Neutral and cationic group 4 complexes containing bis(borylamide) ligands, [R2BNCH2CH2NBR2]2-(R = 2,4,6-i-Pr3C2H2, M = Zr; R = cyclohexyl, M = Ti, Zr, Hf)

Article abstract of DOI:10.1021/om970765o

Reaction of Li₂(TripBen)(THF)₄ ([(2,4,6-i-Pr₃C₆H₂)2BNCH₂CH ₂NB(2,4,6-i-Pr₃C₆H₂) ₂]²~ = (TripBen)^2-) with ZrCl₄(THF)₂ in toluene cleanly provides (TripBen)ZrCl₂. (TripBen)ZrCl₂ can be alkylated with Grignard reagents to yield the dimethyl (2), diethyl (3), and dibutyl (4) derivatives, (TripBen)ZrR₂. The structure of (TripBen)Zr(CD₃)₂ reveals that π bonding between boron and nitrogen results in one Trip ring on each boron being oriented over the two alkyl groups. The reaction between (TripBen)ZrMe₂ and B(C₆F₅)₃ in pentane yields colorless crystals of {(TripBen)ZrMe}{MeB(C₆F₅)₃} (5a). The dialkyl complexes 2-4 react with [HNMe₂Ph][B(C⁶F₅)₄] in toluene to yield free dimethylaniline and species that are identical to 5a according to NMR spectra. (CyBen)M(CH₂SiMe₃)₂ complexes ([(cyclohexyl)₂- BNCH₂CH₂NB(cyclohexyl)₂]^2- = (CyBen)^2-) are prepared by the reaction between Li₂(CyBen)- (ether) and M(CH₂SiMe₃)₂Cl₂(ether)_x (M = Ti, Zr, or Hf). (CyBen)MI₂ complexes are formed upon treating (CyBen)M(CH₂SiMe₃)₂ complexes with iodine. A variety of (CyBen )MR₂ species can be prepared, including those in which the alkyl contains β protons. The molecular structure of (CyBen)Zr(CH₂CH₃)₂ shows it to be a relatively symmetric species in which one cyclohexyl ring on each boron is located directly over the two ethyl ligands. Diisobutyl complexes (CyBen)M(CH₂CHMe₂)₂ react with [Ph₃C][B(C₆F₅)₄] in chlorobenzene at -30°C to give the cations [(CyBen)M(CH₂CHMe₂)][B(C₆F₅) ₄], which are believed to be stabilized by coordination of chlorobenzene or toluene. Chlorobenzene solutions of [(CyBen)M(CH₂CHMe₂)- (S)][B(C₆F₅)₄] (M = Zr, Hf; S = chlorobenzene) polymerize 225 equiv of 1-hexene quantitatively at -30°C to give regioregular poly(1-hexene) of high molecular weight (M_n = (0.4- 1.3) × 10⁵ g/mol) and high polydispersity (5-6).

Full text of DOI:10.1021/om970765o

308
Organometallics 1998, 17, 308-321
Neu tr a l a n d Ca tion ic Gr ou p 4 Com p lexes Con ta in in g
Bis(bor yla m id e) Liga n d s, [R2BNCH2CH2NBR2]2- (R )
2,4,6-i-P r 3C6H2, M ) Zr ; R ) Cycloh exyl, M ) Ti, Zr , Hf)
Timothy H. Warren, Richard R. Schrock,* and William M. Davis
Department of Chemistry 6-331, Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139
Received August 27, 1997
Reaction of Li2(TripBen)(THF)4 ([(2,4,6-i-Pr3C6H2)2BNCH2CH2NB(2,4,6-i-Pr3C6H2)2]2-
)
(TripBen)2-) with ZrCl4(THF)2 in toluene cleanly provides (TripBen)ZrCl2. (TripBen)ZrCl2
can be alkylated with Grignard reagents to yield the dimethyl (2), diethyl (3), and dibutyl
(4) derivatives, (TripBen)ZrR2. The structure of (TripBen)Zr(CD3)2 reveals that π bonding
between boron and nitrogen results in one Trip ring on each boron being oriented over the
two alkyl groups. The reaction between (TripBen)ZrMe2 and B(C6F5)3 in pentane yields
colorless crystals of {(TripBen)ZrMe}{MeB(C6F5)3} (5a ). The dialkyl complexes 2-4 react
with [HNMe2Ph][B(C6F5)4] in toluene to yield free dimethylaniline and species that are
identical to 5a according to NMR spectra. (CyBen)M(CH2SiMe3)2 complexes ([(cyclohexyl)2-
BNCH2CH2NB(cyclohexyl)2]2- ) (CyBen)2-) are prepared by the reaction between Li2(CyBen)-
(ether) and M(CH2SiMe3)2Cl2(ether)x (M ) Ti, Zr, or Hf). (CyBen)MI2 complexes are formed
upon treating (CyBen)M(CH2SiMe3)2 complexes with iodine. A variety of (CyBen)MR2 species
can be prepared, including those in which the alkyl contains â protons. The molecular
structure of (CyBen)Zr(CH2CH3)2 shows it to be a relatively symmetric species in which one
cyclohexyl ring on each boron is located directly over the two ethyl ligands. Diisobutyl
complexes (CyBen)M(CH2CHMe2)2 react with [Ph3C][B(C6F5)4] in chlorobenzene at -30 °C
to give the cations [(CyBen)M(CH2CHMe2)][B(C6F5)4], which are believed to be stabilized by
coordination of chlorobenzene or toluene. Chlorobenzene solutions of [(CyBen)M(CH2CHMe2)-
(S)][B(C6F5)4] (M ) Zr, Hf; S ) chlorobenzene) polymerize 225 equiv of 1-hexene quantita-
tively at -30 °C to give regioregular poly(1-hexene) of high molecular weight (Mn ) (0.4-
1.3) × 105 g/mol) and high polydispersity (5-6).
In tr od u ction
We recently reported the synthesis of titanium and
zirconium complexes containing the di(borylamido)
ligand [Mes2BNCH2CH2NBMes2]2- ((MesBen)2-; Mes )
mesityl). We had hoped that dialkyl complexes of the
type (MesBen)MR2 would mimic (cyclopentadienyl)2MR2
species in their ability to form cationic species that
initiate the polymerization of terminal olefins. The
(MesBen)2- ligand is a structurally and electronically
Numerous studies of homogeneous Ziegler-Natta
catalysis by group 4 metallocenes have lead to the
identification of alkyl cations [Cp2MR]+ (Cp ) cyclo-
pentadienyl-like ligand; M ) Ti, Zr, Hf; R ) alkyl) as
the chain-propagating species.1-3 An ever growing
number of studies concern the design of specific mono-
(cyclopentadienyl) or bis(cyclopentadienyl) ligand envi-
ronments that lead to isotactic poly-R-olefins, or syn-
diotactic poly-R-olefins or even “blocky” variations.4 The
success of this chemistry has elicited questions concern-
ing the development of non-cyclopentadienyl complexes
that share some of the properties of metallocenes, in
particular, olefin polymerization activity. In this re-
gard, chelating bis(alkoxide)5-7 and bis(amido)8-19 com-
plexes have been receiving increased attention.
(8) Gue´rin, F.; McConville, D. H.; Payne, N. C. Organometallics
1996, 15, 5085.
(9) Gue´rin, F.; McConville, D. H.; Vittal, J . J . Organometallics 1996,
15, 5586.
(10) Scollard, J . D.; McConville, D. H.; Vittal, J . J . Organometallics
1995, 14, 5478.
(11) Warren, T. H.; Schrock, R. R.; Davis, W. M. Organometallics
1996, 15, 562.
(12) Aoyagi, K.; Gantzel, P. K.; Kalai, K.; Tilley, T. D. Organome-
tallics 1996, 15, 923.
(13) Horton, A. D.; de With, J . J . Chem. Soc., Chem. Commun. 1996,
1375.
(14) Horton, A. D.; de With, J .; van der Linden, A. J .; van de Weg,
H. Organometallics 1996, 15, 2672.
(15) Tinkler, S.; Deeth, R. J .; Duncalf, D. J .; McCamley, A. J . Chem.
Soc., Dalton Trans. 1996, 2623.
(16) Baumann, R.; Davis, W. M.; Schrock, R. R. J . Am. Chem. Soc.
1997, 119, 3830.
(17) Scollard, J . D.; McConville, D. H.; Payne, N. C.; Vittal, J . J .
Macromolecules 1996, 29, 5241.
(18) Scollard, J . D.; McConville, D. H. J . Am. Chem. Soc. 1996, 118,
10008.
(1) Brintzinger, H. H.; Fischer, D.; Mu¨lhaupt, R.; Rieger, B.;
Waymouth, R. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1143.
(2) Bochmann, M. J . Chem. Soc., Dalton Trans. 1996, 255.
(3) Kaminsky, W.; Ardnt, M. Adv. Polym. Sci. 1997, 127, 144.
(4) Kravchenko, R.; Masood, A.; Waymouth, R. M. Organometallics
1997, 16, 3635.
(5) van der Linden, A.; Schaverien, C. J .; Meijboom, N.; Ganter, C.;
Orpen, A. G. J . Am. Chem. Soc. 1995, 117, 3008.
(6) Fokken, S.; Spaniol, T. P.; Kang, H.-C.; Massa, W.; Okuda, J .
Organometallics 1996, 15, 5069.
(7) Tjaden, E. B.; Swenson, D. C.; J ordan, R. F.; Petersen, J . L.
Organometallics 1995, 14, 371.
(19) Mack, H.; Eisen, M. S. J . Organomet. Chem. 1996, 525, 81.
S0276-7333(97)00765-6 CCC: $15.00 © 1998 American Chemical Society
Publication on Web 01/07/1998

Group 4 Complexes Containing [R2BNCH2CH2NBR2]2-
Organometallics, Vol. 17, No. 3, 1998 309
unusual bidentate diamido ligand, since two mesityl
groups should sterically protect two or three coordina-
tion sites in a plane roughly perpendicular to the NCCN
ligand backbone as a consequence of N-B π bonding.20
Unfortunately, (MesBen)TiR2 and especially (MesBen)-
ZrR2 species were found to be relatively unstable. The
mode of decomposition consisted of CH activation in the
ortho methyl group in the mesityl substituent and loss
of 2 equiv of alkane. B(C6F5)3 was found to bind to a
methyl group in (MesBen)MMe2 complexes in dichlo-
romethane, but such compounds showed little polym-
erization activity toward ethylene at 25 °C and 1-2 atm,
presumably as a consequence of strong binding of
[B(C6F5)3Me]- to the metal.
Reaction of Li2(TripBen)(THF)4 with ZrCl4(THF)2 in
toluene cleanly provides (TripBen)ZrCl2 (1), which is
isolated in 85-90% yield from pentane as colorless
prisms containing 1 equiv of pentane per zirconium (eq
2). All resonances except those of the p-CHMe2 groups
Li2(TripBen)(THF)4 + ZrCl4(THF)2 toluene8
(TripBen)ZrCl2
(2)
1
are broadened in the room temperature 1H NMR
spectrum of 1. At 75 °C, a readily interpretable 1H
NMR spectrum is obtained in which a single sharp
backbone resonance is observed along with six doublets
for the isopropyl methyl groups. Six doublet resonances
are consistent with slow rotation about the B-N bonds
and slow rotation about the B-Caryl bonds on the NMR
time scale. We assume that the structure of 1 is
analogous to that of the dimethyl derivative (see below
and Figure 1) and that at low temperatures some lower
symmetry conformation (e.g., a preference for the Trip2B
groups to be oriented as shown in Figure 1) leads to
more complex spectra.
In this paper, we report two variations of [R2BN-
CH2CH2NBR2]2- ligands and group 4 metal complexes
thereof, one in which R ) 2,4,6-i-Pr3C6H2 ((TripBen)2-
)
and another in which R ) cyclohexyl ((CyBen)2-). The
large o-isopropyl substituents in TripBen might be
expected to produce a more crowded coordination envi-
ronment and thereby weaken the binding of an anion
such as [MeB(C6F5)3]- to the metal center. The
(CyBen)2- ligand was prepared because it was simple
to do so, because we expected its complexes to be highly
crystalline, and because cyclohexyl has no CH bonds
that should be activated readily. The small size of the
cyclohexyl groups also would test the possible benefits
of a more open coordination sphere. We report here the
synthesis and solution behavior of neutral and cationic
(TripBen)ZrR2 complexes and (CyBen)MR2 complexes
(M ) Ti, Zr, and Hf) and a brief exploration of their
polymerization activity with ethylene and 1-hexene.
Reaction of 1 in ether at -40 °C with the appropriate
Grignard reagent cleanly provides the dimethyl (2),
diethyl (3), and dibutyl (4) derivatives (eq 3). The
(TripBen)ZrCl2 + 2RMgCl ether/-40 °C8
(TripBen)ZrR2
(3)
R ) Me (2), Et (3), n-Bu (4)
dimethyl complex is isolated in high yield from pentane
as colorless crystals that contain 1 equiv of pentane per
Zr. However, 3 and 4 crystallize as fine, fibrous needles
which are difficult to isolate. They are more easily
recovered as off-white powders by exhaustive removal
of pentane from pentane solutions in vacuo. With the
Resu lts
Th e Syn th esis of H2(Tr ip Ben ) a n d (Tr ip Ben )Zr
Com p lexes. [Trip2BNHCH2CH2NHBTrip2] (H2(Trip-
Ben)) may be prepared in a manner analogous to that
reported for H2(MesBen), substituting Trip2BF for Mes2-
BF (Mes ) 2,4,6-trimethylphenyl; eq 1). Although the
1
exception of the distinct p-CHMe2 groups, H and 13C
NMR spectra of 2, 3, and 4 are again broad at room
temperature, presumably as a consequence of various
hindered rotations, as discussed above for 1. However,
all broad resonances sharpen upon heating a sample to
50-70 °C to give a spectrum consistent with a molecule
that contains two planes of symmetry but in which
rotation of the Trip group about the B-C bond and
rotation of the B(Trip)2 group about the B-N bond are
still slow on the NMR time scale, as observed in 1.
Resonances for the coordinated alkyl groups are also
broadened and in the case of 3 and 4 are partially
obscured by the CHMe2 resonances. In order to confirm
the identity of the ethyl resonances, (TripBen)Zr(CD2-
H2NCH2CH2NH2 (1) 2 BuLi/THF8
(2) 2 Trip2BF
Trip2BNHCH2CH2NHBTrip2 (1)
high solubility of H2(TripBen) in pentane hampers its
isolation in good yield, it need not be isolated. Depro-
tonation of crude H2(TripBen) with 2 equiv of butyl-
lithium in pentane containing a small amount of THF
proceeds cleanly to yield colorless crystals of Li2(TripBen)-
(THF)4. Room temperature 1H NMR spectra of H2-
(TripBen) and Li2(TripBen)(THF)4 consist of two sets
of sharp isopropyl resonances; other resonances are
broad. At ∼70 °C, two chemically inequivalent triiso-
propylphenyl groups may be readily identified and other
resonances sharpen to patterns consistent with a freely
rotating symmetric molecule. We propose that broad
resonances result from hindered aryl group rotation
about the B-Cipso bonds as a consequence of a steric
interaction between opposing o-isopropyl groups, al-
though there are other possibilities for the lithium
derivative such as rearrangement of the Li2N2 core.
2
CD3)2 was prepared and shown to have broad H NMR
resonances for the R and â deuterons in the ethyl groups
at δ 0.47 and 0.96 ppm, respectively, at room temper-
ature, which sharpen with increasing temperature. It
is important to note that 2, 3, and 4 are all much more
thermally stable toward formation of alkane via activa-
tion of benzylic protons than (MesBen)ZrR2 complexes,11
presumably because benzylic protons are relatively
inaccessible in the (TripBen)2- ligand. However, (Trip-
Ben)ZrR2 complexes will decompose upon heating, with
the rate of decomposition of 4 being qualitatively the
fastest.
(20) Bartlett, R. A.; Feng, X.; Olmstead, M. M.; Power, P. P.; Weese,
K. J . J . Am. Chem. Soc. 1987, 109, 4851.

310 Organometallics, Vol. 17, No. 3, 1998
Warren et al.
Ta ble 1. Cr ysta llogr a p h ic Da ta , Collection
P a r a m eter s, a n d Refin em en t P a r a m eter s for
(Tr ip Ben )Zr (CD3)2‚p en ta n e a n d
(CyBen )Zr (CH2CH3)2 (10a )
(TripBen)Zr(CD3)2
(CyBen)Zr(CH2CH3)2
empirical formula C69H114B2N2Zr
C30H58B2N2Zr
559.62
fw
1084.46
diffractometer
cryst color,
morphology
Siemens SMART/CCD Siemens SMART/CCD
colorless, plate
colorless, prismatic
cryst dimens, mm 0.30 × 0.22 × 0.12
0.28 × 0.24 × 0.18
triclinic
cryst syst
a, Å
triclinic
11.6475(7)
16.4594(10)
18.4034(10)
101.0020(10)
100.0960(10)
98.1700(10)
3352.9(3)
P1h
9.9911(2)
12.23940(10)
13.82900(10)
71.1280(10)
85.0270(10)
77.2990(10)
1560.80(4)
P1h
b, Å
c, Å
R, deg
â, deg
γ, deg
V, Å3
space group
Z
2
2
Dcalc
F000
1.074 g/cm3
1184
1.191 g/cm3
604
µ(Mo KR), mm-1 0.202
0.372
ω scans
183(2)
scan type
temp, K
total no.
ω scans
188(2)
7126
4364
unique reflns
no. of variables
643
317
R
Rw
GoF
0.0654
0.0927
1.093
0.0345
0.0970
1.234
Ta ble 2. Selected Bon d Dista n ces (In tr a m olecu la r
Con ta cts) (Å) a n d An gles (d eg) for
(Tr ip Ben )Zr (CD3)2 a n d (CyBen )Zr (CH2CH3)2
(TripBen)Zr(CD3)2 (CyBen)Zr(CH2CH3)2
Distances
Zr-N(1)
2.102(4)
2.111(4)
2.232(6)
2.229(6)
1.404(8)
1.398(8)
3.84
3.85
3.37
3.36
2.067(2)
2.070(2)
2.247(3)
2.253(3)
1.392(4)
1.398(4)
Zr-N(2)
Zr-C(1)
Zr-C(2)
N(1)-B(1)
N(2)-B(2)
Zr‚‚‚C(59)
Zr‚‚‚C(18)
C(1)‚‚‚C(59)
C(2)‚‚‚C(18)
Angles
78.4(2)
N(1)-Zr-N(2)
C(1)-Zr-C(2 or 3)
N(1)-Zr-C(1)
78.87(2)
100.6(2)
112.2(2)
127.6(2)
126.5(2)
113.4(2)
119.53(12)
115.13(10)
111.12(11)
110.92(11)
114.43(11)
114.8(2)
N(2)-Zr-C(1)
F igu r e 1. (a) ORTEP drawing of the structure of [Trip-
Ben]Zr(CD3)2. (b) Chem3D top view of the structure of
[TripBen]Zr(CD3)2 with the isopropyl groups removed.
N(1)-Zr-C(2 or 3)
N(2)-Zr-C(2or 3)
Zr-C(1)-C(2)
Zr-C(3)-C(4)
119.7(2)
Two views of the structure of (TripBen)Zr(CD3)2 are
shown in Figures 1a and 1b. (See also Tables 1 and 2.
The choice of the CD3 over the CH3 derivative was
circumstantial.) A space-filling view including hydrogen
atoms shows that this molecule is impressively crowded.
The triisopropylphenyl rings on a given boron are
twisted with respect to one another, and the two Trip2B
units are twisted with respect to one another. There-
fore, two triisopropylphenyl rings are located approxi-
mately above or below a methyl group on the metal. The
TripBen ligand symmetrically chelates the zirconium
center with a bite angle of 78.4(2)° and Zr-N distances
of 2.102(4) and 2.111(4) Å (Table 2). These distances
are similar to those found in the doubly metalated
species formed upon decomposition of (MesBen)ZrR2
(2.088(6) and 2.106(6) Å).11 Borylamide B-N π-interac-
tions are significant, as shown by the B-N distances
(1.404(8) and 1.398(8) Å). The Zr-Me distances (2.229-
(6) and 2.232(6) Å) are essentially identical to the Zr-C
distance in (t-Bu3SiNH)3ZrMe (2.231(7) Å).21 The short
Zr‚‚‚C(18) and Zr‚‚‚C(59) separations (3.84 and 3.85 Å)
and space-filling views reveal that the sides of the
coordination wedge are effectively blocked by the methyl
groups of the o-isopropyl substituents. Furthermore,
the relatively short separations between the isopropyl
and metal-bound methyl groups (C(1)‚‚‚C(59) ) 3.37 Å
and C(2)‚‚‚C(18) ) 3.36 Å) show that the groups in the
coordination wedge strongly feel the presence of the aryl
o-isopropyl substituents, which may help explain the
C(1)-Zr-C(2) angle of only 100.6(2)° compared to a
(21) Cummins, C. C.; Duyne, G. D. V.; Schaller, C. P.; Wolczanski,
P. T. Organometallics 1991, 10, 164.

Group 4 Complexes Containing [R2BNCH2CH2NBR2]2-
Organometallics, Vol. 17, No. 3, 1998 311
C-Zr-C angle of ∼120° in a related CyBen complex (see
below). However, a C(1)-Zr-C(2) angle of ∼100° is
similar to that typically found in a dialkylzirconocene22
or recently in a diamido complex reported by McCon-
ville.9
Gen er a tion of “Ca tion s”. The reaction between 2
and B(C6F5)3 in pentane results in formation of an oil
that crystallizes upon standing at -40 °C to afford
colorless crystals of 5a in 87% yield (eq 4). In addition
pentane provides a sample of 5b as a powder. 1H NMR
spectra of this solid are identical to those taken of 5b
prepared in situ, except that resonances corresponding
to dimethylaniline are absent. Spectra of 5b generated
in chlorobenzene-d5 are similar to those generated in
toluene-d8 and similar to NMR spectra of 5a . In fact,
1H and 13C NMR spectra of a mixture of 5a and 5b,
separately prepared in toluene-d8, show only resonances
for a single cationic species between 0 and 60 °C. It
seems unlikely that the two anions would behave
analogously in terms of binding to the metal to produce
an unsymmetric species. Therefore, the asymmetry
cannot be ascribed to coordination of the anion to the
metal, i.e., the cation in 5a and 5b are one in the same.
Several explanations that do not involve anion binding
are possible, however. (i) The solvent (toluene or
chlorobenzene) might coordinate to the metal. (ii) One
(reversibly) or two of the methyl groups in an o-isopropyl
group might be interacting in an agostic manner27 with
the metal center. In the model shown in Figure 1a,
C(18) (for example) can virtually take the place of C(2)
after a slight rotation about the B-Cipso bond. (iii) The
cation may actually be a nonplanar three-coordinate
species that inverts its configuration via a planar form
and, therefore, yield NMR spectra at higher tempera-
tures that are consistent with a higher symmetry
species and a symmetric pattern for the backbone proton
and carbon atoms. In the absence of an X-ray structure,
we have little hope of distinguishing between these
possibilities. Unfortunately, no suitable crystals have
been obtained to date.
to a slightly broadened AA′BB′ pattern ascribed to the
ligand backbone in 5a , 10 or 11 CHMe2 resonances can
be detected out of the 12 expected for an unsymmetri-
cally substituted coordination wedge, i.e., no CHMe2
groups within a given BTrip2 unit are equivalent and
the two BTrip2 groups are related by a mirror plane.
On the basis of several observations of “partial abstrac-
tion” of a methyl group by B(C6F5)3 in zirconium methyl
complexes23-26 and recently in a zirconium complex that
contains a tridentate diamido ether ligand,16 the
[MeB(C6F5)3]- “ligand” could be occupying one of the two
carbon-based coordination positions in the pseudotet-
rahedral species, as shown in eq 4. Heating an NMR
sample of 5a in toluene leads to a symmetric backbone
on the NMR time scale, a fact that would be consistent
with either loss of the B(C6F5)3 group or the [MeB-
(C6F5)3]- ion on the NMR time scale. However, other
explanations are possible and are delayed until the
discussion of 5b below.
Attempts to generate cationic species by adding
[Ph3C][B(C6F5)4] to TripBen complexes appeared to fail.
No species analogous to those prepared using B(C6F5)3
or [HNMe2Ph][B(C6F5)4] were obtained.
Room temperature 1H NMR spectra of 6 and 7 in
toluene-d8 exhibit an additional feature not present in
the spectra of 5a and 5b. Each displays a slightly
broadened upfield resonance, which in the case of 6 is
a broadened triplet at δ -0.64. In [(TripBen)Zr(CD2-
CD3)][B(C6F5)4], the upfield resonance is absent. We
assign the upfield resonance to the â-H atoms of the
ethyl ligand. This situation is related to that observed
in complexes of the type [Cp2M(R)(PMe3)][BPh4] (R )
Et, n-Bu, i-Bu),28,29 in which the â-H resonance is shifted
upfield in each alkyl as a consequence of â-agostic
interactions.27 The dense 13C NMR spectrum of 6
unfortunately prevents unambiguous assignment of the
ZrCH2CH3 resonances and confirmation of â-agostic
behavior. The AA′BB′ pattern of the ligand backbone
in 5b, 6, and 7 coalesces into a broad resonance in the
temperature range 60-80 °C; the exact coalescence
temperature depends on the nature of the complex.
Further sharpening of the backbone resonance is ob-
served with increasing temperature. Unfortunately,
additional resonances irreversibly grow in that signal
the onset of decomposition, and the origin of the
temperature-dependence remains obscure. In addition
to temperature dependent processes analogous to those
The dialkyls 2-4 react with [HNMe2Ph][B(C6F5)4] in
toluene to yield species that behave qualitatively like
5a behaves (eq 5). The dimethylaniline generated in the
(TripBen)ZrR2 + [HNMe2Ph][B(C6F5)4] toluene/-40 °C8
-RH
-Me2NPh
[(TripBen)ZrR][B(C6F5)4]
(5)
R ) Me (5b), Et (6), n-Bu (7)
protonolysis does not coordinate to the metal in cations
5b, 6, and 7; the chemical shift of its methyl groups is
characteristic of “free” dimethylaniline. We have not
been able to isolate 5b, 6, or 7 in crystalline form,
although concentration of a toluene solution of 5b to an
oil followed by exhaustive washing of this oil with
(22) Collman, J . P.; Hegedus, L. S.; Norton, J . R.; Finke, R. G.
Principles and Applications of Organotransition Metal Chemistry, 2nd
ed.; University Science Books: Mill Valley, CA, 1987.
(23) Yang, X.; Stern, C. L.; Marks, T. J . J . Am. Chem. Soc. 1991,
113, 3623.
(24) Yang, X.; Stern, C. L.; Marks, T. J . J . Am. Chem. Soc. 1994,
116, 10015.
(25) Bochmann, M. Angew. Chem., Int. Ed. Engl. 1992, 31, 1181.
(26) Bochmann, M.; Lancaster, S. J .; Hursthouse, M. B.; Malik, K.
M. A. Organometallics 1994, 13, 2235.
(27) Brookhart, M.; Green, M. L. H.; Wong, L.-L. Prog. Inorg. Chem.
1988, 36, 1.
(28) J ordan, R. F.; Bradley, P. K.; Baenziger, N. C.; LaPointe, R. E.
J . Am. Chem. Soc. 1990, 112, 1289.
(29) Guo, Z.; Swenson, D. C.; J ordan, R. F. Organometallics 1994,
13, 1424.

312 Organometallics, Vol. 17, No. 3, 1998
Warren et al.
Sch em e 1. Syn th esis of H2(CyBen ) a n d
Li2(CyBen )(eth er )
tures, a significant amount of a byproduct that is
believed to be (CyBen)2Ti severely limits this method
as a synthetic route to (CyBen)TiCl2. Reaction of ZrCl4
or ZrCl4(THF)2 with Li2(CyBen)(ether) in a variety of
solvents in our hands does not cleanly lead to the
desired dichloro complexes (CyBen)ZrCl2(THF)x (x ) 1
or 2). These findings contrast with those obtained in
the MesBen11 or TripBen systems and we assume result
from a smaller steric demand of the CyBen ligand.
Similar results have been reported for other diamido
ligand systems.12
Preliminary studies showed that Li2(CyBen)(ether)
reacts with ZrR2Cl2(ether)2 (R ) CH2Ph, CH2CMe3)31
to provide the corresponding (CyBen)ZrR2 complexes,
but addition of Li2(CyBen)(ether) to solutions of M(CH2-
32
SiMe3)2Cl2(ether)2 prepared in situ proved to be the
most convenient synthetic entry into group 4 metal
chemistry (eq 6). The zirconium and hafnium deriva-
Li (CyBen)(ether)8
2
M(CH2SiMe3)2Cl2(ether)x
M ) Zr (x ) 2), Hf (x ) 2), or Ti
ether
(CyBen)M(CH2SiMe3)2
(6)
M ) Ti (8c), Zr (8a ), Hf (8b)
tives may be isolated in 70-80% yield as colorless
crystals from pentane, but the yield of the titanium
derivative (8c) is only ∼50%. The (trimethylsilyl)methyl
protecting groups can be cleaved off by adding 2
equivalents of I2 to afford 9a and 9b as colorless crystals
in 80-90% yield and 9c as golden yellow needles in 55%
yield (not optimized; eq 7).
2I2/CH2Cl2
(CyBen)M(CH2SiMe3)2 -2Me SiCH I8
proposed in 1, various possible agostic interactions
would also lead to temperature-dependent NMR behav-
ior.
3
2
(CyBen)MI2
M ) Ti (9c), Zr (9a ), Hf (9b)
(7)
Solutions of 5b, 6, and 7 are not active for the
polymerization of ethylene. Only traces of polyethylene
are recovered from solutions exposed to one atm of
ethylene over a period of 15 min to 1 h. We speculate
that agostic interactions prevent olefin attack at the
metal, a circumstance that is encouraged by the ex-
tremely crowded TripBen coordination sphere. This
finding led to our increased interest in the more open
diamido coordination environment offered by CyBen
complexes.
Syn th esis of H2(CyBen ) a n d (CyBen )Zr Com -
p lexes. Hydroboration of cyclohexene with commer-
cially available H2BCl(SMe2) is a convenient route to
an ethereal solution of Cy2BCl,30 which may be added
to a suspension of LiNHCH2CH2NHLi (generated in situ
from butyllithium and ethylenediamine) to yield crys-
talline H2(CyBen) in ∼70% yield (Scheme 1). The
lithium salt, Li2(CyBen)(ether), may be prepared in 95%
yield by deprotonating H2(CyBen) with 2 equiv of
butyllithium in pentane containing ∼5% ether (Scheme
1). Both syntheses can be carried out readily on a large
scale.
The zirconium and hafnium diiodides are useful
precursors to a variety of dialkyl complexes (eq
8). Compounds 12a and 12b are oils, but the others can
(CyBen)MI2 + 2RMgX f (CyBen)MR2
(8)
M ) Zr (a ) or Hf (b); R ) ethyl (10a ,b), CH2CHMe2
(11a ,b), CH2CH2Bu (12a ,b), CHBu2 (13a ); X )
Cl or Br
be isolated as colorless crystals in 70-90% yield from
pentane at -30 °C. Attempts to prepare methyl deriva-
tives employing MeMgBr or MgMe2 were unsuccessful.
For example, although (CyBen)ZrMe2 could be identified
in product mixtures by its characteristic 1H NMR
resonances, it could not be isolated from a significant
amount of Cy2BMe (δ 0.68 ppm, BMe). These results
suggest that alkylation at boron competes with alkyla-
tion at zirconium when methylating agents are em-
ployed. In contrast, pale yellow crystalline (CyBen)-
TiMe2 (14c) may be prepared in 90% yield from 9c and
Addition of Li2(CyBen)(ether) to TiCl4 in pentane does
not lead to clean formation of (CyBen)TiCl2. Although
(CyBen)TiCl2 may be isolated from the reaction mix-
(31) Wengrovius, J . H.; Schrock, R. R. J . Organomet. Chem. 1981,
205, 319.
(30) Brown, H. C.; Ravindran, N.; Kulkarni, S. U. J . Org. Chem.
1979, 44, 2417.
(32) Brand, H.; Capriotti, J . A.; Arnold, J . Organometallics 1994,
13, 4469.

Group 4 Complexes Containing [R2BNCH2CH2NBR2]2-
Organometallics, Vol. 17, No. 3, 1998 313
nances at room temperature, as expected if B-N bond
rotation is slow on the NMR time scale. Upon heating
samples of 10a and 11a in bromobenzene-d5 to 50-60
°C, the eight cyclohexyl resonances coalesce into four
and barriers for B-N bond rotation (∆Gqrot) of 16.2(2)
and 16.3(2) kcal/mol, respectively, can be calculated. In
contrast, each titanium complex exhibits three relatively
sharp 13C cyclohexyl resonances (the B-C resonance
was generally too broad to be observed at room tem-
perature), indicating more facile rotation of the dicy-
clohexylboryl units relative to those of the zirconium and
hafnium derivatives.
The molecular structure of 10a is shown in Figure 2,
and distances and angles are listed in Table 2. The two
planar boryl fragments are linked by the puckered,
ethylene bridge and are oriented so that one cyclohexyl
ring is located directly over the two ethyl ligands and
one ring is located over the ethylene bridge; the dihedral
angle between the two BC2 planes is only 10°, as
opposed to 62° in (TripBen)Zr(CD3)2. The Zr-N dis-
tances (2.067(2) and 2.070(2) Å), Zr-C distances (2.247-
(3) and 2.253(3) Å), and N(1)-Zr-N(2) angle (78.87(2)°)
are similar to those in (TripBen)Zr(CD3)2 (Figure 1). For
comparison, in (MesBen)Ti(benzyl)Cl11 the N(1)-Ti-
N(2) angle is slightly larger (89.2(2)°) and the C-Ti-Cl
angle (125.5(2)°) is slightly larger than the C-Zr-C
angles, perhaps in part because the titanium is inter-
acting with Cipso in the benzyl ligand. The “in,in”
conformation of the two ethyl ligands in 10a is not a
consequence of any Câ-H interaction with the metal,
since the Zr-CR-Câ angles are 114.8(2)° and 119.7(2)°
and the Zr‚‚‚Câ distances are 3.20 and 3.28 Å, respec-
tively. The in,in orientation appears to be much less
desirable from a steric point of view, so it would appear
that some weak attractive interaction between the two
cyclohexyl rings and the two ethyl ligands would have
to be invoked in order to explain the observed structure.
It should be noted that the cyclohexyl rings do not
protect the N2ZrC face of the pseudotetrahedron to the
degree that is found in MesBen11 or TripBen complexes.
However, accessibility of the benzylic CH bonds in the
dimesitylboryl complexes leads to loss of alkane and
orthometalation, even when the alkyl on the metal (Ti
or Zr) contains no â hydrogens (e.g., (trimethylsilyl)-
methyl).11 In the CyBen complexes, there is no readily
accessible CH bond that could be activated and cleaved.
F or m a tion of Alk yl Ca tion s. Addition of [Ph3C]-
[B(C6F5)4]33-35 to 11a in chlorobenzene-d5 (L) at -30 °C
cleanly produces what we propose to be a cationic
complex, 11a +(L) (eq 10; Figure 3; Table 3), along with
1 equiv of Ph3CH and variable amounts of isobutylene.
F igu r e 2. (a) ORTEP drawing of the structure of (CyBen)-
Zr(CH2CH3)2 (10a ). (b) Chem3D top view of the structure
of (CyBen)Zr(CH2CH3)2 (10a ).
Me2Mg in dichloromethane (eq 9). Although 9c reacts
(CyBen)TiI2 + MgR2 dichloromethane8
-30 °C
(CyBen)TiR2
R ) CH2CHMe2 (11c)
(9)
Me (14c)
with Mg(i-Bu)2 to afford (CyBen)Ti(CH2CHMe2)2 (11c)
in 87% yield, addition of MgEt2 to 9c in dichloromethane
at -30 °C leads to an initially light yellow reaction
mixture that turns brown at room temperature and
from which putative (CyBen)Ti(CH2CH3)2 could not be
isolated. Solutions of the zirconium and hafnium
complexes show no signs of significant decomposition
upon standing overnight in benzene-d6 at room tem-
perature, while 11c decomposes to give a black suspen-
sion under the same conditions.
1H NMR spectra of the (CyBen)MR2 complexes show
a single sharp resonance for the CyBen ligand backbone
methylene protons and resonances for the alkyl group
that are consistent with the presence of two mirror
planes in a pseudotetrahedral species. The cyclohexyl
groups are inequivalent in the Zr and Hf complexes, as
judged by the presence of eight cyclohexyl 13C reso-
We assume that 1 equiv of isobutylene is formed,
although the detected amount is always less than 1
equiv. At -30 °C the 1H NMR spectrum of 11a +(L)
shows sharp, well-resolved AA′BB′ backbone resonances
(at 3.46 and 3.79 ppm) consistent with a cation that

314 Organometallics, Vol. 17, No. 3, 1998
Warren et al.
o-C6H4)2O]2- ([NON]2-) ligand16,36 and 15-20 ppm
downfield of where it appears in the neutral isobutyl
complex.
Reaction of [Ph3C][B(C6F5)4] with 11a in toluene-d8
at room temperature yields what we propose is the
cationic complex [(CyBen)Zr(CH2CHMe2)(toluene)]-
[B(C6F5)4] (11a +(tol)) (Table 3) along with 1.0 equiv of
Ph3CH and a variable amount of isobutylene. The
ligand backbone methylene resonances are revealed as
a sharp, well-resolved AA′BB′ pattern (at 3.27 and 3.78
ppm) analogous to that observed in a sample prepared
in chlorobenzene-d5 at -30 °C, consistent with the
presence of a relatively non-labile coordinating ligand
(L) next to the alkyl ligand in the coordination wedge.
19F NMR spectra show three sharp resonances for the
[B(C6F5)4]- anion, consistent with its isotropic character
on the NMR time scale. The chemical shift of the
isobutyl Câ-H resonance in 11a +(tol) at δ 1.97 ppm is
not consistent with any â-agostic C-H interaction.27,28
The temperature at which the right and left methyl-
ene backbone protons exchange on the NMR time scale
in 11a + in chlorobenzene-d5 increases steadily by 10-
20 °C upon addition of small amounts of toluene (0.5-2
equiv), i.e., toluene effectively slows the methylene
backbone proton equilibration process at a given tem-
perature by binding to the metal more strongly than
chlorobenzene binds. The addition of 1-2 equiv of
toluene to a 0.01-0.1 M solution of 11a + in chloroben-
zene-d5 should not change the bulk properties of the
solvent sufficiently to alter the degree of association of
the anion with the metal, if the anion were indeed the
ligand L. Finally, addition of 1.1 equiv of dimethyla-
niline to a sample of 11a + in chlorobenzene-d5 leads to
formation of a species that contains relatively strongly
coordinated dimethylaniline (Me at 2.80 ppm; Table 3),
along with a trace of free aniline (Me at 2.67 ppm). Free
and coordinated aniline do not exchange rapidly on the
NMR time scale. All of these observations are consis-
tent with the binding of some base (solvent or added
base) to the metal center in 11a +, presumably because
no agostic interactions in the cyclohexyl ring or the
isobutyl ligand are favorable, and the large [B(C6F5)4]-
anion binds much more weakly to the metal than
solvent.
F igu r e 3. Variable-temperature 1H NMR (300 MHz)
spectra of 11a + in chlorobenzene-d5.
contains only one plane of symmetry, as expected if only
one alkyl is present in a pseudotetrahedral cationic
species. The fourth position in the pseudotetrahedron,
we propose, is occupied by a chlorobenzene ligand. At
room temperature or above, the spectrum simplifies to
one that is consistent with a pseudotetrahedral species
that contains two mirror planes analogous to the
spectrum of 11a itself. (Isobutylene is not observed in
solution at 40 °C, presumably because it is too volatile.)
Except for some variation in the amount of isobutylene
in solution, the low-temperature spectrum is regener-
ated upon cooling the sample. We propose that the
“right” and “left” set of backbone methylene protons
equilibrate as a consequence of loss of ligand L and
formation of a pseudotrigonal planar intermediate. On
the basis of coalescence of the AA′ and BB′ backbone
resonances at 4 °C (at 300 MHz), we can calculate a
value for ∆Gq of 12.8(3) kcal/mol for the equilibration
process. The coalescence temperature did not change
over a 6-fold change in concentration of 11a ((1.3-8.0)
× 10-2 M). Low-temperature 19F NMR spectra of these
chlorobenzene-d5 solutions are superimposable upon
spectra of [Ph3C][B(C6F5)4] under the same conditions
and show no appreciable broadening other than that
which may be attributed to increased solvent viscosity
at low temperature. On the basis of this result and data
in other solvents reported below, the [B(C6F5)4]- ion does
not appear to interact significantly with the metal. The
13C NMR spectrum reveals the isobutyl R carbon atom
at 86.24 ppm, close to where it is observed in a 13C-
labeled zirconium cation that contains the [(t-Bu-d6-N-
The cationic hafnium (11b+) and titanium (11c+)
isobutyl complexes can be generated by analogous
methods. In the 13C NMR spectrum of 11b+ in chlo-
robenzene-d5 at -30 °C the HfCH2CHMe2 resonance is
observed at 85.27 ppm, ∼15-20 ppm downfield of where
it appears in the neutral isobutyl complex (Table 3).
1
Variable-temperature H NMR spectra reveal that the
barrier to symmetrization increases in the order Ti <
Zr < Hf (Table 3), which is what one would expect on
the basis of the strength of the M-Base bond and a rate-
limiting loss of base. This same trend is observed
qualitatively in terms of the increase in Tc upon adding
toluene to 11+ in chlorobenzene-d5; titanium is the least
sensitive to added toluene, while hafnium is the most
sensitive.
The reaction of 10a and 10b or 12a and 12b with
[Ph3C][B(C6F5)4] in chlorobenzene-d5 at -30 °C affords
the corresponding ethyl and hexyl cations, respectively.
The reaction between (CyBen)ZrEt2 and [Ph3C][B(C6F5)4]
(33) Chien, J . C. W.; Tsai, W. M.; Rausch, M. D. J . Am. Chem. Soc.
1991, 113, 8570.
(34) Bochmann, M.; Lancaster, S. J . J . Organomet. Chem. 1992, 434,
C1.
(35) J ordan, R. F. Adv. Organomet. Chem. 1991, 32, 325.
(36) Baumann, R.; Schrock, R. R. J . Organomet. Chem., in press.

Group 4 Complexes Containing [R2BNCH2CH2NBR2]2-
Organometallics, Vol. 17, No. 3, 1998 315
Ta ble 3. 1H NMR P a r a m eter s for Alk yl Ca tion s
solvent
(added base)
∆Gq
sym
other resonances
complex temp (°C)
δΜe(coord)
δMe(free) δ (AA′) δ (BB′) Tc (°C) (kcal/mol)
11a +
-30
C6D5Cl
3.79
3.46
4
12.8(3)
2.01 (m)
1.07 (d)
0.85 (d)
86.24 (13C)
1.97 (m)
0.96 (d)
0.85 (d)
2.08 (m)
1.08 (d)
0.93 (d)
CHMe2
ZrCH2
CHMe2
ZrCH2
CHMe2
ZrCH2
CHMe2
CHMe2
ZrCH2
CHMe2
11a +
11a +
25
25
C6D5CD3
3.78
3.99
3.27
3.48
C6D5Cl
(1.1 Me2NPh)
2.80
2.67
-0.44 (br m) BCH
11b+
11c+
-30
-30
C6D5Cl
C6D5Cl
3.78
4.13
3.19
3.65
32
14.3(3)
12.2(3)
2.14 (m)
0.93 (d)
CHMe2
CHMe2
HfCH2
HfCH2
CHMe2
TiCH2
CHMe2
TiCH2
ZrCH2CH3
BCH
HfCH2CH3
HfCH2
BCH
CH2CH3
BCH
CH2CH3
HfCH2
BCH
0.88 (d)
85.27 (13C)
2.10 (br)
1.89 (br)
0.77 (br)
112.71 (13C)
1.39 (t)
-13
10a +
10b+
-30
-30
C6D5Cl
(5 C6H5CH3)
C6D5Cl
2.21
2.26
2.17
2.18
3.80
3.86
3.29
3.35
28
61
14.1(3)
15.8(3)
-0.47 (br)
1.54 (t)
0.45 (br q)
-0.24 (br)
0.87 (t)
(5 C6H5CH3)
12a +
12b+
-30
-30
C6D5Cl
(5 C6H5CH3)
C6D5Cl
2.26
2.29
2.18
2.17
3.86
3.87
3.33
3.35
39
72
14.7(3)
16.3 (3)
-0.29 (br)
0.86 (t)
(5 C6H5CH3)
0.58 (br m)
-0.21 (br)
produces 0.5 equiv of Ph3CEt and 0.5 equiv of Ph3CH
plus a variable amount of ethylene in toluene. When
(CyBen)Zr(CD2CD3)2 was employed, Ph3C(CD2CD3) and
Ph3CD were formed, consistent with competition be-
tween a combination of an ethyl radical and trityl
radical and abstraction of a deuterium atom by a trityl
radical. In the case of 12a and 12b, only Ph3CH is
observed, consistent with predominant abstraction of a
hydrogen atom from a hexyl radical versus a radical/
radical combination. The ethyl and hexyl cations are
less stable than the isobutyl cations, however. (We
speculate that â-agostic interactions in the alkyl ligands
in 10+ and 12+ lead to activation of the â hydrogens and
migration of one of them to some part of the CyBen
ligand, perhaps to the amido nitrogen, which would
constitute a “reduction” of the metal, as in any â-hy-
drogen elimination process.) However, if 10+ and 12+
are generated in the presence of 5 equiv of toluene at
-30 °C, then 1H NMR spectra can be obtained (Table
3) in which the exchange of free and coordinated toluene
is slow on the NMR time scale. Toluene methyl
resonances for both bound (δ 2.29-2.21 ppm) and free
(δ 2.18 ppm) toluene are observed at -30 °C in each
complex. At slightly higher temperatures, only a single
resonance is observed for free and coordinated toluene,
consistent with rapid exchange of toluene on the NMR
time scale. As in the case of cations 10a ,b+, the metal-
arene interaction is stronger for 10b+ and 12b+ than
for 10a + and 12a +. Perhaps in contrast to what might
be expected on steric grounds, toluene binds most
strongly to the hexyl derivatives than to the ethyl
derivatives. That would be consistent with a competi-
tion between a â-agostic interaction and solvent binding,
since the â-agostic interaction is believed to be most
efficient in an ethyl complex.27,37
served in all cases. (CyBen)Zr(CHBu2)2 (13a ) and
(CyBen)TiMe2 (14c) did not react cleanly with [Ph3C]-
[B(C6F5)4] in chlorobenzene, even in the presence of 5
equiv of toluene. Although Ph3CH and Ph3CMe could
be identified in 1H NMR spectra of reaction mixtures
of 13a and 14c with [Ph3C][B(C6F5)4], respectively, no
CyBen-based cation could be identified unambiguously
in either case. Cationic species also could not be isolated
employing [Me2PhNH][B(C6F5)4] or B(C6F5)3 as activa-
tors, in spite of the apparent formation of a dimethyl-
aniline adduct of a cation as noted above. The reasons
are not known.
P olym er iza tion of 1-Hexen e a n d Eth ylen e. Po-
lymerization of 1-hexene was explored under conditions
where cations 10a +, 10b+, or 11a + could be generated
and observed in the NMR studies discussed above.
Addition of 1-hexene to such samples at temperatures
between 0 and 25 °C irreproducibly produced little poly-
(1-hexene) (yields 5-25%) when the monomer/solvent
volume ratio was lower than 1/7. GPC analysis of
polymers generated at 25 °C with a monomer/cation (1-
hexene/11a +) ratio of 200-300 showed that the molec-
ular weights (Mn ) 3 × 105 to 2 × 106 g/mol) and
polydispersity indices (2.5-6) of these polymers were
much higher than expected. We conclude that although
initiator cations can be observed at 25 °C in the absence
of 1-hexene, propagating cations must not be stable in
the presence of dilute 1-hexene at 25 °C.
Increasing the concentration of 1-hexene to 1.5 mL
per 5 mL of chlorobenzene and lowering the tempera-
ture to -30 °C led to almost quantitative yields of poly-
(1-hexene) with 10a +, 10b+, and 11a + as initiators
(Table 4). However, the molecular weight distribution
of the poly(1-hexene) in each case was very broad and
(37) Burger, B. J .; Thompson, M. E.; Cotter, W. D.; Bercaw, J . E. J .
Am. Chem. Soc. 1990, 112, 1566.
Clean formation of cationic complexes was not ob-

316 Organometallics, Vol. 17, No. 3, 1998
Warren et al.
Ta ble 4. P olym er iza tion of 1-Hexen e w ith
Activa ted (CyBen )MR2 Sp ecies
No polymerization activity toward ethylene was noted
for any species containing the TripBen ligand. Polym-
erization activity was assessed qualitatively for [(CyBen)-
Zr(CH2CHMe2)][B(C6F5)4]. In both toluene and chlo-
robenzene ethylene was polymerized immediately but
attempts to prepare polymer on a preparative scale did
not produce more than 100-150 mg of polyethylene as
a consequence of catalyst decomposition. The molecular
weight distribution was not determined.
temp yield 104Mn 104Mw
(CyBen)MR2 solvent
(°C)
(%)
(g/mol) (g/mol) PDI
10a
10b
11a
11a
11a
PhCla
-30
-30
-30
0
96
96
97
45
58
3.58
12.7
5.62
12.4
2.42
20.7
71.9
30.9
16.2
5.78b
5.68b
5.49
PhCla
PhCla
CH2Cl2
CH2Cl2
c
1.31d
1.53d
0
3.67
a
See Experimental Section for a complete description of the
polymerization experiments; 1.50 mL of hexene (225 equiv), 5.0
mL of solvent, 60 min reaction time, quenched with 1 mL of MeOH,
Mn(expected) ) 1.89 × 104 if PDI ) 1.0. b A resolved, low molecular
weight peak consisting of less than 7% of sample weight was
excluded from analysis. c Generated in the presence of 1-hexene
Discu ssion
The stability of complexes prepared here that contain
â protons seems unusual. However, some metallocene
dialkyls have been isolated when the alkyl contains a â
proton,29,41 and stable diamido complexes have been
known for some time.42,43 It is beginning to appear that
stability of diamido group 4 complexes to â elimination
is more often the rule than the exception, at least for
Zr and Hf, the least reducible of the group 4 metals.
d
and quenched after 10 min. Only the low molecular weight peak
was used in calculating PDI.
biased toward low molecular weights. The 13C NMR
spectra (in chloroform-d1) of the poly(1-hexene) prepared
from 11b+ and 11a + show primarily six resonances
(40.24, 34.60, 32.37, 28.71, 23.23, and 14.18 ppm)
characteristic of some degree of isotacticity.5,38-40 1H
NMR spectra of the poly(1-hexene) prepared from 11a +
as the initiator showed the presence of both disubsti-
tuted terminal (δ 4.6, 4.7) and 1,2-disubstituted internal
(δ 5.3) olefinic protons,38 which correspond roughly to
one double bond for every 300 monomer repeat units.
We think it likely that â-hydride elimination is the
origin of the formation of olefins. Intermediates in the
polymerization process could not be observed, and we
made no effort to further elucidate the details of the
decomposition process.
Although [(CyBen)Zr(CH2CHMe2)]+ (11a +) could not
be observed in dichloromethane-d2, even in the presence
of 15 equiv of toluene, addition of [Ph3C][B(C6F5)4] to
11a in dichloromethane followed by 1-hexene does lead
to rapid heat evolution and formation of poly(1-hexene)
in yields between 50% and 75%. In a typical experi-
ment, approximately one-half of the 1-hexene was
polymerized after 10 min at room temperature. GPC
analysis showed that the molecular weight distribution
of this polymer was trimodal, with most of the polymer
having a relatively low molecular weight approxmiately
corresponding to the number of monomers added.
Although Mn for this peak was only somewhat lower
A variety of group 4 metal cations have now been
observed in solution or isolated as base adducts.35 In
some cases the base has been an arene. For example,
[CpMMe2(arene)][MeB(C6F5)4] (Cp
)
Cp*,44 1,3-
(TMS)2C5H345) complexes are known in which coordi-
nated and free toluene do not exchange rapidly at room
temperature. The linked Cp-amide cation, {[(Me4C5)-
SiMe2(NCMe3)]ZrMe}+, also binds toluene in the coor-
dination site adjacent to the Zr-Me group and only
reluctantly undergoes exchange at elevated tempera-
tures in toluene-d8.46 Agostic interactions27 (both R47
and â35) are now believed to be relatively common in
neutral or cationic species that contain metals as far to
the left as scandium37 and as far to the right as cobalt.48
In comparison, however, group 4 transition-metal cat-
ions that do not contain at least one cyclopentadienyl
ring are rarely observed.13,14 Perhaps the most inter-
esting are pseudo-four-coordinate zirconium cations that
contain the [(t-Bu-d6-N-o-C6H4)2O]2- ([NON]2-) ligand,16
as they are relatively stable toward â-hydride elimina-
tion but they are still active for the polymerization of
1-hexene and can be observed in the form of living
polymers during a 1-hexene polymerization reaction.36
Titanium complexes that contain diamido ligands such
as [ArNCH2CH2CH2NAr]2- (Ar ) 2,6-i-Pr2C6H3)17,18 are
also initiators for the living polymerization of 1-hexene
in the presence of excess 1-hexene. The cationic species
in the titanium system is believed to be an unstable
pseudo-three-coordinate species, although it has not yet
been observed directly by NMR methods. In these
diamido catalyst systems, the amido substituents are
relatively bulky, thereby reducing the possibility of side
reactions involving the amido nitrogen lone pair.
than that found for corresponding polymerizations in
4
chlorobenzene (3.65 × 10
g/mol), it was far more
narrow with a PDI of 1.18. The polymerization process
was not improved by cooling the reaction to -30 °C or
by adding toluene (5-15 equiv) to the reaction medium.
1H and 13C NMR spectra (in chloroform-d1) of the poly-
(1-hexene) produced in dichloromethane reveal prima-
rily internal olefinic resonances (5.5-5.0 ppm) in 1H
NMR spectra and extremely broad signals in 13C NMR
spectra. Poly(1-hexene) prepared in dichloromethane
is a viscous oil, in contrast to the rubbery solid produced
in chlorobenzene. We conclude that dichloromethane
is a poor solvent for the polymerization of 1-hexene, and
we have considerable doubt concerning the nature of the
catalytically active species.
(41) Erker, G.; Schlund, R.; Kru¨ger, C. Organometallics 1989, 8,
2349.
(42) Planalp, R. P.; Andersen, R. A.; Zalkin, A. Organometallics
1983, 2, 16.
(43) Andersen, R. A. Inorg. Chem. 1979, 18, 2928.
(44) Gillis, D. J .; Tudoret, M.-J .; Baird, M. C. J . Am. Chem. Soc.
1993, 115, 2543.
(45) Lancaster, S. J .; Robinson, O. B.; Bochmann, M.; Coles, S. J .;
Hursthouse, M. B. Organometallics 1995, 14, 2456.
(46) J ia, L.; Yang, X.; Stern, C. L.; Marks, T. J . Organometallics
1997, 16, 842.
(47) Grubbs, R. H.; Coates, G. W. Acc. Chem. Res. 1996, 29, 85.
(48) Tanner, M. J .; Brookhart, M.; DeSimone, J . M. J . Am. Chem.
Soc. 1997, 119, 7617.
(38) Babu, G. N.; Newmark, R. A.; Chien, J . C. Macromolecules 1994,
27, 3383.
(39) Asakura, T.; Demura, M.; Nishiyama, Y. Macromolecules 1991,
24, 2334.
(40) Coughlin, E. B.; Bercaw, J . E. J . Am. Chem. Soc. 1992, 114,
7606.

Group 4 Complexes Containing [R2BNCH2CH2NBR2]2-
Organometallics, Vol. 17, No. 3, 1998 317
placed in the freezer (∼-35 °C) overnight. Colorless crystals
were isolated by filtration the next day and were washed with
pentane. Subsequent concentration and cooling of the mother
liquor gave a total of 17.4 g (83%) of the product in three
crops: 1H NMR (C6D6, 70 °C) δ 7.095 (s, 4, m-H), 7.069 (s, 4,
m-H), 4.042 (br, 4, o-CHMe2), 3.681 (s, 4, NCH2), 3.539 (br m,
4, o-CHMe2), 3.347 (br m, 16, THF), 2.878 (sept, 2, p-CHMe2),
2.794 (sept, 2, p-CHMe2), 1.471 (d, 12, p-CHMe2), 1.366 (br,
24, o-CHMe2), 1.293 (d, 12, p-CHMe2), 1.205 (d, 12, o-CHMe2),
1.171 (br m, 28, THF and o-CHMe2); 13C NMR δ 153.70, 151.56
(Co), 147.21, 126.2 (Cp), 121.46, 120.16 (Cm), 67.95 (THF), 59.91
(NCH2), 34.65, 34.57 (p-CHMe2), 33.15, 32.56 (o-CHMe2), 27.11
(CHMe2), 25.56 (THF), 25.48, 24.51, 24.46, 24.25 (CHMe2).
Anal. Calcd for C78H128N2B2Li2O4: C, 78.51; H, 10.80; N, 2.35.
Found: C, 79.08; H, 11.31; N, 2.46.
(Tr ip Ben )Zr Cl2 (1). ZrCl4(THF)2 (0.990 g, 2.62 mmol) was
added slowly to a solution of Li2(TripBen)(THF)4 (3.13 g, 2.62
mmol) in toluene (35 mL) at -40 °C. The solution was allowed
to warm to room temperature and was stirred overnight. The
volatile components were removed in vacuo, and the residue
was triturated with pentane (2 × 5 mL). The residue was
subsequently extracted with pentane (50 mL), and the extract
was cooled to -40 °C to produce a crop of clear, colorless
crystals which were isolated by filtration and gently washed
with cold pentane. Further concentration and cooling of the
mother liquors provided two additional crops; total 2.27 g
(87%). The product contains ∼1 equiv of pentane per zirco-
nium: 1H NMR (C6D6, 75 °C) δ 7.152 (s, 4, m-H), 7.132 (s, 4,
m-H), 4.276 (s, 4, NCH2), 3.334 (br m, 8, o-CHMe2), 2.836 (sept,
2, p-CHMe2), 2.703 (sept, 2, p-CHMe2), 1.599 (d, 12, o-CHMe2),
1.341 (d, 12, o-CHMe2), 1.245 (d, 12, p-CHMe2), 1.185 (d, 12,
o-CHMe2), 1.120 (d, 12, p-CHMe2), 0.910 (br d, 12, o-CHMe2);
13C NMR δ 157.37 (Co), 153.52 (Cp), 152.61 (Co), 149.72 (Cp),
136.82, 134.05 (Ci), 124.26, 121.85 (Cm), 56.95 (NCH2), 35.29
(o-CHMe2), 34.75, 34.54 (p-CHMe2), 33.19 (o-CHMe2), 27.37,
26.13 (br), 25.77, 24.16, 24.02, 23.67 (CHMe2). Anal. Calcd
for C67H108N2B2Cl2Zr: C, 71.52; H, 9.66; N, 2.49. Found: C,
71.29; H, 9.90; N, 2.28.
(Tr ip Ben )Zr Me2 (2). Methylmagnesium bromide (0.312
mL, 1.02 mmol, 3.26 M in ether) was added to a solution of 1
(0.546 g, 0.485 mmol) in ether (5 mL) at -40 °C; a precipitate
formed immediately. The mixture was allowed to stand at -40
°C for 30 min, dioxane was added, and the mixture was filtered
through Celite. The volatile components were removed from
the filtrate in vacuo, and the residue was extracted with
pentane (10 mL). The extract was filtered through Celite,
concentrated to ∼1 mL in vacuo, and allowed to stand at -40
°C overnight to yield 0.439 g (83%) of colorless crystals
containing 1 equiv of pentane per Zr: 1H NMR (C6D6, 50 °C)
δ 7.166 (s, 4, m-H), 7.096 (s, 4, m-H), 4.106 (s, 4, NCH2), 3.422
(m, 8, o-CHMe2), 2.848 (sept, 2, p-CHMe2), 2.711 (sept, 2,
p-CHMe2), 1.482 (d, 12, o-CHMe2), 1.375 (d, 12, o-CHMe2),
1.259 (d, 12, p-CHMe2), 1.217 (br d, 12, o-CHMe2), 1.129 (d,
12, p-CHMe2), 0.929 (br, 12, o-CHMe2), 0.012 (br, 6, Zr-CH3);
13C NMR (70 °C) δ 156.35 (Co), 152.42 (Co), 152.02 (Cp), 149.23
(Cp), 123.17, 121.62 (Cm), 54.72 (1J CH ) 137.9 Hz, NCH2), 52.50
(1J CH ) 114.7 Hz, Zr-CH3), 34.97 (o-CHMe2), 34.72, 34.62 (p-
CHMe2), 32.97 (o-CHMe2), 27.54, 26.05, 25.62, 24.14, 24.10,
23.96 (CHMe2). Anal. Calcd for C69H114N2B2Zr: C, 76.42; H,
10.59; N, 2.58. Found: C, 76.49; H, 10.96; N, 2.50.
The key idea at the beginning of this work was to
employ B-N π bonding in a dialkyl or diarylboron
substituent on amido nitrogens to both orient the bulky
boron substituents and to increase the electrophilicity
of the metal in a pseudo-three-coordinate cationic rela-
tive of a pseudo-three-coordinate metallocene alkyl
cation. In light of the success obtained with ordinary
diamido ligands noted above, that tactic would now
appear to be unnecessary and in fact apparently yields
species which decompose too readily compared to reac-
tion with olefins to yield polymers. Metal-nitrogen π
bonding may be helpful in stabilizing intermediate
cationic species and slowing the rate of â-hydride
elimination in the growing polymer chain compared to
further reaction of the cation with olefin. Although
boryl-substituted amido ligands do not appear to be
advantageous for group 4 metal chemistry related to
olefin polymerization catalysts, it is conceivable that
boryl-substituted chelating diamido ligands of the type
described here might be useful in circumstances where
a more electron-poor metal is desired than in an
analogous chelating diamido complex in which the
amido nitrogen contains a carbon-based substituent.
Exp er im en ta l Section
Gen er a l P r oced u r es. All experiments were performed
under nitrogen in a Vacuum Atmospheres drybox or under
argon using standard Schlenk techniques. All solvents were
purified by standard techniques, and deuterated NMR solvents
were dried and stored over 4 Å molecular sieves before use.
Butyllithium, H2BCl(SMe2), and all Grignard reagents were
obtained from commercial sources and titrated immediately
before use. Ethylenediamine containing 0.6% water was
stored over 4 Å molecular sieves before use. [HNMe2Ph]-
[B(C6F5)4],49 [Ph3C][B(C6F5)4], and B(C6F5)350 were donated by
Exxon. Trip2BF was prepared in an identical fashion to that
52
described for Xylyl2BF.51 ZrCl4(THF)2 was prepared as
described in the literature, as were M(CH2SiMe3)2Cl2(ether)2
(M ) Zr or Hf)32 derivatives.
1H, 13C, and 19F spectra were recorded at 300, 75.4, and 282
MHz, respectively. Proton spectra were referenced internally
by the residual solvent proton signal relative to tetramethyl-
silane. Carbon spectra were referenced internally relative to
the 13C signal of the NMR solvent relative to tetramethylsilane.
Fluorine spectra were referenced externally to neat CFCl3.
Elemental analyses were performed on a Perkin-Elmer PE2400
microanalyzer in our laboratories.
Li2(Tr ip Ben )(THF )4. Butyllithium (13.9 mL, 35.3 mmol,
2.53 M in hexane) was added to a solution of ethylenediamine
(1.06 g, 17.6 mmol) in THF (150 mL) that had been chilled to
-40 °C. A flocculent white precipitate soon formed. The
reaction was allowed to warm to room temperature and was
stirred for 3 h. The reaction was chilled to -40 °C again and
Trip2BF (15.4 g, 35.28 mmol) was added over a period of 5
min. The flocculent solid was soon consumed to give a clear
viscous suspension of LiF. The solution was stirred overnight,
concentrated to ∼25 mL in vacuo, and combined with 125 mL
of pentane. The mixture was filtered through Celite, and the
Celite was washed with pentane (3 × 30 mL) and THF (10
mL). The combined filtrates were then chilled to -40 °C.
Butyllithium (13.9 mL, 35.3 mmol, 2.53 M in hexane) was
added to the chilled filtrate with stirring, and the mixture was
(Tr ip Ben )Zr (CD3)2 was prepared analogously, employing
2.1 equiv of LiCD3‚LiI in ether with the exception that a total
of 4 pentane extraction cycles were performed in order to
completely remove all lithium halides.
(Tr ip Ben )Zr (CH2CH3)2 (3). EtMgCl (0.328 mL, 0.692
mmol, 2.11 M in ether) was added to a solution of 1 (0.371 g,
0.330 mmol) in ether (5 mL) at -40 °C. The reaction mixture
was allowed to stand at -40 °C for 30 min, and dioxane (0.060
mL, 0.70 mmol) was added. The mixture was filtered through
Celite, and the filtrate was concentrated to dryness. The
residue was extracted with pentane (5 mL). The pentane
(49) Turner, H. W.; Hlatky, G. G. (Exxon) Int. Appl. WO 88/05793
(Eur. Pat. Appl. EP 211004), 1988.
(50) Massey, A. G.; Park, A. J . J . Organomet. Chem. 1964, 2, 245.
(51) Chen, H.; Power, P. P.; Shoner, S. C. Inorg. Chem. 1991, 30,
2884.
(52) Manzer, L. E. Inorg. Synth. 1982, 21, 135.

318 Organometallics, Vol. 17, No. 3, 1998
Warren et al.
extracts were concentrated to dryness, and the residue was
again extracted with pentane. After the mixture was filtered
once again, the filtrate was concentrated to dryness to afford
3 as a fluffy powder (0.269 g, 78% yield): 1H NMR (C6D6, 70
°C) δ 7.147 (s, 4, m-H), 7.081 (s, 4, m-H), 4.127 (s, 4, NCH2),
3.417 (br m, 8, o-CHMe2), 2.837 (sept, 2, p-CHMe2), 2.754 (sept,
2, p-CHMe2), 1.474 (d, 12, o-CHMe2), 1.371 (d, 12, o-CHMe2),
1.249 (d, 12, p-CHMe2), 1.175 (d, 24, p-CHMe2 and o-CHMe2),
1.061 (br t, 6, Zr-CH2CH3), 0.961 (br, 12, o-CHMe2), 0.585 (br,
4, Zr-CH2CH3); 13C NMR δ 154.71 (Co), 152.17 (Cp), 150.77
(Co), 149.01 (Cp), 138.35 (br, Ci), 122.43, 121.60 (Cm), 64.52 (br,
Zr-CH2CH3), 53.18 (NCH2), 34.69 (p-CHMe2), 34.53, 33.00 (o-
CHMe2), 27.49, 26.22 (br), 25.52, 24.17, 24.11, 23.99 (CHMe2),
11.68 (br, Zr-CH2CH3). Anal. Calcd for C66H106N2B2Zr: C,
76.20; H, 10.26; N, 2.69. Found: C, 76.68; H, 10.45; N, 2.63
solution of 6 along with free Me2NPh was generated by adding
[HNMe2Ph][B(C6F5)4] (0.074 g, 0.092 mmol) with stirring to
the toluene-d8 solution of freshly prepared 3 at -40 °C: 1H
NMR (C6D6, 0 °C) δ 7.225 (s, 2, m-H), 7.200 (s, 2, m-H), 7.155
(s, 2, m-H), 7.050 (s, 2, m-H), 4.096 (br AA′BB′, 2, NCH2), 3.745
(br AA′BB′, 2, NCH2), 2.944 (m, 4, o-CHMe2), 2.816 (sept, 2,
CHMe2), 2.700 (br m, 6, CHMe2), 1.416 (d, 6, CHMe2), 1.327
(d, 6, CHMe2), 1.301 (d, 6, CHMe2), 1.252 (d, 12, CHMe2), 1.205
(d, 6, CHMe2), 1.121 (d, 6, CHMe2), 1.085 (d, 6, CHMe2), 1.048
(d, 6, CHMe2), 0.954 (d, 6, CHMe2), 0.878 (d, 6, CHMe2), 0.368
3
(d, 6, CHMe2), -0.655 (t, J HH ) 8.0 Hz, 3, Zr-CH2CH3). The
Zr-CH2CH3 resonance is obscured by the o-CHMe2 resonances.
[(Tr ipBen )Zr (CH2CH2CH2CH3)][B(C6F5)4] (7). A toluene-
d8 solution (0.5 mL) of 4 was prepared from 1 (0.100 g, 0.092
mmol) and BuMgCl (0.086 mL, 0.193 mmol, 2.25 M in ether).
A solution of 7 plus free Me2NPh was generated by adding
[HNMe2Ph][B(C6F5)4] (0.074 g, 0.092 mmol) with stirring to
the toluene-d8 solution of freshly prepared 4 at -40 °C: 1H
NMR (C6D6) δ 7.272 (s, 2, m-H), 7.197 (s, 4, m-H), 7.056 (s, 2,
m-H), 4.096 (br AA′BB′, 2, NCH2), 3.659 (br AA′BB′, 2, NCH2),
2.98 (m, 8, o-CHMe2), 2.819 (sept, 2, CHMe2), 2.719 (br m, 2,
CHMe2), 1.5-0.75 (m, CHMe2 and Zr-CH2CH2CH2CH3), 0.593
(t, 3J HH ) 7.0 Hz, Zr-CH2CH2CH2CH3), 0.41 (br d, 6, CHMe2),
-0.11 (br, 2, Zr-CH2CH2).
(Tr ip Ben )Zr (CD2CD3)2 was prepared analogously from 2
and 2.5 equiv of CD3CD2MgI with the exception that a total
of four pentane extraction cycles were performed in order to
completely remove all magnesium halides: 2H NMR (70 °C) δ
0.938 (Zr-CD2CD3), 0.482 (br, Zr-CD2CD3).
(Tr ip Ben )Zr Bu 2 (4). BuMgCl (0.105 mL, 0.235 mmol, 2.25
M in ether) was added to a solution of 1 (0.126 g, 0.112 mmol)
in ether (5 mL) at -40 °C. The product was obtained as a
fluffy powder (0.082g, 65% yield) after an isolation procedure
analogous to that described for 3: 1H NMR (C6D6, 50 °C) δ
7.150 (s, 4, m-H), 7.085 (s, 4, m-H), 4.117 (s, 4, NCH2), 3.406
(br m, 8, o-CHMe2), 2.832 (sept, 2, p-CHMe2), 2.778 (sept, 2,
p-CHMe2), 1.495 (d, 12, o-CHMe2), 1.385 (d, 12, o-CHMe2),
1.245 (d, 12, p-CHMe2), 1.215 (d, 12, p-CHMe2), 1.166 (br d,
12, o-CHMe2), 0.973 (br, 12, o-CHMe2), 0.817 (br m, 10, Zr-
CH2CH2CH3 and Zr-CH2CH2CH3), 0.654 (br, 4, Zr-CH2).
Cy2BNHCH2CH2NHBCy2 (H2(CyBen )). An ethereal so-
lution of Cy2BCl30 was prepared by addition of H2BCl(SMe2)
(25.00 g, 226 mmol) in ether (45 mL) to a stirring solution of
cyclohexene (40.9 g, 498 mmol) in ether (150 mL) maintained
at 0 °C. The solution became cloudy and was stirred at 0 °C
for 1 h. The reaction mixture was then allowed to warm to
room temperature and was stirred overnight. A suspension
of [LiNHCH2CH2NHLi]x was prepared by adding butyllithium
(95.1 mL, 238 mmol, 2.5 M in hexane) to a stirred solution of
ethylenediamine (7.15 g, 118.9 mmol) in THF (375 mL) at -78
°C. After the addition was complete, the reaction mixture was
stirred for 30 min at -78 °C, allowed to warm to room
temperature, and stirred an additional 90 min. After cooling
the suspension of [LiNHCH2CH2NHLi]x once again to -78 °C,
the clear solution of Cy2BCl was added, consuming the white
suspension to give a light yellow solution. The solution was
then warmed to room temperature and stirred overnight. The
volatile components were removed in vacuo, and the residue
was extracted with pentane (∼400 mL). The extracts were
filtered through Celite and concentrated to ∼100 mL to yield
a large amount of crystalline solid. The filtrate was then
cooled to -30 °C for 3 h, and the crystalline solid was collected
on a large frit, washed with pentane (∼20 mL), and dried in
vacuo to give 31.75 g (68%) of colorless crystals: 1H NMR δ
(CDCl3) 3.872 (br t, 2, NH), 2.993 (pseudo t, 4, NCH2), 1.7-
0.9 (br resonances, 44, Cy); 13C NMR δ 44.88 (NCH2), 29.66
(Câ), 28.95 (B-C), 28.72 (Câ), 28.36 (B-C), 28.06, 28.00 (Cγ),
27.25, 27.12 (Cδ); 11B NMR δ 46.3; IR (Nujol/KBr) 3396 ν(NH)
cm-1. Anal. Calcd for C26H50N2B2: C, 75.75; H, 12.21; N, 6.80.
Found: C, 76.01; H, 12.33; N, 6.83.
Li2(CyBen )(eth er ). A solution of tert-butyllithium (22.0
mL, 36.08 mmol, 1.64 M in hexane) was slowly added over 15
min to a chilled (-30 °C), stirred solution of H2(CyBen) (7.26
g, 17.61 mmol) in pentane (200 mL) with a small amount of
ether (10 mL) added. After the addition was complete, the
suspension was allowed to stir for 2 h at room temperature.
The mixture was concentrated to one-half of its original volume
in vacuo and the flask was cooled to -30 °C for 2 h. The white
solid was collected by filtration, washed liberally with pentane,
and dried in vacuo, to afford 8.30 g (95%) of the product: 1H
NMR (C6D6 plux 6 drops THF-d8) δ 3.589 (br, 4, NCH2), 3.252
(q, 4, OCH2CH3), 2.0-1.2 (br resonances, 44, Cy), 1.089 (t, 6,
OCH2CH3); 13C NMR δ 65.86 (OCH2CH3), 49.81 (NCH2), 33.92
(br, B-C), 32.40 (Cy), 31.60 (br, B-C), 30.71 (Cy), 49.48 (2
Cy), 28.42 (Cy), 28.30 (Cy), 15.49 (OCH2CH3). Elemental
analyses (C and H) were consistently low, consistent with
gradual loss of ether in the solid state.
[(Tr ip Ben )Zr Me][MeB(C6F 5)3] (5a ). A chilled (-40 °C)
solution of B(C6F5)3 (0.047 g, 0.092 mmol) in pentane (5 mL)
was added dropwise to a solution of 2 (0.100 g, 0.092 mmol)
in pentane (5 mL) at -40 °C. During the addition of the
triarylborane, the solution became turbid, and after complete
addition, a colorless oil separated on the walls of the reaction
vial. After the mixture was allowed to stand for 36 h at -40
°C, the oil crystallized to afford 0.128 g (87%) of colorless
crystals of 5a : 1H NMR (C7D8) δ 7.184 (s, 6, m-H), 7.074 (s, 2,
m-H), 3.961 (br AA′BB′, 2, NCH2), 3.489 (br AA′BB′, 2, NCH2),
2.875 (m, 4, o-CHMe2), 2.826 (sept, 2, CHMe2), 2.726 (br m, 2,
CHMe2), 2.590 (sept, 2, CHMe2), 2.436 (br m, 2, CHMe2), 1.413
(d, 6, CHMe2), 1.402 (br, 3, B-Me), 1.390 (br d, 6, CHMe2),
1.254 (d, 18, CHMe2), 1.201 (d, 6, CHMe2), 1.142 (d, 6, CHMe2),
1.109 (d, 6, CHMe2), 0.971 (d, 6, CHMe2), 0.894 (d, 6, CHMe2),
0.755 (d, 6, CHMe2), 0.387 (d, 6, CHMe2), 0.201 (s, 3, Zr-CH3);
3
3
19F NMR δ -131.77 (d, J FF ) 21.2, o-C6F5), -164.25 (t, J FF
3
) 21.0, p-C6F5), -166.67 (m, J FF (average) ) 22.0, m-C6F5).
Anal. Calcd for C82H102N2B3F15Zr: C, 64.61; H, 6.74; N, 1.84.
Found: C, 64.50; H, 7.06; N, 1.80
[(Tr ip Ben )Zr Me][B(C6F 5)4] (5b). A solution of 5b con-
taining free Me2NPh was generated by stirring [HNMe2Ph]-
[B(C6F5)4] (0.076 g, 0.095 mmol) with 2 (0.100 g, 0.095 mmol)
in toluene-d8 (0.5 mL). Gas evolved vigorously. 5b could be
obtained as a powder after exhaustive removal of the toluene
in vacuo and washing the precipitate with minimal pentane
(3 × 2 mL) to remove the aniline: 1H NMR (C7D8) δ 7.188 (s,
2, m-H), 7.174 (s, 4, m-H), 7.079 (s, 2, m-H), 3.976 (br AA′BB′,
2, NCH2), 3.503 (br AA′BB′, 2, NCH2), 2.891 (m, 4, o-CHMe2),
2.818 (sept, 2, CHMe2), 2.732 (br m, 2, CHMe2), 2.574 (sept,
2, CHMe2), 2.452 (br m, 2, CHMe2), 1.420 (d, 6, CHMe2), 1.381
(br d, 6, CHMe2), 1.253 (d, 18, CHMe2), 1.200 (d, 6, CHMe2),
1.129 (d, 6, CHMe2), 1.116 (d, 6, CHMe2), 0.971 (d, 6, CHMe2),
0.904 (d, 6, CHMe2), 0.756 (d, 6, CHMe2), 0.399 (d, 6, CHMe2),
0.213 (s, 3, Zr-CH3); 19F NMR δ -131.80 (br, o-C6F5), -163.11
(m, p-C6F5), -166.85 (br, m-C6F5).
[(Tr ip Ben )Zr (CH2CH3)][B(C6F 5)4] (6). A toluene-d8 solu-
tion (0.5 mL) of 3 was prepared from 1 (0.100 g, 0.092 mmol)
and EtMgCl (0.072 mL, 0.193 mmol, 2.67 M in ether).
A

Group 4 Complexes Containing [R2BNCH2CH2NBR2]2-
Organometallics, Vol. 17, No. 3, 1998 319
(CyBen )Zr (CH2SiMe3)2 (8a ). Zr(CH2SiMe3)2Cl2(ether)2
was first prepared in situ as follows: under reduced lighting,
a solution of LiCH2SiMe3 (3.00 g, 31.9 mmol) in ether (95 mL)
was added dropwise over a period of 1 h to a vigorously stirred
suspension of ZrCl4 (3.71 g, 15.9 mmol) in ether (75 mL) at
room temperature. After the reaction was stirred an ad-
ditional 30 min, the reaction mixture was filtered through
Celite and the filter cake was washed with ether (2 × 20 mL)
to give a clear, light yellow solution of Zr(CH2SiMe3)2Cl2-
(ether)2. This solution was chilled to -30 °C, and solid Li2-
(CyBen)(ether) (7.94 g, 15.9 mmol) was added with stirring.
The mixture was stirred overnight. The volatile components
were removed in vacuo. The residue was first titurated with
pentane (25 mL) and then extracted with pentane (350 mL).
The extract was filtered through Celite, and the filtrate was
concentrated in vacuo to ∼25 mL. The mixture was allowed
to stand for several hours at -30 °C in order to complete
crystallization. The product was collected by filtration, washed
with pentane (3 × 5 mL), and dried in vacuo to yield 8.41 g
(76%) of colorless crystals: 1H NMR (C6D6) δ 3.766 (s, 4,
NCH2), 2.05-1.20 (br m, 44, Cy), 0.590 (s, 4, Zr-CH2), 0.185
(s, 18, CH2SiMe3); 13C NMR δ 55.50, 53.31, 32.78 (B-CH),
30.07 (B-CH), 29.84, 28.62, 28.53, 28.00, 27.60, 26.76 (Cy).
Anal. Calcd for C34H70N2B2Si2Zr: C, 60.42; H, 10.43; N, 4.15.
Found: C, 60.63; H, 10.68; N, 3.96.
(br m, 44, Cy), 1.751 (s, 4, Zr-CH2), 0.138 (s, 18, CH2SiMe3);
13C NMR δ 77.69, 51.02 (NCH2), 28.94, 28.66, 27.61 (Cy), 3.42
(SiMe3). Anal. Calcd for C34H70N2B2Si2Ti: C, 64.56; H, 11.14;
N, 4.43. Found: C, 63.83; H, 11.39; N, 4.57.
(CyBen )Zr I2 (9a ). Iodine crystals (3.35 g, 13.20 mmol)
were added slowly to a rapidly stirred solution of (CyBen)Zr-
(CH2SiMe3)2 (4.46 g, 6.60 mmol) in dichloromethane (200 mL)
at room temperature until the color of the iodine was no longer
discharged. (Approximately 0.06 g of iodine remained.) The
mixture was stirred for a further 45 min, and the light orange
solution was concentrated to ∼20 mL in vacuo. The concen-
trated mixture was allowed to stand at -30 °C overnight. The
crystals that formed were collected on a frit and washed with
pentane (3 × 10 mL). Two further crops were obtained from
the mother liquors combined with the pentane washings to
give a total of 4.42 g (88%) of product: 1H NMR (C6D6) δ 3.579
(s, 4, NCH2), 2.35-2.15 (br, 4, Cy), 1.90-1.20 (br m, 40, Cy);
13C NMR δ 56.89 (NCH2), 32.44 (B-CH), 30.28 (Cy), 30.00
(BCH), 28.16, 28.02, 27.35, 27.18, 26.29 (Cy). Anal. Calcd for
C
26H48N2B2I2Zr: C, 41.35; H, 6.40; N, 3.71. Found: C, 41.40;
H, 6.81; N, 3.61.
(CyBen )HfI2 (9b). Iodine crystals (4.92 g, 19.4 mmol) were
added slowly to a rapidly stirred solution of (CyBen)Hf(CH2-
SiMe3)2 (7.40 g, 9.70 mmol) in dichloromethane (250 mL) at
room temperature until the color of the iodine was no longer
discharged upon dissolution. (Approximately 0.20 g of iodine
remained.) The product was isolated as described for 9a ; total
yield 6.76 g (83%): 1H NMR (C6D6) δ 3.579 (s, 4, NCH2), 2.35-
2.15 (br, 4, Cy), 1.90-1.20 (br m, 40, Cy); 13C NMR δ 56.89
(NCH2), 32.44 (B-CH), 30.28 (Cy), 30.00 (BCH), 28.16, 28.02,
27.35, 27.18, 26.29 (Cy). Anal. Calcd for C26H48N2B2I2Hf: C,
37.06; H, 5.74; N, 3.33. Found: C, 36.78; H, 5.65; N, 3.28.
(CyBen )TiI2 (9c). Iodine crystals (3.03 g, 11.95 mmol) were
added slowly to a rapidly stirred solution of (CyBen)Ti(CH2-
SiMe3)2 (4.46 g, 6.60 mmol) in dichloromethane (300 mL) at
room temperature. At the end of the addition, the color of the
solution indicated the presence of excess iodine. The dark
solution was filtered through Celite and concentrated to ∼25
mL in vacuo. Pentane (∼25 mL) was added to assist precipi-
tation of the product. The crude product (contaminated with
some I2) was collected on a frit and washed with pentane (2 ×
5 mL). It was recrystallized from a mixture of pentane and
dichloromethane (50/50, ∼150 mL) to give 2.40 g (56%) of
golden orange needles: 1H NMR (C6D6) δ 3.579 (s, 4, NCH2),
2.35-2.15 (br, 4, Cy), 1.90-1.20 (br m, 40, Cy); 13C NMR δ
56.89 (NCH2), 32.44 (BCH), 30.28 (Cy), 30.00 (BCH), 28.16,
28.02, 27.35, 27.18, 26.29 (Cy). Anal. Calcd for C26H48N2B2I2-
Ti: C, 43.86; H, 6.79; N, 3.94. Found: C, 43.72; H, 7.11; N,
3.93.
(CyBen )Zr (CH2CH3)2 (10a ). A solution of CH3CH2MgCl
(1.37 mL, 2.78 mmol, 2.03 M in ether) was added to a stirred,
chilled (-30 °C) solution of (CyBen)ZrI2 (1.00 g, 1.32 mmol)
in dichloromethane (35 mL). The reaction mixture was
allowed to stand at -30 °C for 30 min, and the magnesium
salts were precipitated by the adding 1,4-dioxane (0.28 g, 3.2
mmol). The volatile components were removed in vacuo, and
the residue was triturated with pentane (10 mL). The powdery
residue was extracted with pentane, the extracts were filtered
through Celite, and the filtrates were concentrated to dryness
in vacuo. The resulting white solid was recrystallized from
pentane at -30 °C to afford 0.597 g (81%) of white crystals in
two crops: 1H NMR (C6D6) δ 3.804 (s, 4, NCH2), 1.90-1.00
(br m, 44, Cy), 1.538 (t, 6, CH2CH3), 0.965 (q, 4, ZrCH2); 13C
NMR δ 53.88 (NCH2), 49.03 (ZrCH2), 32.38, 30.06 (BCH),
29.77, 28.72, 28.59, 28.14, 28.59, 27.64, 26.87 (Cy), 9.95 (Zr-
CH2CH3). Anal. Calcd for C30H58N2B2Zr: C, 64.39; H, 10.44;
N, 5.01. Found: C, 64.35; H, 10.44; N, 5.07.
(CyBen )Hf(CH2SiMe3)2 (8b). Hf(CH2SiMe3)2Cl2(OEt2)2
was first prepared in situ as follows: under reduced lighting,
a solution of LiCH2SiMe3 (2.70 g, 28.7 mmol) in ether (90 mL)
was added dropwise over 1 h to a vigorously stirring suspen-
sion of HfCl4 (4.59 g, 14.3 mmol) in ether (75 mL) at room
temperature. The reaction mixture was stirred for an ad-
ditional 30 min and filtered through Celite. The filter cake
was washed with ether (2 × 20 mL) to give a clear, light yellow
filtrate. This solution of Hf(CH2SiMe3)2Cl2(ether)2 was chilled
to -30 °C, and solid Li2(CyBen)‚OEt2 (7.00 g, 14.1 mmol) was
added with stirring. The product was isolated as described
for 8a ; yield 7.81 g (74%): 1H NMR (C6D6) δ 3.786 (s, 4, NCH2),
2.05-1.20 (br m, 44, Cy), 0.186 (s, 18, CH2SiMe3), 0.157 (s, 4,
Hf-CH2); 13C NMR δ 64.42, 49.83, 32.91 (B-CH), 30.20 (B-
CH), 29.54, 28.57, 28.08, 27.61, 26.84 (Cy), 3.95 (SiMe3). Anal.
Calcd for
C34H70N2B2Si2Hf: C, 53.51; H, 9.24; N, 3.67.
Found: C, 53.36; H, 9.26; N, 3.53.
(CyBen )Ti(CH2SiMe3)2 (8c). Ti(CH2SiMe3)2Cl253 was first
prepared by the adding a solution of TiCl4 (1.14 g, 6.02 mmol)
in dichloromethane (20 mL) to a solution of Ti(CH2SiMe3)4
(2.39 g, 6.02 mmol) in dichloromethane (50 mL) at room
temperature. An aliquot taken from the bright orange solution
indicated that the reaction was 80% complete, as judged by
integration of the TiCH2SiMe3 resonances in the mixture of
the Ti(CH2SiMe3)4-xClx (x ) 0-3) products. After an additional
1
20% of TiCl4 (0.234 g) was added to the reaction mixture, H
NMR analysis of a subsequent aliquot indicated the reaction
was complete. The volatile components were removed in
vacuo, and the deep orange oil was added to chilled (-30 °C)
ether (100 mL). Solid Li2(CyBen)(ether) (5.84 g, 11.7 mmol)
was then added in two batches to the stirred ethereal solution
of Ti(CH2SiMe3)2Cl2. After the addition was complete, the
mixture was stirred for 1 h, a small amount of dichloromethane
was added, and the volatile components were removed in
vacuo. The residue was extracted with ∼350 mL of dichlo-
romethane, and the extract was filtered through Celite. The
brown filtrate was concentrated to ∼50 mL and was allowed
to stand at -30 °C for 1 h to complete crystallization. Thin
needles were collected by filtration, which were washed
liberally with pentane and dried in vacuo to afford 4.01 g (53%)
of dark yellow needles which were used without further
purification: 1H NMR (C6D6) δ 3.746 (s, 4, NCH2), 2.05-1.20
(CyBen )Hf(CH2CH3)2 (10b). This compound was prepared
as described for 10a from (CyBen)HfI2 (0.750 g, 0.890 mmol)
and CH3CH2MgCl (0.886 mL, 1.87 mmol, 2.11 M in ether);
yield 0.448 g (78%) of colorless crystals in two crops: 1H NMR
(53) Beilin, S. I.; Bondarenko, G. N.; Vdovin, V. M.; Dolgoplosk, B.
A.; Markevich, I. N.; Nametkin, N. S.; Poletaev, V. A.; Svergun, V. I.;
Sergeeva, M. B. Dokl. Akad. Nauk SSSR 1974, 218, 1347.

320 Organometallics, Vol. 17, No. 3, 1998
Warren et al.
(C6D6) δ 3.778 (s, 4, NCH2), 1.95-1.20 (br m, 44, Cy), 1.540 (t,
6, CH2CH3), 0.853 (q, 4, Zr-CH2); 13C NMR δ 61.71 (HfCH2),
50.66 (NCH2), 32.49, 30.35 (BCH), 29.52, 28.67, 28.58, 28.18,
27.62, 26.91, 10.07 (Hf-CH2CH3). Anal. Calcd for C30H58N2B2-
Hf: C, 55.70; H, 9.03; N, 4.33. Found: C, 55.74; H, 8.90; N,
4.33.
26.91, 23.09, 14.33. Anal. Calcd for C38H74N2B2Zr: C, 67.94;
H, 11.09; N, 4.17. Found: C, 67.90; H, 11.40; N, 3.98.
(CyBen )Hf(CH2CH2Bu )2 (12b). This product was pre-
pared in a manner analogous to that described for 12a from
(CyBen)HfI2 (0.425 g, 0.504 mmol) and BuCH2CH2MgBr (0.512
mL, 1.06 mmol, 2.07 M in ether); yield 0.334 g (87%) of the
product as a slightly tan, analytically pure oil: 1H NMR (C6D6)
δ 3.811 (s, 4, NCH2), 2.05-1.20 (br m, 60, Cy and hexyl), 1.063
(m, 4, ZrCH2), 0.912 (t, 6, CH3), (m, 4, Hf-CH2); 13C NMR δ
73.57 (HfCH2), 50.52 (NCH2), 36.49, 32.91 (BCH), 32.18, 30.36
(BCH), 29.59, 28.69, 28.58, 28.19, 27.61, 27.54, 26.93, 23.11,
14.34. Anal. Calcd for C38H74N2B2Hf: C, 60.13; H, 9.82; N,
3.69. Found: C, 60.06; H, 9.93; N, 3.42.
(CyBen )Zr (CHBu 2)2 (13a ). Bu2CHMgBr was prepared in
three steps from 5-nonanone: (1) 5-nonanone (10.73 g, 78.79
mmol) was reduced with lithium aluminum hydride (1.50 g,
39.4 mmol) in ether54 to give 10.81 g (99%) of 5-nonanol after
workup with 2 N H2SO4, extraction with ether, and removal
of all volatiles by rotary evaporation under reduced pressure
(water aspirator). (2) The nonanol was degassed and stood
over 4 Å molecular sieves for 12 h. 5-Nonanol (5.25 g, 38.00
mmol) was placed in dry, degassed DMF (50 mL), and the
solution was chilled to 0 °C under argon. Solid Ph3PBr2 (17.58
g, 41.65 mmol)55 was added to the solution over a period of 45
min, after which time the solution was gently heated to 45 °C
for 2 h. Hexane (300 mL) and water (200 mL) were then
added, and the organic fraction was collected and washed with
water (2 × 200 mL). The hexane layer was dried with MgSO4,
filtered, concentrated to ∼5 mL, filtered through activated
alumina, and then concentrated in vacuo to a constant weight
to yield 4.70 g (61%) of 5-bromononane as an oil. (3) Dry
5-bromononane (2.05 g, 10.17) was added slowly over a period
of 2 h to 0.60 g of magnesium powder (activated by heating
followed by the addition of an I2 crystal after the addition of
ether) in ether (10 mL). The Grignard solution was filtered,
concentrated to <10 mL, transferred to a 10 mL volumetric
flask, and diluted to 10.0 mL. Titration with n-propanol
indicated the yield of Grignard reagent was 45%.
A solution of Bu2CHMgBr (3.24 mL, 1.46 mmol, 0.45 M in
ether) was added to a stirred, chilled (-30 °C) solution of
(CyBen)ZrI2 (0.500 g, 0.662 mmol) in dichloromethane (15 mL).
The reaction mixture was allowed to stand at -30 °C for 5 h
and the magnesium salts were precipitated by the addition of
1,4-dioxane (0.35 g, 4.0 mmol). The volatiles were removed
in vacuo, and the residue was extracted with pentane, filtered
through Celite, and concentrated to dryness in vacuo. 1H NMR
analysis of this residue indicated that alkylation was not
complete, showing the presence of ∼10% (CyBen)Zr(CHBu2)I.
This mixture was dissolved in ether (10 mL) and chilled to
-30 °C, and an additional 0.35 mL of the Bu2CHMgBr solution
was added. After 5 h, more dioxane (35 µL) was added and
the mixture was concentrated to dryness. The residue was
extracted with pentane (10 mL), the extracts were filtered
through Celite, and the filtrate was concentrated to dryness
in vacuo. The residue was recrystallized from pentane at -30
°C to yield 0.354 g (71%) of thin, colorless needles in four
crops: 1H NMR (C6D6) δ 3.939 (s, 4, NCH2), 2.2-1.2 (br m,
68, Cy and Bu), 0.999 (t, 12, CH2Me), 0.732 (br q, 2, ZrCHBu2);
13C NMR δ 76.23 (Zr-CHBu2), 53.50 (NCH2), 34.61, 33.67
(BCH), 33.08, 30.94, 30.17 (BCH), 29.56, 28.68, 27.96, 27.69,
27.02, 23.85 (CH2CH3), 14.43 (CH3). Anal. Calcd for
(CyBen )Zr (CH2CHMe2)2 (11a ). A solution of Me2CHCH2-
MgCl (1.11 mL, 2.66 mmol, 2.39 M in ether) was added to a
stirred, chilled (-30 °C) solution of (CyBen)ZrI2 (0.980 g, 1.30
mmol) in dichloromethane (35 mL). The reaction mixture was
allowed to stand at -30 °C for 45 min, and the magnesium
salts were precipitated by the addition of 1,4-dioxane (0.25 g,
2.8 mmol). The volatile components were removed in vacuo,
and the residue was triturated with pentane (5 mL). The
powdery residue was then extracted with pentane (25 mL),
and the extracts were filtered through Celite and concentrated
to dryness in vacuo. The resulting white solid was recrystal-
lized from pentane at -30 °C to afford 0.687 g (86%) of white
crystals in three crops: 1H NMR (C6D6) δ 3.777 (s, 4, NCH2),
2.326 (m, 2, CHMe2), 2.05-1.30 (br m, 44, Cy), 1.096 (d, 12,
CHMe2), 1.004 (d, 4, Zr-CH2); 13C NMR δ 76.60 (ZrCH2), 53.56
(NCH2), 33.31, 30.07 (B-CH), 29.76, 29.62, 28.85, 28.65, 28.59,
27.93, 27.63, 26.89 (Cy and CHMe2). Anal. Calcd for
C
34H66N2B2Zr: C, 66.33; H, 10.79; N, 4.55. Found: C, 66.52;
H, 10.85; N, 4.52.
(CyBen )Hf(CH 2CHMe2)2 (11b). This compound was pre-
pared as described for 11a from (CyBen)HfI2 (0.750 g, 0.890
mmol) and Me2CHCH2MgCl (0.861 mL, 1.87 mmol, 2.17 M in
ether); yield 0.535 g (85%) of white crystals in two crops: 1H
NMR (C6D6) δ 3.778 (s, 4, NCH2), 2.387 (m, 2, CHMe2), 2.1-
1.3 (br m, 44, Cy), 1.103 (d, 12, CHMe2), 0.741 (d, 4, HfCH2);
13C NMR δ 88.34 (HfCH2), 50.22 (NCH2), 33.30 (B-CH), 30.34
(BCH), 29.97, 29.64, 29.51, 28.59, 28.00, 27.62, 26.94. Anal.
Calcd for C34H66N2B2Hf: C, 58.09; H, 9.45; N, 3.99. Found:
C, 58.22; H, 9.83; N, 3.96.
(CyBen )Ti(CH2CHMe2)2 (11c). A solution of diisobutyl-
magnesium (3.13 mL, 0.938 mmol, 0.30 M in ether) was added
with stirring to a chilled (-30 °C) solution of (CyBen)TiI2
(0.607 g, 0.853 mmol) in dichloromethane (25 mL). The
reaction mixture was allowed to stand at -30 °C for 10 min,
and the magnesium salts were precipitated by the addition of
1,4-dioxane (0.10 mL, 1.2 mmol). The volatile components
were removed in vacuo, and the residue was triturated with
pentane (5 mL). The yellow, powdery residue was extracted
with pentane (∼25 mL), and the extracts were filtered through
Celite and concentrated to dryness in vaacuo. The resulting
yellow solid was recrystallized at -30 °C from pentane after
filtering to afford 0.424 g (87%) of yellow crystals in three
crops: 1H NMR (C6D6) δ 3.802 (s, 4, NCH2), 2.387 (m, 2,
CHMe2), 2.15-1.20 (br m, 44, Cy), 1.750 (d, 4, TiCH2), 1.007
(d, 12, CHMe2); 13C NMR δ 94.29 (TiCH2), 52.28 (NCH2), 32.39,
28.97 (br), 28.65, 27.76, 27.67 (Cy and CHMe2). (The BC
resonance was not identified.) Anal. Calcd for C34H66N2B2Ti:
C, 71.35; H, 11.61; N, 4.90. Found: C, 71.34; H, 11.60; N, 4.91.
(CyBen )Zr (CH2CH2Bu )2 (12a ). A solution of BuCH2CH2-
MgBr (0.506 mL, 1.05 mmol, 2.07 M in ether) was added to a
stirred, chilled (-30 °C) solution of (CyBen)ZrI2 (0.377 g, 0.499
mmol) in dichloromethane (15 mL). The solution was allowed
to stand at -30 °C for 45 min before the magnesium salts were
precipitated by the addition of 1,4-dioxane (0.10 mL, 1.2 mmol).
The volatiles were removed in vacuo, and the oily residue was
triturated with pentane (5 mL) and then extracted with
pentane (10 mL). The extract was filtered through Celite, and
the filtrate was concentrated to dryness. The residue was
extracted with pentane, and the filtrate was concentrated to
give 0.297 g (98%) of the product as a slightly tan, analytically
pure oil: 1H NMR (C6D6) δ 3.825 (s, 4, NCH2), 1.9-1.2 (br m,
60, Cy and hexyl), 1.063 (m, 4, ZrCH2), 0.910 (t, 6, CH3); 13C
NMR δ 61.47 (ZrCH2), 53.77 (NCH2), 35.75, 32.89 (BCH),
32.12, 30.15 (BCH), 29.85, 28.78, 28.60, 28.17, 27.63, 27.40,
C
44H86N2B2Zr: C, 69.91; H, 11.46; N, 3.71. Found: C, 70.15;
H, 11.59; N, 3.68.
(CyBen )TiMe2 (14c). A solution of dimethylmagnesium
(1.22 mL, 0.973 mmol, 0.80 M in ether) was added to a stirred,
chilled (-30 °C) solution of (CyBen)TiI2 (0.660 g, 0.927 mmol)
in dichloromethane (50 mL). The solution immediately turned
(54) Noyce, D. S.; Denney, D. B. J . Am. Chem. Soc. 1950, 72, 5743.
(55) Wiley, G. A.; Hershkowitz, R. L.; Rein, B. M.; Chung, B. C. J .
Am. Chem. Soc. 1964, 64, 964.

Group 4 Complexes Containing [R2BNCH2CH2NBR2]2-
Organometallics, Vol. 17, No. 3, 1998 321
light yellow. The reaction mixture was allowed to stand at
-30 °C for 10 min and the Mg salts were precipitated by the
addition of 1,4-dioxane (85 µL, 1.0 mmol). The volatile
components were removed in vacuo, and the residue was
triturated with pentane (5 mL). The yellow, powdery residue
was extracted with pentane (50 mL), the extracts were filtered
through Celite, and the filtrate was concentrated to dryness.
The resulting white solid was recrystallized at -30 °C from
pentane after filtering to afford 0.406 g (90%) of pale yellow
crystals in three crops: 1H NMR (C6D6) δ 3.766 (s, 4, NCH2),
2.0-1.2 (br m, 44, Cy), 1.138 (s, 6, TiMe); 13C NMR δ 54.01,
53.47 (TiMe and NCH2), 28.73, 28.36, 27.35 (Cy). (The BC
resonances were not identified.) Anal. Calcd for C28H54N2B2-
Ti: C, 68.89; H, 11.14; N, 5.74. Found: C, 68.76; H, 11.36; N,
5.83.
the freezer. In cases where polymerization ensued, the solu-
tion became extremely viscous within 5-10 min. After the
mixture was allowed to stand at -30 °C for 1 h, it was
quenched with 1.0 mL of methanol. The viscous solution was
transferred to a preweighed flask, assisted by washing with
dichloromethane, and most of the volatile components were
removed by rotary evacuation under reduced pressure (water
aspirator). The polymer was then dried at 60 mTorr and 110
°C for 8-12 h. The yields reported are not corrected for extra
weight due to the metal complex and activator fragments.
X-r a y str u ctu r es of (Tr ip Ben )Zr (CD3)2 a n d (CyBen )-
Zr (CH2CH3)2. A description of a typical procedure can be
found in a recent publication,56 and the relevant crystal-
lographic data can be found in Table 1. See Supporting
Information for more details.
P r ep a r a tion a n d NMR Ch a r a cter iza tion of th e Alk yl
Ca tion s. A sample of 11 (0.027 mmol) was dissolved in
chlorobenzene-d5, and the solution was chilled to -30 °C. Solid
[Ph3C][B(C6F5)4] (0.025 g, 0.027 mmol) was added to the stirred
solution. The light yellow solution was transferred to an NMR
tube and was frozen at -78 °C before being inserted into a
precooled NMR probe at -30 °C. 1H NMR data are presented
in Table 3: 19F NMR (11a +, -30 °C) δ -137.97 (br s, 2, o-F),
-167.83 (t, 1, p-F), -171.75 (br s, 2, m-F).
1-Hexen e P olym er iza tion s. A sample of a dialkyl com-
plex (0.30-0.38 mg) was dissolved in 5.0 mL of chlorobenzene,
and the solution was chilled in the freezer (-30 °C) for ∼10
min. The solution was briefly removed from the freezer and
solid [Ph3C][B(C6F5)4] (1 equiv) was added with stirring. After
all of the [Ph3C][B(C6F5)4] had dissolved, the solution was
returned to the freezer for 2 min prior to the addition of
1-hexene (1.50 mL). The vial was immediately returned to
Ack n ow led gm en t. R.R.S. thanks the National Sci-
ence Foundation for research support (Grant Nos. CHE
91-22827 and 97-00736) and for funds to help purchase
a departmental Siemens SMART/CCD diffractometer.
We also thank Exxon for gifts of [HNMe2Ph][B(C6F5)4],
[Ph3C][B(C6F5)4], and B(C6F5)3.
Su p p or tin g In for m a tion Ava ila ble: Tables of final
positional parameters, final thermal parameters, and bond
lengths and angles for (TripBen)Zr(CD3)2 and (CyBen)Zr(CH2-
CH3)2 (6 pages). Ordering information is given on any current
masthead page.
OM970765O
(56) Rosenberger, C.; Schrock, R. R.; Davis, W. M. Inorg. Chem.
1997, 36, 123.