Structure/activity relationships for the living Ziegler-Natta polymerization of 1-hexene by the series of cationic monocyclopentadienyl zirconium acetamidinate complexes

Article abstract of DOI:10.1016/S0020-1693(02)01288-4

Protonation of (η⁵-C₅Me₅)ZrMe₂[N(t-Bu)C(Me) N(CH₂R)] (1) [R=(a) Me, (b) i-Pr, (c) t-Bu, (d) Ph, (e) 2-ClC₆H₄, (f) 3-MeC₆H₄ and (g) 2,4,6-Me₃C₆H₂] by [PhNMe₂H][B(C₆F₅)₄] in chlorobenzene at -10°C provides the titled cationic complexes 2a-g. Both 2b and 2d are active initiators for the living polymerization of 1-hexene at -10°C, however, whereas 2d functions much like 2a in terms of activity and isoselectivity, propagation from 2b is much less active and it produces atactic material. Complexes 2d and 2g are inactive, presumably due to steric reasons, while intramolecular chloride coordination appears responsible for inhibiting 2e. Finally, 2f, while active as an initiator, was found not to provide a living system as a result of C-H activation processes that terminate propagation. A structural study of the precatalysts 1b-d sheds additional light on the nature of ligand effects on activity and stereocontrol for the family of initiators represented by 2.

Full text of DOI:10.1016/S0020-1693(02)01288-4

Inorganica Chimica Acta 345 (2003) 121ꢀ129
/
www.elsevier.com/locate/ica
Structure/activity relationships for the living Zieglerꢀ
/
Natta
polymerization of 1-hexene by the series of cationic
monocyclopentadienyl zirconium acetamidinate complexes,
[(h5-C5Me5)ZrMe{N(CH2R)C(Me)N(t-Bu)}][B(C6F5)4] (Rꢁ
i-Pr, t-Bu, Ph, 2-ClC6H4, 3-MeC6H4 and 2,4,6-Me3C6H2)
/
Me,
Denis A. Kissounko, James C. Fettinger, Lawrence R. Sita *
Department of Chemistry and Biochemistry, University of Maryland, College Park, MD 20742, USA
Received 24 May 2002; accepted 23 July 2002
Dedicated to Professor Richard R. Schrock
Abstract
Protonation of (h5-C5Me5)ZrMe2[N(t-Bu)C(Me)N(CH2R)] (1) [Rꢁ
and (g) 2,4,6-Me3C6H2] by [PhNMe2H][B(C6F5)4] in chlorobenzene at ꢂ
and 2d are active initiators for the living polymerization of 1-hexene at ꢂ
/
(a) Me, (b) i-Pr, (c) t-Bu, (d) Ph, (e) 2-ClC6H4, (f) 3-MeC6H4
10 8C provides the titled cationic complexes 2aꢀg. Both 2b
10 8C, however, whereas 2d functions much like 2a in
/
/
/
terms of activity and isoselectivity, propagation from 2b is much less active and it produces atactic material. Complexes 2d and 2g
are inactive, presumably due to steric reasons, while intramolecular chloride coordination appears responsible for inhibiting 2e.
Finally, 2f, while active as an initiator, was found not to provide a living system as a result of CÃ
/
H activation processes that
terminate propagation. A structural study of the precatalysts 1bꢀd sheds additional light on the nature of ligand effects on activity
/
and stereocontrol for the family of initiators represented by 2.
# 2002 Elsevier Science B.V. All rights reserved.
Keywords: ZieglerꢀNatta polymerization; Living polymerization; Poly(1-hexene); Amidinate complexes
/
1. Introduction
syndiotactic [3b,3c,3d] polymer microstructure. Regard-
ing these efforts, we recently reported that activation of
neutral dimethyl monocyclopentadienyl zirconium acet-
amidinates of the general structure, (h5-C5R5)ZrMe2-
Following the initial reports of McConville and
Scollard [1] and Schrock et al. [2], several groups have
now similarly disclosed that certain classes of non-
metallocene based Group 4 complexes can serve as
[N(R1)C(Me)N(R2)] (1) (Rꢁ
/
H or Me), by borate salts
produces highly to extremely active cationic initiators
for the living ZieglerꢀNatta polymerization of a-olefins
initiators for the living ZieglerꢀNatta polymerization of
/
/
propene and long chain a-olefins (e.g. 1-hexene) upon
activation by either excess methylaluminoxane (MAO)
or a stoichiometric amount of a borane or borate
cocatalyst [e.g. B(C6F5)3 and [PhNMe2H][B(C6F5)4],
respectively] [3,4]. Some of these systems are also known
to effect polymerization with a very high degree of
stereocontrol, producing either an isotactic [3a,4] or
and for the living cyclopolymerization of a,v-non-
conjugated dienes [4]. Importantly, depending upon
the nature of the cyclopentadienyl groups, R, and the
two acetamidinate substituents, R1 and R2, either a very
high degree of stereocontrol could be achieved (i.e.
isotactic when Rꢁ
/
Me and R1"
/
R2) or none at all (i.e.
Rꢁ
/
H and/or R1ꢁ
/
R2); except in the case of vinylcy-
clohexane where a highly isotactic microstructure is
obtained regardless of the nature of the initiator,
presumably due to stereocontrol by a chain-end control
* Corresponding author.
E-mail address: 15214@umail.umd.edu (L.R. Sita).
0020-1693/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 0 2 0 - 1 6 9 3 ( 0 2 ) 0 1 2 8 8 - 4

122
D.A. Kissounko et al. / Inorganica Chimica Acta 345 (2003) 121ꢀ129
/
mechanism. In the specific example where Rꢁ
/
Me,
R1ꢁEt, and R2ꢁ
through extensive 2D NMR characterization of a
solvents were employed throughout. Diethyl ether
(Et2O), tetrahydrofuran (THF) and C5H12 were distilled
from Na/benzophenone (with a few milliliters of tri-
glyme being added to the pot in the case of C5H12), and
chlorobenzene (PhCl) was refluxed for several days over
CaH2 before being used for polymerizations. The
monomer 1-hexene (99%) was obtained from Aldrich
and stirred over NaK alloy (1:1) overnight before being
vacuum-transferred. Benzene-d6 was likewise vacuum
transferred from NaK prior to being used for NMR
spectroscopy. Chlorobenzene-d5 was dried over CaH2
prior to vacuum transfer. 1-(1,1-Dimethylethyl)-3-(2-
/
/t-Bu (1a), we have now determined,
purposefully synthesized low molecular weight, narrow
1800, Mw/
polydispersity poly(1-hexene) sample (Mnꢁ
/
Mnꢁ1.06), that propagation proceeds in an isospecific
/
(i.e. 100% stereoselective) and regiospecific (i.e. only 1,2-
insertion) manner [5]. Accordingly, in this system,
propagation must be proceeding through a single highly
stereodifferentiating olefin binding site as opposed to
the well established two-site stereocontrol model of
zirconocenes [6]. This requirement, in turn, necessitates
that rapid site-isomerization following olefin insertion
must be occurring [7], and that epimerization of the
zirconium center through ‘amidinate ring-flipping’ [8]
must be far slower than propagation. Putting these
processes together in terms of their relative rates, the
methylpropyl)carbodiimide,
(2,2-dimethyl-propyl)carbodiimide,
1-(1,1-dimethylethyl)-3-
1-(1,1-dimethy-
lethyl)-3-benzylcarbodiim-ide, 1-(1,1-dimethylethyl)-3-
(2-chlorobenzyl)carbodiimide, 1-(1,1-dimethylethyl)-3-
(3-methyl-benzyl)carbodiimide and 1-(1,1-dimethy-
lethyl)-3-(2,4,6-trimethylbenzyl)carbodiimide were all
prepared through a previously published heterocumu-
lene metathesis route [9]. (h5-C5Me5)ZrCl3 and
[PhNMe2H][B(C6F5)4] were obtained from Strem Che-
micals and Boulder Scientific, respectively, and used
without further purification.
following order must then be true: kiso[Zr]ꢀ
/
kp[Zr][M]ꢀkepi[Zr], where kiso, kp and kepi are the
/
respective rate constants for site-isomerization, propa-
gation, and metal centered epimerization. Given that the
magnitude of kiso, kp, and kepi, and the degree of olefin
face selectivity, should be dependent upon the steric
bulk of the ligand environment about the metal center,
we set out to map, in a systematic manner, the structure/
activity relationships for the family of initiators derived
from 1. In the present work, we present the results of a
study in which 1-hexene polymerization activity and
degree of stereocontrol is correlated with a variation in
the steric bulk of the ethyl substituent in 1a through the
additional synthesis and activation of the precatalysts
GPC analyses were performed using a Waters GPC
system equipped with a column oven and differential
refractometer both maintained at 40 8C and four
˚
columns (Waters Ultrastyragel 500 A, Waters Styragel
HR3, Waters Styragel HR4, and Shodex K-806M).
THF was used as the eluant at a flow rate of 1.1 ml
minꢂ1. Mn and Mw/Mn values were obtained using the
Waters GPC software and seven different polystyrene
standards (Polymer Laboratories). Elemental analyses
were performed by Midwest Microlab Inc. 1H and
13C{1H} NMR spectra were recorded at 400 and 100
MHz, respectively, using C6H6-d6 or PhCl-d5 as the
solvent.
1bꢀ
state geometrical parameters of three of these precata-
lysts, 1bꢀ1d, were compared to those previously re-
/g shown in Eq. (1). As a part of this study, the solid-
/
ported for 1a [4c].
2.2. General procedure for the synthesis of (h5-
C5Me5)ZrMe2[N(CH2R)C(Me)N(t-Bu)] (Rꢁ
Bu, Ph, 2-ClC6H5, 3-MeC6H5 and 2,4,6-Me3C6H2) (1bꢀ
g)
/i-Pr, t-
/
In the glovebox, 5.6 ml of methyllithium (1.6 M in
Et2O), cooled to approximately ꢂ30 8C in an internal
refrigerator, was slowly added dropwise by pipette to a
/
ð1Þ
solution of 1.0 g (3 mmol) of (h5-C5Me5)ZrCl3 in 30 ml
of Et2O, also precooled to approximately ꢂ30 8C. The
/
reaction mixture was then allowed to warm to room
temperature (r.t.) for 1 h, after which time, 3 mmol
(neat) of the requisite carbodiimide was added by
pipette. After stirring at r.t. for 12 h, the volatiles were
removed in vacuo, the oily residue extracted by C5H12,
the C5H12 extract filtered through a glass frit and
concentrated to a volume of approximately 3 ml.
2. Experimental
2.1. General procedures
Manipulations were performed under an inert atmo-
sphere of dinitrogen using standard Schlenk techniques
or a vacuum atmospheres glovebox. Dry, oxygen-free
Overnight cooling of this solution at ꢂ
/
30 8C then
afforded the desired complex as a white crystalline

D.A. Kissounko et al. / Inorganica Chimica Acta 345 (2003) 121ꢀ
/
129
123
material that was recrystallized to obtain analytically
pure compound.
For compound 1b: isolated yield 47%. 1H NMR
(C6H2); 172.4 (CMe). Anal. Calc. for C28H46N2Zr: C,
67.0; H, 9.2; N, 5.6. Found: C, 67.1; H, 9.2; N, 5.5%.
(C6H6-d6, 25 8C): d 0.15 (s, 6H, ZrMe); 0.75 (d, 6H,
2.3. General procedure for in situ generation of initiators
Jꢁ/6.4 Hz, CHMe2); 1.12 (s, 9H, CMe3); 1.61 (m, 1H,
2bꢀg and polymerization of 1-hexene
/
CHMe2); 1.70 (s, 3H, CMe); 1.95 (s, 15H, C5Me5); 2.73
(s, 2H, CH2). 13C{1H} NMR (C6H6-d6, 25 8C): d 11.6
(CMe3); 15.7 (ZrMe); 20.4 (CHMe2); 32.1 (C5Me5); 30.4
(CH2); 45.3 (CMe); 52.6 (CMe3); 55.1 (CHMe2); 119.2
(C5Me5); 173.5 (CMe). Anal. Calc. for C22H42N2Zr: C,
62.1; H, 9.9; N, 6.6. Found: C, 62.0; H, 9.9; N, 6.4%.
For compound 1c: isolated yield 81%. 1H NMR
(C6H6-d6, 25 8C): d 0.21 (s, 6H, ZrMe); 0.80 (s, 9H,
CH2CMe3); 1.13 (s, 9H, CMe3); 1.74 (s, 3H, CMe); 1.93
(s, 15H, C5Me5); 2.85 (s, 2H, CH2). 13C{1H} NMR
(C6H6-d6, 25 8C): d 11.8 (CMe3); 16.3 (ZrMe); 28.6
(CH2CMe3); 32.3 (C5Me5); 33.2 (CH2); 46.9 (CMe);
53.1 (CMe3); 58.2 (CH2CMe3); 119.1 (C5Me5); 173.8
(CMe). Anal. Calc. for C23H44N2Zr: C, 62.8; H, 10.1; N,
6.4. Found: C, 62.9; H, 10.1; N, 6.2%.
Within the internal refrigerator of a glovebox, main-
tained at ꢂ10 8C and equipped with a low profile mini
magnetic stirrer, the precatalyst, 1b (0.02 mmol) was
dissolved in 2 ml of PhCl within a glass vial. In a
separate vial, 0.016 g (0.02 mmol) of [PhNHMe2]-
[B(C6F5)4] (cocatalyst) was dissolved in 8 ml of PhCl.
/
After cooling to ꢂ10 8C, the solution of precatalyst
/
was added rapidly by pipette to the cocatalyst to give a
clear yellow solution of the corresponding cationic
species 2b. To this rapidly stirred solution of initiator,
0.55 ml (4 mmol) of 1-hexene was then rapidly added all
at once and the reaction mixture was stirred at ꢂ10 8C
/
for 2 h, whereupon it was quenched by the addition of
silica gel (degassed). After filtering, all the volatiles were
removed in vacuo, the oily residue redissolved in
approximately 5 ml of CHCl3 and the poly(1-hexene)
precipitated by dropwise addition into 400 ml of
acidified (HCl) MeOH. The purified polymer was then
isolated and dried overnight at 70 8C/0.01 mmHg.
For compound 1d: isolated yield 53%. 1H NMR
(C6H6-d6, 25 8C): d 0.20 (s, 6H, ZrMe); 1.11 (s, 9H,
CMe3); 1.60 (s, 3H, CMe); 1.93 (s, 15H, C5Me5); 4.13 (s,
2H, CH2); 7.10ꢀ
7.13 (m, 5H, C6H5). 13C{1H} NMR
/
(C6H6-d6, 25 8C): d 11.5 (CMe3); 16.0 (ZrMe); 31.8
(C5Me5); 44.6 (CMe); 50.7 (CH2); 52.7 (CMe3); 119.5
1
For obtaining the H NMR spectra of the initiators,
(C5Me5); 126.0ꢀ/142.7 (C6H5); 174.6 (CMe). Anal. Calc.
2bꢀg, a solution of the precatalyst (0.02 mmol in
/
for C25H40N2Zr: C, 65.3; H, 8.8; N, 6.1. Found: C, 65.5;
approximately 0.5 ml of PhCl-d5, was added to a
suspension of 16 mg (0.02 mmol) of [PhNHMe2]-
[B(C6F5)4] in approximately 0.5 ml of PhCl-d5, also
H, 8.8; N, 6.0%.
For compound 1e: isolated yield 67%. 1H NMR
(C6H6-d6, 25 8C): d 0.22 (s, 6H, ZrMe); 1.10 (s, 9H,
CMe3); 1.51 (s, 3H, CMe); 1.89 (s, 15H, C5Me5); 4.33 (s,
precooled to ꢂ10 8C. The resulting clear yellow solu-
/
tion of the corresponding cationic species was then
transferred to a NMR tube that was kept cooled to
2H, CH2); 6.72ꢀ
7.30 (m, 4H, C6H4). 13C{1H} NMR
/
(C6H6-d6, 25 8C): d 11.4 (CMe3); 15.5 (ZrMe); 31.8
(C5Me5); 45.1 (CMe); 48.5 (CH2); 52.7 (CMe3); 119.6
ꢂ10 8C in a dewar before being inserted into the NMR
probe that was maintained at the same temperature.
/
(C5Me5); 123.1ꢀ139.7 (C6H4); 175.0 (CMe). Anal. Calc.
/
for C25H39ClN2Zr: C, 60.8; H, 8.0; N, 5.7. Found: C,
2.4. Crystal structure determinations
60.7; H, 7.9; N, 5.7%.
For compound 1f: isolated yield 46%. 1H NMR
(C6H6-d6, 25 8C): d 0.21 (s, 6H, ZrMe); 1.12 (s, 9H,
CMe3); 1.64 (s, 3H, CMe); 1.94 (s, 15H, C5Me5); 2.12 (s,
Data was collected on a Bruker SMART CCD system
operating at ꢂ80 8C. All crystallographic calculations
/
were performed on a Personal computer (PC) with a
Pentium 1.80 GHz processor and 512 MB of extended
memory. The SHELXTL program package was imple-
mented to determine the probable space group and set
up the initial files. Table 1 provides information on the
data collection and refinement parameters for com-
3H, C6H4Me); 4.14 (s, 2H, CH2); 6.83ꢀ67.11 (m, 4H,
/
C6H4). 13C{1H} NMR (C6H6-d6, 25 8C): d 11.5
(CMe3); 16.0 (ZrMe); 21.1 (C6H4Me); 31.8 (C5Me5);
44.6 (CMe); 50.6 (CH2); 52.7 (CMe3); 119.5 (C5Me5);
123.1ꢀ142.5 (C6H4); 174.7 (CMe). Anal. Calc. for
/
C26H42N2Zr: C, 65.9; H, 8.9; N, 5.9. Found: C, 65.72;
pounds 1bꢀd.
/
H, 8.91; N, 5.87%.
For compound 1g: isolated yield 54%. 1H NMR
(C6H6-d6, 25 8C): d 0.17 (s, 6H, ZrMe); 1.18 (s, 9H,
CMe3); 1.70 (s, 3H, CMe); 1.85 (s, 15H, C5Me5); 2.10 (s,
3H, p-MeC6H2); 2.22 (s, 6H, o-MeC6H2); 4.18 (s, 2H,
CH2); 6.74 (s, 2H, C6H2). 13C{1H} NMR (C6H6-d6,
25 8C): d 11.4 (CMe3); 16.4 (ZrMe); 20.5 (p- MeC6H2);
20.6 (o-MeC6H2); 32.0 (C5Me5); 44.9 (CMe); 47.2
2.4.1. Crystal structure determination of compound 1b
A colorless plate with approximate orthogonal dimen-
sions 0.492ꢃ
/
0.306ꢃ
/
0.092 mm3 was placed and opti-
cally centered on the Bruker SMART CCD system at ꢂ
/
80 8C. The initial unit cell was indexed using a least-
squares analysis of a random set of reflections collected
from three series of 0.38 wide v-scans, 10 s per frame,
and 25 frames per series that were well distributed in
(CH2); 52.9 (CMe3); 119.5 (C5Me5); 127.4, ꢂ136.3
/

124
D.A. Kissounko et al. / Inorganica Chimica Acta 345 (2003) 121ꢀ129
/
Table 1
Crystallographic data and details of refinement for compounds 1bꢀ
/
d
1b
1c
1d
Empirical formula
Formula weight
Temperature (K)
Crystal system
Space group
C
22H42N2Zr
C23H44N2Zr
439.82
193(2)
C25H40N2Zr
459.81
193(2)
425.80
193(2)
triclinic
monoclinic
P21/c
monoclinic
P21/c
¯
/
P1/
˚
a (A)
8.7213(4)
9.8836(4)
14.5957(6)
72.8290(10)
76.9400(10)
82.8010(10)
1168.63(9)
2
15.9229(6)
17.7627(6)
17.6180(6)
90
8.5629(5)
29.7302(16)
10.1619(6)
90
˚
b (A)
˚
c (A)
a (8)
b (8)
g (8)
100.3150(10)
90
106.1140(10)
90
3
˚
V (A )
Z
4902.4(3)
8
1.192
2485.3(2)
4
1.229
rcalc (g cmꢂ3
)
1.210
0.477
m(Mo) (mmꢂ1
)
0.457
0.454
Crystal size (mm)
2u Range (8)
Total reflections
0.492ꢃ
/
0.306ꢃ
/0.092
0.607ꢃ
/
0.388ꢃ
/0.286
0.617ꢃ
/
0.274ꢃ0.108
/
2.26ꢀ27.49
/
1.64ꢀ27.50
/
2.20ꢀ27.50
/
18 551
78 069
39 635
Independent reflections
Data/restraints/parameters
5330 [Rint
ꢁ
/
0.0182]
11 257 [Rint
ꢁ
/
0.0269]
5712 [Rint
ꢁ0.0379]
/
5330/3/398
R1ꢁ0.0215, wR2ꢁ
R1ꢁ0.0245, wR2ꢁ
1.098
11 257/6/828
R1ꢁ0.0322, wR2ꢁ
R1ꢁ0.0391, wR2ꢁ
1.079
5712/3/417
R1ꢁ0.0307, wR2ꢁ
R1ꢁ0.0416, wR2ꢁ
1.086
Final R indices [I ꢁ
/
2s(I)]
/
/
0.0575
/
/0.0871
/
/
0.0832
R indices (all data)
/
/0.0596
/
/0.0924
/
/
0.0902
Goodness-of-fit
reciprocal space. Data frames were collected [Mo Ka]
with 0.38 wide v-scans, 14 s per frame and 606 frames
per series. Five complete series were collected at varying
2.4.2. Crystal structure determination of compound 1c
A colorless irregular block with approximate ortho-
gonal dimensions 0.607ꢃ
/
0.388ꢃ
0.286 mm3 was placed
/
and optically centered on the diffractometer. The initial
unit cell was indexed using a least-squares analysis of a
random set of reflections collected from three series of
0.38 wide v-scans, 10 s per frame, and 25 frames per
series that were well distributed in reciprocal space.
Data frames were collected [Mo Ka] with 0.38 wide v-
scans, 10 s per frame and 606 frames per series. Five
f angles (fꢁ
detector distance was 4.888 cm, thus providing a
complete sphere of data to 2umax 50.08. A total of
18 551 reflections were collected and corrected for Lp
effects and absorption using Blessing’s method as
incorporated into the program SADABS [1,2] with 5330
/
0, 72, 144, 216, 2888). The crystal to
ꢁ
/
unique [Rint
tematic absences and intensity statistics indicated the
ꢁ0.0192]. System symmetry, lack of sys-
/
complete series were collected at varying f angles (fꢁ
/
0, 72, 144, 216, 2888). An additional 200 frames, a repeat
of the first series for redundancy and decay purposes,
were also collected. The crystal to detector distance was
4.813 cm, thus providing a complete sphere of data to
chiral triclinic space group P1 (no. 1); however the
¯
centric triclinic space group P1 (no. 2) was chosen. The
structure was determined by direct methods with the
successful location of nearly all non-hydrogen atoms
using the program XS [4]. The structure was refined with
XL [5]. One least-squares difference-Fourier cycle was
required to locate the remaining non-hydrogen atoms.
All non-hydrogen atoms were refined anisotropically.
The hydrogen atoms were initially placed in calculated
positions but allowed to refine freely during the final
stages of refinement (xyzU). The final structure was
2umax
lected and corrected for Lp effects and absorption using
ꢁ55.08. A total of 78 069 reflections were col-
/
Blessing’s method as incorporated into the program
SADABS with 11 257 unique [Rint
ꢁ0.0269]. System
/
symmetry, systematic absences and intensity statistics
indicated the unique centric monoclinic space group
P21/c (no. 14). The structure was determined by direct
methods with the successful location of nearly all non-
hydrogen atoms using the program XS. The structure
was refined with XL. One least-squares difference-Four-
ier cycle was required to locate the remaining non-
hydrogen atoms. All non-hydrogen atoms were refined
anisotropically. Hydrogen atoms were initially placed in
calculated positions; then allowed to refine freely during
refined to convergence [D/s 5
2.45%, wR(F2)ꢁ
5.96%, GOFꢁ
unique reflections [R(F)ꢁ
2.15%, wR(F2)ꢁ
those 4951 data with Foꢁ
/
0.002] with R(F)ꢁ
/
/
/1.098 for all 5330
/
/5.75% for
/
4s(Fo)]. The final difference-
Fourier map was featureless indicating that the structure
is both correct and complete.

D.A. Kissounko et al. / Inorganica Chimica Acta 345 (2003) 121ꢀ
/
129
125
the final refinement stages. The final structure was
conveniently prepared in high yield through carbodii-
mide insertion into a zirconiumÃ
methyl bond of (h5-
refined to convergence [D/s 5
3.91%, wR(F2)ꢁ
9.24%, GOFꢁ
unique reflections [R(F)ꢁ
3.22%, wR(F2)ꢁ
those 9832 data with Foꢁ
Fourier map possessed one large peak, 1.25 e/A , within
/
0.003] with R(F)ꢁ
/
/
/
/1.079 for all 11 257
C5R5)ZrMe3 that is generated in situ from the corre-
sponding (h5-C5R5)ZrCl3 reagent according to Eq. (2)
[4a,4d,8]. Furthermore, the required carbodiimide for
this process, if not commercially available, can be
readily obtained through a tin(II)-mediated heterocu-
mulene metathesis between a N-trimethylsilyl amine,
R1NH(SiMe3), and an isocyanate, R2NCO [9]. For the
/
/8.71% for
/
4s(Fo)]. The final difference-
2
˚
˚
0.8 A of the central Zr(1), labeled atom while the
remainder of the map was featureless indicating that the
structure is both correct and complete.
present study, the precatalysts 1bꢀg were all obtained
/
2.4.3. Crystal structure determination of compound 1d
A colorless plate with approximate orthogonal dimen-
through this carbodiimide insertion route in unopti-
mized isolated yields ranging from 45 to 81% as
analytically pure white crystalline materials.
sions 0.617ꢃ
/
0.274ꢃ
0.108 mm3 was placed and opti-
/
cally centered on the diffractometer. The initial unit cell
was indexed using a least-squares analysis of a random
set of reflections collected from three series of 0.38 wide
v-scans, 10 s per frame, and 25 frames per series that
were well distributed in reciprocal space. Data frames
were collected [Mo Ka] with 0.38 wide v-scans, 16 s per
frame and 606 frames per series. Five complete series
ð2Þ
were collected at varying f angles (fꢁ0, 72, 144, 216,
/
2888). An additional 200 frames, a repeat of the first
series for redundancy and decay purposes, were also
collected. The crystal to detector distance was 4.813 cm,
3.2. Crystal structures of compounds 1bꢀd
/
thus providing a complete sphere of data to 2umax
ꢁ
/
As it has proven not to be routinely possible to obtain
single crystals of the cationic initiators themselves, we
sought to examine what consequences an increase in the
steric bulk of the ethyl substituent of 2a might have on
the nature of bonding of the monocyclopentadienyl/
acetamidinate ligand set within the series of new cationic
55.08. A total of 39 635 reflections were collected and
corrected for Lp effects and absorption using Blessing’s
method as incorporated into the program SADABS with
5712 unique [Rint
ꢁ0.0379]. System symmetry, systema-
/
tic absences and intensity statistics indicated the unique
centric monoclinic space group P21/c (no. 14). The
structure was determined by direct methods with the
successful location of nearly all non-hydrogen atoms
using the program XS. The structure was refined with
XL. One least-squares difference-Fourier cycle was
required to locate the remaining non-hydrogen atoms.
All non-hydrogen atoms were refined anisotropically.
Hydrogen atoms were initially placed in calculated
positions; then allowed to refine freely during the final
refinement stages. A centroid for the pentamethylcyclo-
pentadienyl ligand was calculated. The final structure
species, 2bꢀ
obtained for the precatalysts 1bꢀ
obtained for the precatalyst 1a [4c]. Figs. 1ꢀ
the molecular structures for compounds 1bꢀd, respec-
/
g, by comparing the geometrical parameters
d with those previously
3 present
/
/
/
tively, as obtained from single-crystal X-ray crystal-
lographic analyses, and Table 2 provides selected
geometrical parameters as derived from these structures.
Using the geometrical parameters of 1a as a bench-
mark, several key features of the molecular structures of
was refined to convergence [D/s 5
4.16%, wR(F2)ꢁ
9.02%, GOFꢁ1.086 for all 5712
unique reflections [R(F)ꢁ 8.32% for
3.07%, wR(F2)ꢁ
those 4725 data with Foꢁ4s(Fo)]. The final difference-
/
0.001] with R(F)ꢁ
/
/
/
/
/
/
Fourier map was featureless indicating that the structure
is both correct and complete.
3. Results and discussion
3.1. Preparation of precatalysts 1bꢀg
/
As previously described, neutral dimethyl monocy-
clopentadienyl zirconium acetamidinates of the general
formula, (h5-C5R5)ZrMe2[N(R1)C(Me)N(R2)], can be
Fig. 1. Molecular structure of compound 1b (30% thermal ellipsoids).
Hydrogen atoms have been removed for the sake of clarity.

126
D.A. Kissounko et al. / Inorganica Chimica Acta 345 (2003) 121ꢀ129
/
1a. For the nitrogen atom bearing the t-butyl group,
N(2), once more the a uN(2) values are nearly identical
for 1a and 1d, 357.26 and 357.688, respectively, whereas
in 1b and 1c, it is somewhat smaller at 356.2 and
356.478, respectively. A final consequence of strong non-
bonded interactions that is evident in the molecular
structure of 1c is the significant degree of non-planarity
displayed by the four-membered ring of the amidinate
fragment as measured by the angle between the planes
defined by Zr(1)Ã
/
N(1)Ã
/
C(11) and Zr(1)Ã
/
N(2)Ã/C(11).
In 1c, this ‘envelope’ angle is 33.68, whereas in 1a and
1d, it is nearly identical at 18.3 and 18.18, respectively.
Interestingly, the corresponding value for 1b is the
smallest at only 14.98. In summary then, while the
molecular structure of 1d is virtually identical to 2a with
respect to these specific geometrical parameters, that of
1c is quite different and 1b is closer to 1a than to 1c.
3.3. Solution configurational stabilities of precatalysts
Fig. 2. Molecular structure of compound 1c (30% thermal ellipsoids).
Hydrogen atoms have been removed for the sake of clarity.
1bꢀ
/
g and cationic initiators 2bꢀg
/
In keeping with previous observations [4], the C1-
symmetric precatalysts 1bꢀg are all configurationally
unstable in solution at room temperature with respect to
racemization of the zirconium center that occurs
through a facile amidinate ring-flipping process that
serves to exchange the diastereotopic methylene protons
1bꢀd can be pointed out. To begin, the most obvious
/
/
differences are to be found between compounds 1a and
1c, presumably as a result of a significant increase in
non-bonded steric interactions between substituents of
the latter. For instance, the Zr(1)Ã
/
N(1) bond distance of
˚
1c is considerably longer at 2.3234(15) A than the
of the NÃ
/
CHaHbR substituent. For precatalyst 1a, the
%
˚
corresponding distance of 2.251(3) A in 1a. The nitrogen
barrier to this racemization process, DGc , was pre-
viously found to be only 10.9 kcal molꢂ1 at a low Tc of
223 K [4c], and therefore, it is clear that even a
atom that bears the neopentyl group in 1c, N(1), is also
substantially more pyramidalized than the nitrogen
atom in 1a that bears the ethyl group as revealed by a
comparison of the sum of the bond angles about each
substantial increase in steric bulk of the NÃ
/
CH2R
group, such as that presented by 1c, does not serve to
raise this barrier significantly.
In contrast to the configurational instability of the
nitrogen atom (c.f. a uN(1)
ꢁ
/
353.52 vs. 343.5128 in 1a
and 1c, respectively). In contrast, both the Zr(1)Ã
/N(1)
precatalysts, the cationic initiators 2bꢀ/d were all found
bond distances and the a uN(1) values for 1b and 1d,
˚
˚
to be configurationally stable at ꢂ10 8C on the NMR
/
2.251 A, 353.628 and 2.552(17) A; 353.278, respectively,
are very similar to the corresponding values found for
timeframe as indicated by diagnostic resonances with
narrow line widths now being observed for the diaster-
eotopic protons mentioned above. Fig. 4 presents the 1H
NMR spectrum for 2d where these diastereotopic
resonances show up as an AB quartet (see inset).
Surprisingly, the cationic species 2e, which has a 2-
chlorobenzyl substituent, was found to be configura-
tionally unstable by 1H NMR spectroscopy down to
ꢂ30 8C, which is the temperature limit of the solvent.
/
The origin of this difference in configurationally be-
tween 2d and 2e may be due to the ability of the 2-
chlorobenzyl substituent to alleviate the electron defi-
ciency of the cationic zirconium center through an
intramolecular coordination of the chlorine atom to
zirconium, thereby, decreasing the barrier to amidinate
ring-flipping through a lengthening of zirconiumÃ
gen bonds. Finally, similar 1H NMR studies of 2f and 2g
found that these initiators are chemically unstable at ꢂ
10 8C, potentially as a result of facile CÃH activation
/
nitro-
/
Fig. 3. Molecular structure of compound 1d (30% thermal ellipsoids).
Hydrogen atoms have been removed for the sake of clarity.
/

D.A. Kissounko et al. / Inorganica Chimica Acta 345 (2003) 121ꢀ
/
129
127
Table 2
Selected bond distances (A) and angles (8) for compounds 1aꢀ
˚
/
d
1b
1c
1d
Bond distances
Zr(1)Ã
Zr(1)Ã
Zr(1)Ã
Zr(1)Ã
N(1)Ã
N(1)Ã
N(2)Ã
N(2)Ã
/
N(1)
N(2)
C(21)
C(22)
2.2465(11)
2.2829(11)
2.2714(16)
2.2717(16)
1.3328(17)
1.4640(17)
1.3355(16)
1.4851(16)
Zr(1)Ã
Zr(1)Ã
Zr(1)Ã
Zr(1)Ã
N(1)Ã
N(1)Ã
N(2)Ã
N(2)Ã
/
N(1)
N(2)
C(22)
C(23)
2.3234(15)
2.2439(15)
2.269(2)
2.263(2)
1.333(2)
1.462(2)
1.342(2)
1.485(2)
Zr(1)Ã
Zr(1)Ã
Zr(1)Ã
Zr(1)Ã
N(1)Ã
N(1)Ã
N(2)Ã
N(2)Ã
/
N(1)
N(2)
C(22)
C(23)
2.2552(17)
2.2910(16)
2.274(2)
/
/
/
/
/
/
/
/
/
2.2910(16)
1.341(3)
/
C(11)
C(13)
C(11)
C(17)
/
C(11)
C(13)
C(11)
C(18)
/
C(11)
C(13)
C(11)
C(20)
/
/
/
1.454(3)
/
/
/
1.330(3)
/
/
/
1.486(2)
Bond angles
C(11)ÃN(1)Ã
C(11)ÃN(1)Ã
C(13)ÃN(1)Ã
C(11)ÃN(2)Ã
C(11)ÃN(2)Ã
C(18)ÃN(2)Ã
Zr(1)ÃN(1)Ã
Zr(1)ÃN(2)Ã
/
/
C(13)
Zr(1)
Zr(1)
C(17)
Zr(1)
Zr(1)
122.70(12)
94.37(8)
136.58(9)
124.27(11)
92.67(8)
139.18(8)
14.9
C(11)Ã
C(11)Ã
C(13)Ã
C(11)Ã
C(11)Ã
C(18)Ã
Zr(1)ÃN(1)Ã
Zr(1)ÃN(2)Ã
/
N(1)Ã
N(1)Ã
N(1)Ã
N(2)Ã
N(2)Ã
N(2)Ã
/
C(13)
Zr(1)
Zr(1)
C(18)
Zr(1)
Zr(1)
122.46(16)
88.05(10)
133.00(12)
124.72(16)
90.79(11)
140.96(12)
33.6
C(11)Ã
C(11)Ã
C(13)Ã
C(11)Ã
C(11)Ã
C(20)Ã
Zr(1)ÃN(1)Ã
Zr(1)ÃN(2)Ã
/
N(1)Ã
N(1)Ã
N(1)Ã
N(2)Ã
N(2)Ã
N(2)Ã
/
C(13)
Zr(1)
Zr(1)
C(20)
Zr(1)
Zr(1)
122.14(19)
94.42(12)
136.71(16)
125.07(17)
93.11(12)
139.50(14)
18.1
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
C(11) and
/
/
C(11) and
C(11)
/
/
C(11) and
/
/
C(11)
/
/
/
/C(11)
processes involving the methyl substituents on the
phenyl rings. Schrock et al. [10] have previously seen
ment most likely plays a key role in shutting down
activity, while for 1e, intramolecular coordination of the
chlorobenzyl substituent may be responsible for block-
ing olefin binding. Finally, for initiator 1f, which is
decomposition through similar types of CÃ
/
H activation
of mesityl substituents in their cationic initiators.
unstable at ꢂ10 8C in the absence of olefin, a high yield
/
of isotactic poly(1-hexene) can still be obtained, but as
the polydispersity value reveals, this polymerization is
no longer living.
3.4. Polymerization of 1-hexene by initiators 2aꢀg
/
Table 3 presents the results of a study of 1-hexene
polymerization by initiators 2aꢀg that was conducted at
10 8C. Where polymer was obtained, this table also
/
ꢂ
/
presents the mean average molecular weight, Mn, and
polydispersity, Mw/Mn, of the material as obtained by
GPC analysis and the microstructure of the poly(1-
hexene) as revealed by 13C NMR spectroscopy [4a].
Using the results obtained with 1a as a point of
reference, it can be clearly seen that the benzyl-sub-
stituted initiator 1d behaves in a nearly identical fashion
by quantitatively providing isotactic poly(1-hexene) of
similar Mn and Mw/Mn values. Although we have not
carried out a kinetic evaluation of 1-hexene polymeriza-
tion using 1d, we are confident that, as found for 1a,
polymerizations are living with this initiator under these
specific conditions. In contrast to the high activities
expressed by 1a and 1d where 1-hexene polymerizations
4. Conclusions
The present structure/activity study serves to show
how subtle differences in the ligand environment of the
amidinate-based family of ZieglerꢀNatta initiators de-
/
rived from 1 can have a profound influence on both
activity and the degree of stereocontrol that can be
obtained. Surprisingly, it was found that while the
benzyl substituent of 2d can adopt a conformation
that apparently does not interfere with the nature of
the olefin binding site established by 2a (i.e. with respect
to both the direction of site-isomerization and the level
of non-bonded steric interactions with the bound olefin),
an isobutyl substituent cannot. Importantly, we believe
that the loss of stereocontrol that is observed in the
latter case with 2b is due to the site-isomerization
process that no longer has a clear discriminating choice
of placing the growing polymer chain towards one
amidinate substituent or the other, which in turn leads
to promiscuous olefin binding and an atactic polymer
microstructure. The present study has also served to
further outline the steric limitations of the amidinate
substituents and their ability to engage in either
intramolecular coordination to the electron-deficient
are complete within 2 h at ꢂ
/
10 8C (90ꢀ95% isolated
/
yields), the polymerization mediated by 1b is consider-
ably more sluggish as evidenced by both a low 57% yield
of isolated material that possesses a correspondingly
much lower Mn value (see Table 3). Interestingly, while
the narrow polydispersity that is observed with this
initiator is indicative that polymerizations with 1b are
also living, it was surprising to find that an atactic
polymer microstructure is now obtained. For 1d and 1g,
significant steric congestion within the ligand environ-

128
D.A. Kissounko et al. / Inorganica Chimica Acta 345 (2003) 121ꢀ129
/
Fig. 4. 1H NMR (500 MHz, chlorobenze-d5, ꢂ
10 8C) spectrum of 2d.
/
Road, Cambridge CB2 1EZ, UK (fax: ꢄ44-1223-336-
033; e-mail: deposit@ccdc.cam.ac.uk or www: http://
www.ccdc.cam.ac.uk).
/
Table 3
Polymerization activity and polymer characteristics for initiators 2aꢀg
/
a
Initiator
Mn
Mw/Mn
Microstructure
2a
2b
2c
2d
2e
2f
19 800
7940
1.03
1.03
isotactic
atactic
Acknowledgements
b
na
19 340
na
1.10
1.48
isotactic
isotactic
This work was supported by the National Science
Foundation (CHM-0092493) for which we are grateful.
23 580
na
2g
a
As determined by 13C NMR (100 MHz, CDCl3, 25 8C) spectro-
References
scopy.
b
Not active.
[1] J.D. Scollard, D.H. McConville, J. Am. Chem. Soc. 118 (1996)
10008.
metal center or in undesirable and terminating CÃ
/
H
[2] R. Baumann, W.M. Davis, R.R. Schrock, J. Am. Chem. Soc. 119
(1997) 3830.
activation processes.
[3] (a) E.Y. Tshuva, I. Goldberg, M. Kol, J. Am. Chem. Soc. 122
(2000) 10706;
(b) J. Tian, P.D. Hustad, G.W. Coates, Angew. Chem., Int. Ed.
Engl. 39 (2000) 3626;
5. Supplementary material
(c) J. Saito, M. Mitani, J. Mohri, Y. Yoshida, S. Matsui, S. Ishii,
S. Kojoh, N. Kashiwa, T. Fujita, Angew. Chem., Int. Ed. Engl. 40
(2001) 2918;
Crystallographic data (excluding structure factors)
for the structures in this paper have been deposited
with the Cambridge Crystallographic Data Centre,
(d) J. Saito, M. Mitani, J. Mohri, S. Ishii, Y. Yoshida, T. Matsugi,
S. Kojoh, N. Kashiwa, T. Fujita, Chem. Lett. (2001) 576.
[4] (a) K.C. Jayaratne, L.R. Sita, J. Am. Chem. Soc. 122 (2000) 958;
(b) K.C. Jayaratne, R.J. Keaton, D.A. Henningsen, L.R. Sita, J.
Am. Chem. Soc. 122 (2000) 10490;
CCDC Nos. 197488ꢀ197490 for compounds 1b, 1c
/
and 1d, respectively. Copies of this data can be obtained
free of charge from The Director, CCDC, 12 Union

D.A. Kissounko et al. / Inorganica Chimica Acta 345 (2003) 121ꢀ
/
129
129
(c) R.J. Keaton, K.C. Jayaratne, J.C. Fettinger, L.R. Sita, J. Am.
Chem. Soc. 122 (2000) 12909;
[7] (a) J.A. Ewen, J. Mol. Catal. A: Chem. 128 (1998) 103;
(b) M.A. Giardello, M.S. Eisen, C.L. Stern, T.J. Marks, J. Am.
Chem. Soc. 117 (1995) 12114.
(d) R.J. Keaton, K.C. Jayaratne, D.A. Henningsen, L.A. Koter-
was, L.R. Sita, J. Am. Chem. Soc. 123 (2001) 6197;
(e) K.C. Jayaratne, L.R. Sita, J. Am. Chem. Soc. 123 (2001)
10754.
[8] (a) L.R. Sita, J.R. Babcock, Organometallics 17 (1998) 5228;
(b) L.A. Koterwas, J.C. Fettinger, L.R. Sita, Organometallics 18
(1999) 4183.
[5] K.C. Jayaratne, Y.F. Lam, L.R. Sita, results to be published.
[6] H.H. Brintzinger, D. Fischer, R. Mullhaupt, B. Rieger, R.M.
¨
[9] J.R. Babcock, L.R. Sita, J. Am. Chem. Soc. 120 (1998) 5585.
[10] R.R. Schrock, P.J. Bonitatebus, Jr., Y. Schrodi, Organometallics
Waymouth, Angew. Chem., Int. Ed. Engl. 34 (1995) 1143.
20 (2001) 1056ꢀ1058.
/