pat. Appl.
EP 667,357, 1995.

Bifunctional Lewis Acids
Organometallics, Vol. 17, No. 16, 1998 3565
(245 mg, 0.643 mmol) at -78 °C, and Me2Sn(CtCBut)2 (100
mg, 0.322 mmol) dissolved in hexane (10 mL) was added
dropwise. After the reaction solution was stirred for 1 h at
-78 °C, it was warmed to 25 °C and stirred for a further hour.
The hexane was removed, and the Me2SnCl2 was separated
by sublimation under full vacuum at 30 °C. The residue was
recrystallized at -30 °C in benzene/hexane (1:10) and yielded
ButCtCB(C6F5)2 (230 mg, 85%). 1H NMR: δ 1.19 (s, 9H,
CCH3). 13C{1H} NMR: δ 30.4 (CCH3), 30.2 (CCH3). 11B
NMR: δ 48.5. 19F NMR: δ -129.1 (m, 4F, o-F), -146.5 (m, 2
F, p-F), -161.8 (m, 4F, m-F). MS: 426 (25) [M]+; 411(70) [M
P r ep a r a tion of Me2Sn (CtCC6F 5)2. Ether (20 mL) was
condensed into an evacuated flask, and HCtCC6F5 (200 mg,
1.04 mmol) was added at -78 °C. After addition of BuLi/
hexane (0.65 mL, 1.04 mmol, 1.6M) at -78 °C, it was warmed
to 25 °C and stirred for 30 min. Me2SnCl2 (114 mg, 0.52 mmol)
was added in one portion, and the mixture was stirred
overnight. A white precipitate was formed. The ether was
removed in vacuo, and the residue was dissolved in hexane
(20 mL). After filtration, the solvent was removed in vacuo,
yielding a colorless powder. Crystallization at -30 °C in
hexane yielded Me2Sn(CtCC6F5)2 (166 mg, 60%). 1H NMR:
δ 0.20 (s, 6H, SnCH3). 13C{1H} NMR: δ 148.3 (m, C6F5), 142.1
(m, C6F5), 138.0 (m, C6F5), 106.0 (m, SnCtC), 92.7 (m, SnCt
C), -6.2 (s, SnCH3). 19F NMR: δ -137.0 (m, 4F, o-F), -152.4
(m, 2F, p-F), -162.0 (m, 4F, m-F). IR (KBr pellet): 2155 (w),
1500 (s), 996 (s), 967 (s), 789 (m). EI-MS: M+ - Me - 3F
- Me]+. Anal. Calcd for C18H9BF10
: C, 50.74; H, 2.13.
Found: C, 50.29; H, 2.09.
P r ep a r a tion of Bu t(H)CdC[B(C6F 5)2]2, 1a . Hexane (20
mL) was condensed into an evacuated flask containing ClB-
(C6F5)2 (122 mg, 0.322 mmol) at -78 °C, and Me2Sn(CtCBut)2
(50 mg, 0.161 mmol) dissolved in hexane (10 mL) was added.
After the reaction solution was stirred for 1 h at -78 °C it
was warmed to 25 °C and stirred for a further hour. A clear,
colorless solution was formed. HB(C6F5)2 (111 mg, 0.322
mmol) was added, which was completely dissolved after 5 min.
The hexane was removed, and the Me2SnCl2 was separated
by sublimation under full vacuum at 25 °C. The residue was
recrystallized at -30 °C in benzene/hexane (1:10) and yielded
But(H)CdC[B(C6F5)2]2 (220 mg, 90%). 1H NMR: δ 7.16 (s, 1H,
HCd), 0.87 (s, 9H CCH3). 13C{1H} NMR: δ 183.6 (HCd), 39.0
(CCH3), 29.2 (CCH3). 11B NMR: δ 39.9. 19F NMR: δ -126.5
(m, 4F, o-F), -130.1 (m, 4F, o-F), -143.3 (m, 2 F, p-F), -148.7
(t, 2 F, p-F), -160.5 (m, 4F, m-F), -160.6 (m, 4F, m-F). IR
(KBr pellet): 2972 (w), 1649 (m), 1644 (m), 1550 (m), 1524
(m), 1478 (s), 1391 (m), 1312 (m), 1210 (m), 1178 (m), 975 (s).
EI-MS: 772 (40) [M+], 604 (80) [M - C6F5]+, 346 (30) [HB-
(C6F5)2]+, 258 (50) [M - HB(C6F5)2 - C6F5]+, 57 (80) [But]+.
Anal. Calcd for C30H10B2F20: C, 46.67; H, 1.31. Found: C,
46.47; H, 0.81.
[460 (20)], M+ - Me - C6F5 [348 (10)], C2C6F5 [192 (80)].
+
P r ep a r a tion of C6F 5(H)CdC[B(C6F 5)2]2, 1c. Hexane (20
mL) was condensed into an evacuated flask containing ClB-
(C6F5)2 (103 mg, 0.271 mmol) at -78 °C, and Me2Sn(CtCC6F5)2
(72 mg, 0.136 mmol) dissolved in hexane (10 mL) was added.
After the reaction solution was stirred for 1 h at -78 °C, it
was warmed to 25 °C and stirred for a further hour. A yellow,
clear solution was formed. [HB(C6F5)2]2 (94 mg, 0.271 mmol)
was added, which was completely dissolved after 5 min. The
hexane was removed, and the Me2SnCl2 was separated by
sublimation under full vacuum at 25 °C. The residue was
recrystallized at -30 °C in benzene/hexane (1:10) and yielded
C6F5(H)CdC[B(C6F5)2]2 (155 mg, 65%). 1H NMR: δ 7.87 (s,
1H, HCd). 13C{1H} NMR: δ 148.8 (HCd). 11B NMR: δ 61.8.
19F NMR: δ -126.3 (d, 4F, o-F), -128.9 (d, 4F, o-F), -141.5
(d, 2F, o-F), -142.5 (m, 2F, p-F), -145.3 (m, 2F, p-F), -146.1
(t, 1F, p-F), -158.8 (m, 4F, m-F), -159.9 (m, 2F, m-F), -160.2
(m, 4F, m-F). IR (KBr pellet): 1650 (m), 1644 (m), 1524 (s),
1477 (vs), 1388 (m), 1315 (m), 1165 (m), 975 (s). EI-MS: 882
[M]+. Anal. Calcd for C32HB2F25: C, 43.58; H, 0.11. Found:
C, 43.06; H, 0.00.
Rea ction of Cp 2Zr Me2 w ith 1a (2:1). A sealable 5 mm
NMR tube was loaded with Cp2ZrMe2 (22 mg, 0.087 mmol)
and dissolved in toluene-d8 (ca. 0.3 mL). The tube was
attached to a vacuum line and cooled to -78 °C. A solution of
1a (33 mg, 0.043 mmol) in d8-toluene (ca. 0.4 mL) was added
by syringe and the tube was flame sealed. The tube was
briefly shaken and loaded into a precooled NMR probe (-80
°C) tuned to either 1H or 19F. The ensuing reaction was
monitored spectroscopically as a function of temperature.
Data for 5a (-60 °C). 1H NMR: δ 5.62 (s, 20H, C5H5), 6.31 (s,
1H, HCd), 1.12 (s, 9H, CCH3), -0.10 (s, 6H, ZrCH3), -0.39 (s,
br, 3H, BCH3), -1.23 (s, 3 H, ZrCH3Zr). 19F NMR: δ -123.0
(ABq, br, 1F, J ) 113 Hz), -123.2 (ABq, br, 1F, J ) 113 Hz),
-126.0 (s, br, 1F), -126.3 (s, br, 1F), -126.9 (s, br, 1F), -130.9
(s, br, 1F), -132.7 (s, br, 1F), -138.7 (s, br, 1F), -145.1 (s, br,
1F), -149.9 (s, br, 1F), -157.3 (s, br, 1F), -159.9 (s, br, 1F),
-160.4 (s, br, 1F), -161.2 (s, br, 1F), -161.5 (s, br, 1F), -162.2
(s, br, 1F), -163.0 (s, br, 1F), -163.8 (s, br, 2F), -167.2 (s, br,
1F). Data for Cp2ZrMeC6F5. 1H NMR: δ 5.63 (s, 10H, C5H5),
0.25 (t, 3H, 5J HF ) 3.9 Hz, ZrCH3). 13C NMR: δ 111.4 (C5H5),
45.7 (ZrCH3). 19F NMR: δ -115.2 (m, 2 F, o-F), -156.6 (t, 1
F, p-F), -162.2 (m, 2 F, m-F). Data for Me2BC6F5. 1H NMR:
δ 1.0 (s, br). 19F NMR: δ -130.9 (m, 2 F, o-F), -151.4 (t, 1 F,
p-F), -162.6 (m, 2 F, m-F). 11B NMR: δ 80.2.
P r ep a r a tion of P h CtCB(C6F 5)2, 2b. Hexane (20 mL)
was condensed into an evacuated flask containing ClB(C6F5)2
(217 mg, 0.570 mmol) at -78 °C, and Me2Sn(CtCPh)2 (100
mg, 0.285 mmol) dissolved in hexane (10 mL) was added. After
the reaction solution was stirred for 1 h at -78 °C, it was
warmed to 25 °C and stirred for a further hour. A clear, light
yellow solution was observed. The hexane was removed, and
the Me2SnCl2 was separated by sublimation under full vacuum
at 25 °C. The residue was recrystallized at -30 °C in benzene/
1
hexane (1:10) and yielded PhCtCB(C6F5)2 (190 mg, 75%). H
NMR: δ 7.55 (m, 2H, C6H5), 6.95 (m, 3H, C6H5). 13C{1H}
NMR: δ 134.3, 132.3, 129.4, 122.2 (C6H5). 11B NMR: δ 47.4.
19F NMR: δ -128.5 (m, 4F, o-F), -145.9 (m, 2F, p-F), -161.5
(m, 4F, m-F). MS: 446 (90) [M]+. Anal. Calcd for C20H5-
BF10: C, 53.85; H, 1.13. Found: C, 53.63; H, 0.76.
P r ep a r a tion of P h (H)CdC[B(C6F 5)2]2, 1b. Hexane (20
mL) was condensed into an evacuated flask containing ClB-
(C6F5)2 (108 mg, 0.285 mmol) at -78 °C, and Me2Sn(CtCPh)2
(50 mg, 0.142 mmol) dissolved in hexane (10 mL) was added.
After the reaction solution was stirred for 1 h at -78 °C, it
was warmed to 25 °C and stirred for a further hour. A clear,
light yellow solution was formed. [HB(C6F5)2]2 (99 mg, 0.285
mmol) was added, which was completely dissolved after 5 min.
The hexane was removed, and the Me2SnCl2 was separated
by sublimation under full vacuum at 25 °C. The residue was
recrystallized at -30 °C in benzene/hexane (1:10) and yielded
Ph(H)CdC[B(C6F5)2]2 (180 mg, 80%). 1H NMR: δ 8.27 (s, 1H,
HCd), 6.95 (m, 2H, C6H5), 6.81 (m, 3H, C6H5). 13C{1H} NMR:
δ 173.1 (HCd), 138.6 (Cipso), 133.1, 131.3, 129.2 (C6H5). 11B
NMR: δ 62.7. 19F NMR: δ -127.2 (d, 4F, o-F), -129.8 (m,
4F, o-F), -144.4 (m, 2F, p-F), -147.9 (t, 2F, p-F), -160.3 (m,
4F, m-F), -161.1 (m, 4F, m-F). IR (KBr pellet): 1649 (m), 1644
(m), 1519 (s), 1481 (vs), 1389 (m), 1310 (m), 1202 (m), 1150
(m), 975 (s). EI-MS: 791 (50) [M]+, 77 (100) [C6H5]+. Anal.
Calcd for C32H6B2F20: C, 48.53; H, 0.53. Found: C, 48.07; H,
0.26.
Rea ction of Cp 2Zr Me2 w ith 1a (1:1). The same procedure
as that described above was employed, using 11 mg (0.43
mmol) of Cp2ZrMe2 and 33 mg (0.043 mmol) of 1a . 1H NMR
data for 6a : 5.30 (s, 10H, C5H5), 6.11 (s, 1H, HCd), 1.03 (s,
9H, CCH3), 0.26 (s, 3H, ZrCH3), -0.39 (s, br, 3H, BCH3).
P r ep a r a tion of Cp 2Zr [Bu tCtCB(C6F 5)2], 4a . [Cp2Zr(H)-
CH3]n (92 mg, 0.194 mmol) and ButCtCB(C6F5)2 (165 mg,
0.387 mmol) were dissolved in toluene (20 mL) at -78 °C. After
the reaction solution was stirred for 1 h at -78 °C, it was
warmed to 25 °C. A deep red-brown solution was obtained.

3566 Organometallics, Vol. 17, No. 16, 1998
Ta ble 5. Su m m a r y of Da ta Collection a n d Str u ctu r e Refin em en t Deta ils for 1a -c
Ko¨hler et al.
1a
1b
1c
formula
fw
C
30H10B2F20
C32H6B2F20
791.99
C32HB2F25
881.95
772.00
temp, K
cryst syst
a, Å
b, Å
c, Å
296(2)
160(2)
160(2)
monoclinic
17.8248(5)
9.0087(5)
19.338(1)
99.238(2)
3065.0(2)
P21/c
monoclinic
13.5310(6)
14.0523(6)
16.1446(7)
102.835(2)
2993.1(2)
P21/n
monoclinic
17.318(3)
8.7958(15)
21.475(3)
109.633(4)
3081.0(9)
P21/c
â, °
V, Å3
space group
Z
4
4
4
d
calc, mg m-3
1.673
0.182
1.758
0.189
1.901
0.217
µ, mm-1
cryst size, mm
no. of rflns measd
no. of unique rflns
Rint
0.40 × 0.30 × 0.30
0.50 × 0.40 × 0.35
0.18 × 0.16 × 0.12
10 717
23 300
14 929
3907
0.0482
7135
0.0196
5410
0.0785
no. of variables
max, min e density Å3
R1a
422
0.385/-0.238
0.0746
512
0.317/-0.290
0.0377
537
0.294/-0.298
0.0575
wR2a
0.1763
0.0978
0.1356
gof
1.089
1.031
0.988
a
R1 is a conventional R factor based on F values of reflections having F2 > 2σF2; wR2 is a weighted R factor based on F2 values of all
unique data.
After filtration, the toluene was removed and the residue
washed twice with hexane and vacuum-dried yielding a red-
brown solid of Cp2Zr[ButCtCB(C6F5)2] (180 mg, 72%). 1H
NMR: δ 5.62 (s, 10H, C5H5), 1.16 (s, 9H, CCH3). 13C{1H}
NMR: δ 111.2 (C5H5), 40.9 (CCH3), 32.7 (CCH3). 11B NMR: δ
21.5. 19F NMR: δ -145.6 (s, br, 4F, o-F), -154.8 (t, 2F, p-F),
-161.6 (m, 4F, m-F). EI-MS: 647 (100) [M]+; 590 (20) [M -
But]+, 220 (55), [Cp2Zr]+, 57 (45) [But]+. Anal. Calcd for
of N2 (ca. 5 psi). Polymerizations were terminated by venting
the monomer and rapidly draining the polymer solution into
a small volume of methanol. The polyethylene was filtered,
washed with methanol, and dried in a vacuum oven for 24 h
at 80 °C.
P r oced u r e 2: In Situ Ca ta lyst Gen er a tion . A toluene
solution (25 mL) of AlMe3 and cocatalyst was transferred into
the reactor, containing toluene (450 mL), using a 50 mL sample
vessel. The solution was allowed to stir for at least 30 min at
the desired process temperature while saturating with mono-
mer. Polymerizations were initiated by injection of a solution
of 3 in toluene (25 mL) into the reactor using an over-
pressurized sample cylinder. Polymerizations were terminated
by venting the monomer and rapidly draining the polymer
solution into a small volume of methanol. The polyethylene
was filtered, washed with methanol, and dried in a vacuum
oven for 24 h at 80 °C.
P olym er Ch a r a cter iza tion . Polymer molecular weights
and distributions were determined by gel-permeation chro-
matography (GPC) using a Water 150C chromatograph on a
J ordi mixed-bed column (10-103 Å) employing a differential
refractive index detector at 145 °C in 1,2,4-TCB solution.
Sample dissolution (0.1% w/v) was accomplished by rotating
the samples in an oven operating at 160 °C for a few hours
with 0.1% BHT as the antioxidant. Samples were eluted at a
flow rate of 1.0 mL/min, and columns were calibrated using
both a broad MWD PE and narrow MWD PS standards.
C
28H19BF10Zr: C, 51.94; H, 2.96. Found: C, 51.50; H, 2.71.
X-r a y Cr ysta llogr a p h y. Measurements were made on a
Bruker AXS SMART CCD area-detector diffractometer using
graphite-monochromated Mo KR radiation (λ ) 0.710 73 Å).
The structures were solved by direct methods and refined on
F2 values by full-matrix least-squares for all unique data. Table
5 gives further details. Programs used were standard Bruker
SMART (control) and SAINT (integration); SHELXTL37 for
structure solution, refinement, and molecular graphics; and
local programs.
Eth ylen e P olym er iza tion s. Polymerizations were per-
formed in a 1 L Autoclave Engineers Zipperclave stainless steel
reactor equipped with an overhead stirrer and a spiral-wound
external cooling jacket. All polymerizations were conducted
at a stirring rate of 1000 rpm. A Neslab RTE-110 refrigerating
circulator passed a coolant mixture (ethylene glycol/water)
through the reactor jacket. An RTD-220 temperature control-
ler (Neslab) controlled the temperature ((1 °C), while a remote
sensor (Neslab RS2) monitored the reaction solution at the
temperature and cycled the circulator heater on or off as
required. The flow of monomer into the reactor (to maintain
a given head pressure) was measured using a mass flow meter
and a mass flow controller (Matheson Multiple Dynablender-
8219). The progress of polymerizations was monitored (mono-
mer flow rate(s) and internal temperature) using an IBM PC
with a data acquisition card and associated software.
P r oced u r e 1: P r em ixin g of Ca ta lyst a n d Coca ta lyst.
Toluene (500 mL) was presaturated with monomer at the
specified temperature and pressure (75 psi). A 25 mL amount
of a toluene solution of 3 and cocatalyst (ratio 1:1.2) was
prepared in a glovebox and transferred to a 50 mL stainless
steel sample vessel. Polymerizations were initiated by injec-
tion of the solution into the reactor using a slight overpressure
Ack n ow led gm en t. Funding for this work from the
Natural Sciences and Engineering Research Council of
Canada Strategic Project Grant and Novachem Ltd. of
Calgary, Alberta, Canada, is gratefully acknowledged.
W.C. acknowledges funding from the Engineering and
Physical Sciences Research Council, U.K. W.E.P. also
thanks the Alfred P. Sloan Foundation for a Research
Fellowship (1996-2000). K.K. acknowledges the DFG
for a Postdoctoral Fellowship.
Su p p or tin g In for m a tion Ava ila ble: Full tables of crys-
tallographic data, atomic parameters, hydrogen parameters,
atomic coordinates, anisotropic thermal parameters, and
complete bond distances and angles for 1a -c (26 pages).
Ordering information is given on any current masthead page.
(37) Sheldrick, G. M. SHELXTL, version 5; Bruker AXS Inc.:
Madison, WI, 1994.
OM980259E