Zirconium complexes with versatile β-diketiminate ligands: Synthesis, structure, and ethylene polymerization

Article abstract of DOI:10.1016/j.jorganchem.2008.08.024

A series of zirconium complexes (2c, 2d, 2f, 2g, 2h, 2i) containing symmetrical or unsymmetrical β-diketiminate ligands were synthesized by the reaction of ZrCl₄ · 2THF with lithium salt of the corresponding ligand in 1:2 molar ratio. X-ray crystal structures reveal that complexes 2d and 2g adopt distorted octahedral geometry around the zirconium center. These complexes showed moderate activities for ethylene polymerization, when methylaluminoxane (MAO) was used as cocatalyst. The steric and electronic effects of the substituents at the phenyl rings had considerable influence on the catalytic activities of the metal complex, as well as the molecular weights and molecular weight distributions (MWD) of produced polymers. Introduction of electron-withdrawing CF₃ group to phenyls in the ligand led to a significant increase of catalytic activities, and complex 2f (p-CF₃) exhibited the highest catalytic activity of 7.45 × 10⁵ g PE/mol-Zr · h among the investigated complexes. Complexes 2a-d could produce ultra-high molecular weight polyethylenes (UHMWPE) that were hardly dissolvable in decahydronaphthalene or 1,2-dichlorobenzene under the molecular weight measurement conditions. Nevertheless, polyethylenes with broad MWD could be afforded by complexes 2g-i, which was probably due to the introduction of bulky unsymmetrical ligands leading to the formation of multi active species under polymerization conditions. High-temperature ¹³C NMR data indicate the linear structure of obtained polyethylenes.

Full text of DOI:10.1016/j.jorganchem.2008.08.024

Journal of Organometallic Chemistry 693 (2008) 3509–3518
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Zirconium complexes with versatile b-diketiminate ligands: Synthesis,
structure, and ethylene polymerization
*
*
Shaogang Gong, Haiyan Ma , Jiling Huang
Laboratory of Organometallic Chemistry, East China University of Science and Technology, 130 Meilong Road, Shanghai 200237, People’s Republic of China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 28 May 2008
Received in revised form 4 August 2008
Accepted 18 August 2008
Available online 26 August 2008
A series of zirconium complexes (2c, 2d, 2f, 2g, 2h, 2i) containing symmetrical or unsymmetrical
b-diketiminate ligands were synthesized by the reaction of ZrCl4 Á 2THF with lithium salt of the corre-
sponding ligand in 1:2 molar ratio. X-ray crystal structures reveal that complexes 2d and 2g adopt dis-
torted octahedral geometry around the zirconium center. These complexes showed moderate activities
for ethylene polymerization, when methylaluminoxane (MAO) was used as cocatalyst. The steric and
electronic effects of the substituents at the phenyl rings had considerable influence on the catalytic activ-
ities of the metal complex, as well as the molecular weights and molecular weight distributions (MWD)
of produced polymers. Introduction of electron-withdrawing CF3 group to phenyls in the ligand led to a
significant increase of catalytic activities, and complex 2f (p-CF3) exhibited the highest catalytic activity
of 7.45 Â 105 g PE/mol-Zr Á h among the investigated complexes. Complexes 2a–d could produce ultra-
high molecular weight polyethylenes (UHMWPE) that were hardly dissolvable in decahydronaphthalene
or 1,2-dichlorobenzene under the molecular weight measurement conditions. Nevertheless, polyethyl-
enes with broad MWD could be afforded by complexes 2g–i, which was probably due to the introduction
of bulky unsymmetrical ligands leading to the formation of multi active species under polymerization
conditions. High-temperature 13C NMR data indicate the linear structure of obtained polyethylenes.
Ó 2008 Elsevier B.V. All rights reserved.
Keywords:
b-Diketiminate
Zirconium
Polyethylene
Catalysis
Polymerization
1. Introduction
tantly to explore the potential of metal complexes with other
ligands except for Cp to polymerize ethylene and other olefinic
Over the past decades, considerable development of olefin poly-
merization catalysts of high-performance has created a variety of
polymers with novel properties. For example, ultra-high molecular
weight polyethylenes (UHMWPE) having a molecular weight (Mw)
greater than 3000000 possess excellent abrasion resistance and
impact strength, very low coefficient friction, good self-lubricants
properties, as well as good resistance at low temperature [1–8].
Polyethylenes with broad or multi-model molecular weight distri-
bution (MWD) can meet the requirement for good resin process-
ability, since they show high workability due to the high
molecular weight (HMW) fractions, and at the same time they pro-
vide excellent mechanical properties due to the low molecular
weight (LMW) fractions [9,10]. Therefore, many academic and
industrial research laboratories have engaged in the design and
synthesis of organometallic pre-catalysts of special structure with
the aim to obtain polyolefins possessing high-performance [11].
Recently, interests grow in developing new generation ‘‘non-
metallocene” catalysts, partly to avoid the growing patent mine-
field in group 4 cyclopentadienyl (Cp) systems, but most impor-
monomers [11]. Thereinto, b-diketiminate ligands attract consider-
able attention due to their isoelectronic relationship with cyclo-
pentadienyl anion, easily preparing from cheap and readily
available materials, as well as easy modulation of steric and elec-
tronic requirements by varying the amine moieties and the back-
bone [12–25]. Up to the present, a number of group 4 complexes
bearing b-diketiminate ligands have been synthesized and used
as pre-catalyst for olefin polymerization [17]. For example, in
1996 Lappert’s group [26] synthesized N-trimethylsilyl substituted
b-diketiminate zirconium and hafnium complexes, which showed
high activities in ethylene polymerization and moderate activities
in 1-hexene homopolymerization and copolymerization with eth-
ylene [27]. Then Collins [28] and Andres’s [29] groups synthesized
mono-, bis(b-diketiminate) complexes and monocyclopentadienyl
b-diketiminate mixed complexes of group 4 metals. It was found
that monocyclopentadienyl b-diketiminate zirconium complexes
with electron-withdrawing group in N-aryl moiety showed the
highest catalytic activity for ethylene polymerization [30]. What’s
more, Smith’s [31] and Mindiola’s [32] groups reported the
synthesis of some five- and six-coordinate group 4 b-diketiminate
complexes, where molecular structures in the solid state and in
solution, and intermolecular rearrangement reactions were further
investigated [33]. Besides, Berke et al. [34] found that volatile
* Corresponding authors. Tel./fax: +86 21 64253519.
E-mail addresses: haiyanma@ecust.edu.cn (H. Ma), qianling@online.sh.cn
(J. Huang).
0022-328X/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2008.08.024

3510
S. Gong et al. / Journal of Organometallic Chemistry 693 (2008) 3509–3518
b-diketiminate bis(dialkylamido)zirconium complexes could be
utilized as potential precursors in metal-organic chemical vapor
deposition processes (MOCVD). Novak [35], Liu [36] and Xie’s
[37] groups prepared zirconium, titanium trichloride complexes
with mono b-diketiminate ligand possessing different aromatic
or aliphatic substituents, and found that complexes of aliphatics-
substituted b-diketiminate ligand displayed higher catalytic
activity; fluoro-substituted titanium complex displayed higher
activities for ethylene/1-hexene copolymerization than for ethyl-
ene homopolymerization [37]. In order to further explore the olefin
polymerization behavior of b-diketiminate group 4 complexes, a
series of analogous Ti(III) complexes were synthesized by Bud-
zelaar’s [38] and Theopold’s [39] groups, which produced polyeth-
ylene with broad MWD (MWD = 28.6). Roesky’s [40] and
Mindiola’s [41] groups investigated the stability of b-diketiminate
Ti(III) complexes as well as their intermolecular rearrangement
reactions. More recently, Wu’s [42] and Li’s [43] groups prepared
the bulky bis(b-diketiminate)titanium complexes with trifluoro-
methyl-substituted backbone, which displayed moderate activities
for the homo- and copolymerization of ethylene and norbornene.
As we know, all the previously reported b-diketiminate group 4
complexes possessed symmetrical ligand framework. Enlightened
by Fujita’s FI catalysts [44] with bulky ortho-substitutent at phen-
oxy moiety capable of producing polyethylenes with uni-, bi-, and
trimodal MWD, we introduce bulky b-diketimines bearing two
different N-aryl groups to zirconium metal and conceive that
potential multiple metal species resulting from the versatile
coordination modes of the unsymmetrical ligand could lead to
the produce of polyethylenes with broad or multi-model molecular
weight distribution. Here we report the synthesis of a series of
novel zirconium complexes possessing symmetrical or unsymmet-
rical b-diketiminate ligands; the ethylene polymerization behavior
of these complexes is also studied. To the best of our knowledge,
it’s the first time that the unsymmetrical b-diketiminate ligands
are introduced to zirconium complexes.
R1
R2
R2
R2
R1
R1
N
N
Cl
Cl
(1)
n-BuLi
Zr
NH
N
2THF
(2) 0.5 ZrCl4
2
R2
1a - f
R1
1a: R1 = H, R2 = H
1b: R1 = CH3, R2 = H
1c: R1 = Cl, R2 = H
1d: R1 = F, R2 = H
1e: R1 = H, R2 = F
1f: R1 = CF3, R2 = H
2a - d, 2f
2a: R1 = H, R2 = H
2b: R1 = CH3, R2 = H
2c: R1 = Cl, R2 = H
2d: R1 = F,
R2 = H
2f: R1 = CF3, R2 = H
Scheme 1. The synthetic route of complexes 2a–d and f.
Pri
R2
iPr
R3
N
R1
iPr
N
N
Cl
Cl
(1)
n-BuLi
Zr
R3
NH
2THF
(2) 0.5 ZrCl4
iPr
R 3
2
R3
R2
1g - j
1g: R1 = H, R2 = H, R3 = H
1h: R1 = iPr, R2 = H, R3 = H
1i: R1 = H, R2 = F, R3 = H
1j: R1 = H, R2 = CF3, R3 = H
1k: R1 = H, R2 = H, R3 = CH3
R1
2g - i
2g: R1 = H, R2 = H, R3 = H
2h: R1 = iPr, R2 = H, R3 = H
2i: R1 = H, R2 = F, R3 = H
2. Results and discussion
Scheme 2. The synthetic route of complexes 2g–i.
2.1. Synthesis of b-diketiminate ligands and b-diketiminate zirconium
complexes
and coworkers ever reported [46] that the unsolvate [{(2,6-
Pri2H3C6)N(CH3)C}2CH]Li obtained from toluene existed as dimer or
dodecamer; besides two N-coordination from one b-diketiminate,
each Li+ ion was associated by coordinating to one N-aryl carbon
in another b-diketiminate ligand as evidenced by the X-ray diffrac-
tion measurement. On the other hand, the etherate [{(2,6-
Pri2H3C6)N(CH3)C}2CH]Li(Et2O) crystallized as monomers featuring
the Li+ ion in a distorted trigonal planar geometry [46]. It is conceiv-
able for us that the introduction of strong electron-withdrawing
group CF3 to the para-position of the phenyl ring weakened such
association interaction in toluene and resulted in the decomposition
of the lithium salt of 1f. Due to the coordination effect of solvent, the
etherate [{(4-CF3H4C6)N(CH3)C}2CH]Li(Et2O) could exist stably in
diethyl ether, therefore the desired complex 2f was afforded.
Analytically pure samples of zirconium complexes containing
meta-fluoro substituted b-diketimine 1e and meta-trifluoromethyl
substituted 1j could not be isolated using the similar procedure as
that of 2a due to the poor solubility in hot toluene or thermal insta-
bility. With the consideration that variation of the steric bulkiness
of ligand may lead to profound changes in the performance of cat-
alyst and the microstructure of produced polymers, we attempted
to synthesize zirconium complex bearing 1k. However, no reaction
could be observed between the lithium salt of 1k and ZrCl4 Á 2THF
or ZrCl4, most probably attributed to the steric hindrance [31].
Although similar b-diketiminate titanium complexes were
prepared using CH2Cl2 as extracting solvent [37,42,43], these
In order to investigate in detail the steric and electronic effect of
b-diketiminate ligand on the catalytic behaviors of the correspond-
ing complex, a series of ligands 1a–k [18,19,45] containing differ-
ent ortho-, meta-, and para-substituted N-aryl groups were utilized
to synthesize the zirconium complexes. Complexes 2a [28] and 2b
[31] were prepared according to the reported procedures in the
literature. As depicted in Schemes 1 and 2, zirconium complexes
2c–d and g–i bearing symmetrical or unsymmetrical auxiliary li-
gands were prepared in moderate yields via the reaction of
ZrCl4 Á 2THF and two equivalents of lithium salt of the correspond-
ing ligand in toluene [31]. These complexes with satisfactory ele-
mental analysis were obtained as yellow to orange yellow
crystalline solids after recrystallization from hot toluene. The same
procedure was adopted to prepare complex 2f, and a brown yellow
solid was isolated. The 1H NMR spectrum however showed an un-
known compound instead of the target complex. When diethyl
ether instead of toluene was used as reaction solvent, target com-
plex 2f was isolated successfully as orange yellow crystals and
well-characterized by 1H NMR and elemental analysis. In order to
explore the reason, the lithium salt of 1f prepared in toluene was
hydrolyzed. To our surprise, only the para-trifluoromethyl aniline
and some unknown compounds were detected in the hydrolyzed
mixture. On the contrary, hydrolysis of the lithium salt of 1f
prepared in diethyl ether afforded ligand 1f in high yield. Philip

S. Gong et al. / Journal of Organometallic Chemistry 693 (2008) 3509–3518
3511
zirconium complexes are quite unstable in polar solvents such as
CH2Cl2, CH3CN, and pyridine. In view of the poor solubility of them
in toluene or benzene at room temperature, CDCl3 was utilized in
the 1H NMR measurement despite of the fractional decomposition
of the complexes. 1H NMR data of complexes 2a–d, f at room tem-
perature indicate equivalent methine (c-H) environments in both
ligands, which resonate at ca. 5.4 ppm as a single peak. For com-
plexes 2g–i containing unsymmetrical auxiliary ligand, only one
set of signals for both ligands was shown in the 1H NMR spectra;
wherein two single peaks were observed for the backbone methyls
and one sharp resonance for the methine proton (c-H) at ca.
5.8 ppm, indicating the absence of other isomers at least in solu-
tion on the NMR time scale at room temperature. In comparison
with those of complexes 2a–d and f, the chemical shifts of methine
protons (c-H) in complexes 2g–i move to lower field, which
probably results from the partial gn character of the chelating
N–C–C–C–N ring indicated by the significant deviation of zirco-
nium atom from the ligand backbone plane (vide post).
2.2. Crystal structure of zirconium complexes 2d and 2g
Single crystals of zirconium complexes 2d and 2g suitable for X-
ray diffraction measurement were obtained by slowly cooling a
saturated toluene solution to 0 °C. The selected bond lengths and
angles are summarized in Table 1. The molecular structures of 2d
and 2g are shown in Figs. 1 and 2, respectively. In the solid state,
complex 2d adopts distorted octahedral geometry around the
zirconium center with two cis-chlorine atoms at an angle of
91.08(4)°, in which the N–Zr–N angles involving single diketimi-
nate ligand deviate significantly from 90° (N(1)–Zr–N(2) =
79.92(10)°, N(3)–Zr–N(4) = 80.25(11)°). The shorter Zr–N(3) bond
distance of 2.187(3) Å in comparison with Zr–N(4) = 2.259(3) Å
shows the trans influence of the chlorine atom [28,31]. However,
the almost same bond distances of Zr–N(1) (2.238(3) Å) and Zr–
N(2) (2.236(3) Å) indicate the absence of the expected trans influ-
ence of the chlorine. More obvious difference of two backbone C–C
bond distances (C(1)–C(2) = 1.382(5), C(2)–C(3) = 1.399(5) Å) and
of two N–C bond distances (N(1)–C(1) = 1.349(4), N(2)–
C(3) = 1.327(4) Å) in comparison with those of 2a [28] and 2b
[31] indicates an inferior delocalization effect, probably due to
the introduction of the para-fluorine atoms similar as the reported
fluoro-substituted mono(diiminato)titanium trichloride complex
[37]. The ligand backbone in 2d is non-planar with C(2) deviating
from the least-squares plane (defined by N(1), N(2), C(1), and
C(3)) by 0.1577 Å. The zirconium atom deviates from this plane
by 0.7180 Å.
Fig. 1. ORTEP diagram of the molecular structure of 2d. Thermal ellipsoids are
drawn at the 50% probability level. Hydrogen atoms are omitted for clarity.
Similarly, in the solid state, complex 2g also shows a distorted-
octahedral geometry at the metal center with two cis-chlorine
Table 1
Selected bond distances (Å) and angles (°) for complexes 2d and 2g
Complex 2d
Zr–N(1)
Zr–N(3)
Zr–Cl(1)
N(1)–C(1)
C(1)–C(2)
N(1)–Zr–N(2)
N(2)–Zr–N(3)
2.238(3)
2.187(3)
2.4247(10)
1.349(4)
1.382(5)
79.92(10)
164.48(11)
Zr–N(2)
Zr–N(4)
Zr–Cl(2)
N(2)–C(3)
C(2)–C(3)
N(3)–Zr–N(4)
Cl(1)–Zr–Cl(2)
2.236(3)
2.259(3)
2.4176(10)
1.327(4)
1.399(5)
80.25(11)
91.08(4)
Fig. 2. ORTEP diagram of the molecular structure of 2g. Thermal ellipsoids are
drawn at the 50% probability level. Hydrogen atoms are omitted for clarity.
Complex 2g
Zr–N(1)
Zr–N(3)
2.229(4)
2.229(3)
2.4155(12)
1.335(6)
1.401(6)
81.16(13)
168.76(14)
Zr–N(2)
Zr–N(4)
Zr–Cl(2)
N(2)–C(3)
C(2)–C(3)
N(3)–Zr–N(4)
Cl(1)–Zr–Cl(2)
2.272(4)
2.259(4)
2.3983(15)
1.334(5)
1.395(5)
80.57(14)
89.26(5)
atoms at an angle of 89.26(5)°, which is comparable to those in
2a [28], 2b [31], and 2d (90.0(1)°, 90.13(3)°, and 91.08(4)°, respec-
tively) and slightly larger than that of the titanium congener
(87.88(19)°) [43]. The average Zr–Cl bond distance of 2.4069 Å is
slightly shorter than those observed in 2a, 2b and 2d (2.4325,
2.4371 and 2.4221 Å, respectively). The angles of N(1)–Zr–N(2) =
Zr–Cl(1)
N(1)–C(1)
C(1)–C(2)
N(1)–Zr–N(2)
N(2)–Zr–N(4)

3512
S. Gong et al. / Journal of Organometallic Chemistry 693 (2008) 3509–3518
81.16(13)° and N(3)–Zr–N(4) = 80.57(14)° deviate considerably
from 90°. Although N(1) and N(3) locate at the trans-positions of
two cis-chlorine atoms, Zr–N(2) and Zr–N(4) bond distances of
2.259(4) and 2.272(4) Å are slightly longer than Zr–N(1)
(2.229(4)) and Zr–N(3) (2.229(3) Å), respectively, most likely in-
duced by the bulkier steric hindrance of the two ortho-isopropyl
groups at the same phenyl ring. The much larger angle of N(2)–
Zr–N(4) (168.76(14)°) in comparison with those in the previously
reported 2a [28] and 2b [31] (N(1)–Zr–N(3) = 160.4(1)°, N(2)–Zr–
N(3) = 157.65(6)°, respectively) as well as that in 2d (N(2)–Zr–
N(3) = 164.48(11)°) can also be attributed to the introduction of
two bulky ortho-isopropyls. The very close bond lengths of N(1)–
C(1) (1.335(6) Å) and N(2)–C(3) (1.334(5) Å), as well as C(1)–C(2)
(1.401(6) Å) and C(2)–C(3) (1.395(5) Å), indicate the multiple bond
character and significant delocalization in these bonds. The zirco-
nium atom in 2g situated 0.9753 Å out of the ligand plane defined
by N(1), C(1), C(2), C(3) and N(2), which is bigger than that ob-
served in complex 2d (vide supra). Therefore it is reasonable to con-
sider that the more bulky of the ortho-substituents, the larger the
deviation of zirconium center from the backbone plane will be
induced.
(Fig. 3), complex 2f exhibited the highest catalytic activity of
7.62 Â 105 g PE/mol-Zr Á h among them. These results indicated
that the CF3 substituents in 2f improved significantly the catalytic
activity, probably due to the increase of electrophilicity of metal
center and so accelerating the coordination/insertion rate of ethyl-
ene monomer. This is accordant with the results of previously re-
ported b-diketiminate group 4 complexes [28,42,43]. The lower
activity of 2b than 2a indicated that increase of the electron-donat-
ing ability of the para-substituent was disadvantaged to the cata-
lytic activity [43]. The electron-donating groups at the phenyl
rings decrease the electrophilicity of the zirconium center through
the chelating p-system of the ancillary ligand, obviously unfavor-
able for the coordination and insertion of the ethylene monomer.
Based on this point of view, complexes 2c and 2d bearing elec-
tron-withdrawing para-chloro or fluoro substituent should display
increased catalytic activity. Unexpectedly, the halogen atoms
substituted 2c and 2d showed lower catalytic activities than the
unsubstituted 2a, and complex 2c was slightly more active than
2d. The influence of halogen substitution at the ancillary ligands
on the polymerization performance of the corresponding metal
catalysts for olefins has been investigated in some cases, but con-
flictive results are usually obtained. An increase of ethylene cata-
lytic activity was observed for the mono(b-diiminato) titanium
complex with ortho-fluoro substituent [37]. Yang et al. reported
that ansa-metallocence group 4 complexes with para-fluoro substi-
tuted at phenyl group displayed lower ethylene catalytic activity
than the unsubstituted congener [47]. It’s obvious that the effect
of halogen substitution on catalytic activity is significantly compli-
cated. Generally, halogen atoms are considered to show electro-
negative characteristic, but the lone pair in p-orbital of halogen
atom can also lead to an electron-donating conjugated effect via
2.3. Polymerization of ethylene by complexes 2a–d, f–i/MAO systems
b-Diketiminate zirconium complexes 2a–d, f–i were effective
for the polymerization of ethylene under different conditions with
excess methylaluminoxane (MAO) as cocatalyst. The polymeriza-
tion results are summarized in Tables 2 and 3. For complexes
2a–d and f with symmetrical b-diketiminate ligand, the electronic
nature of the para-substituents at the aromatic rings exerted great
influence on the polymerization of ethylene. The catalytic activity
for ethylene polymerization at 50 °C and 1.0 MPa increased in the
order of 2d (p-F) < 2c (p-Cl) < 2b (p-CH3) < 2a (p-H) < 2f (p-CF3)
p-
p bonding to its para- and ortho-positions at phenyl ring. There-
fore, it seems that the lower activities of complexes 2c and 2d than
Table 2
The polymerization of ethylene with 2a–d, f/MAO as catalytic systemsa
c
d
d
e
Run
Complex
[Al]/[Zr]
Temp (°C)
Yield PE (g)
Activityb 105
M
g
105 g molÀ1
Mw 105 g molÀ1
Mw/Mn
Tm (°C)
1
2
2a
1000
1000
1000
1000
2000
1000
1000
1000
2000
500
1000
1000
1000
2000
4000
1000
1000
1000
1000
2000
1000
1000
1000
1000
2000
4000
50
70
90
70
50
50
70
90
50
50
50
70
90
50
50
30
50
70
90
50
30
50
70
90
70
70
0.6353
0.4575
0.3625
0.1407
1.2250
0.5224
0.4178
0.0504
1.0153
0.0655
0.4604
0.3651
0.3525
0.8525
0.1330
0.2102
0.3514
0.2908
0.2425
0.3453
0.5925
1.1150
1.2801
0.7175
1.8151
0.4951
2.54
1.83
1.45
0.59
4.90
2.08
1.64
0.20
4.06
0.26
1.84
1.46
1.41
3.41
0.53
0.84
1.40
1.16
0.97
1.38
2.37
4.26
5.12
2.87
7.62
1.98
11.5
4.14
3
0.45
4d
5
0.89f
g
–
6
7
8
9
2b
2c
2.62
2.18
3.34
3.05
0.84
g
–
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
3.84
>9.70h
136.2
130.3
>3.00h
1.63
g
–
8.75
9.60
2d
2f
g
–
–
–
g
g
3.73
7.24
6.38
4.85
1.81
4.12
1.78
a
b
c
Conditions: in toluene, V = 25 mL, [Zr] = 2 Â 10À4 mol/L, Pethylene = 1.0 MPa, tp = 0.5 h.
g PE/mol-Zr Á h.
Molecular weights were determined by intrinsic viscosity measurements.
Mw and Mw/Mn were determined by GPC.
d
e
f
Melting temperature was determined by DSC.
Pethylene = 0.5 MPa.
Polymer sample could not be dissolved in decahydronaphthalene at 135 °C.
Obtained for decahydronaphthalene soluble fraction by intrinsic viscosity measurements.
g
h

S. Gong et al. / Journal of Organometallic Chemistry 693 (2008) 3509–3518
3513
Table 3
The polymerization of ethylene with 2g–i/MAO, (BDI-1j)2ZrCl2/LiCl/MAO as catalytic systemsa
Activityb105
M
c105 g molÀ1
g
Mwd105 g molÀ1
Mw/Mn
Tm (°C)
d
e
Run
Complex
[Al]/[Zr]
Temp (°C)
Yield PE (g)
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
2g
1000
1000
1000
1000
2000
1000
1000
1000
2000
500
1000
1000
1000
2000
4000
1000
1000
1000
2000
30
50
70
90
50
50
70
90
50
50
50
70
90
50
50
50
70
90
50
0.1244
0.4405
0.3451
0.1350
0.6725
0.3650
0.2752
0.2608
0.6011
Trace
0.50
1.76
1.38
0.54
2.69
1.46
1.10
1.04
2.40
17.7
17.1
6.45
19.2
14.8
1.48
6.23
f
2h
2i
–
–
137.3
f
2.00
f
–
0.5203
0.4102
0.2725
0.7875
0.3923
0.5902
0.3804
0.1628
0.8900
2.08
1.64
1.09
3.15
1.57
2.36
1.52
0.65
3.56
6.67
>1.89g
1.14
0.94
0.95
10.6
2.37
0.77
6.75
(BDI-1j)2ZrCl2/LiClh
a
b
c
Conditions: in toluene, V = 25 mL, [Zr] = 2 Â 10À4 mol/L, Pethylene = 1.0 MPa, tp = 0.5 h.
g PE/mol-Zr Á h.
Molecular weights were determined by intrinsic viscosity measurements.
Mw and Mw/Mn were determined by GPC.
d
e
f
Melting temperature was determined by DSC.
Polymer sample could not be dissolved in decahydronaphthalene at 135 °C.
Obtained for decahydronaphthalene soluble fraction by intrinsic viscosity measurements.
g
h
The quantity used to carry out the polymerization Runs of (BDI-1j)2ZrCl2 was dependent on the results of elemental analysis.
Fig. 3. Effect of different substitutents of zirconium complexes 2aÀd and f on the
ethylene polymerization catalytic activity (polymerization conditions: in toluene,
V = 25 mL, Tp = 50 °C, [Zr] = 2 Â 10À4 mol/L, Pethylene = 1.0 MPa, tp = 0.5 h, [Al]/
[Zr] = 1000).
Fig. 4. Effect of different substitutents of zirconium complexes 2gÀi and (BDI-
1j)2ZrCl2 on the ethylene polymerization catalytic activity (polymerization condi-
tions: in toluene, V = 25 mL, Tp = 50 °C, [Zr] = 2 Â 10À4 mol/L, Pethylene = 1.0 MPa,
tp = 0.5 h, [Al]/[Zr] = 1000).
the unsubstituted 2a might be attributed to the electron-donating
conjugated effect of fluorine and chlorine atoms.
suggested to be responsible for the lower activity of complex 2d,
the meta-fluoro atom should show dominantly electron-withdraw-
ing effect in 2g, which increased significantly the electrophilicity of
the metal center and led to an increase of catalytic activity than the
unsubstituted analogue. Complex (BDI-1j)2ZrCl2 exhibited the
highest catalytic activity among them due to the presence of strong
electron-withdrawing meta-trifluoromethyl group. When com-
pared with the complexes 2a–d and f, zirconium complexes 2g–i
and (BDI-1j)2ZrCl2 showed slightly lower ethylene catalytic activ-
ity. It’s evident that the bulky 2,6-diisopropyl groups at the aryl
rings blocked the coordination sphere of the metal center to some
extent and restrained the coordination/insertion of ethylene
monomer [35].
As shown in Fig. 4, a comparison of the ethylene polymerization
activities of complexes 2g–i and (BDI-1j)2ZrCl2/LiCl [48] under the
polymerization conditions of 50 °C and 1.0 MPa ethylene pressure
revealed that the catalytic activity increased in the order of 2h
(p-iPr) < 2g (p-H) < 2i (m-F) < (BDI-1j)2ZrCl2/LiCl (m-CF3). Despite
of the introduction of ortho-isopropyls, the influence of the
substituent at the other phenyl group on the catalytic activity
was similar to that observed for the complexes 2a–d and f. The
introduction of electron-donating isopropyl group to the para-
position of one of the aromatic rings in 2h led to a decrease of
activity. The meta-fluoro substituted complex 2i exhibited higher
activity than complex 2g. Although an electron-donating conju-
As listed in Tables 2 and 3, the polymerization temperature had
a remarkable effect on the catalytic behavior of these zirconium
gated effect by the para-fluoro substitution via p-
p bonding was

3514
S. Gong et al. / Journal of Organometallic Chemistry 693 (2008) 3509–3518
complexes. When the polymerization temperature was raised from
30 to 90 °C, these complexes displayed the highest catalytic activ-
ity at about 50 °C except for complex 2f, then the activity de-
creased gradually with the increase of temperature from 70 to
90 °C, which was most likely due to the thermal decomposition
of the active species. Complex 2f showed the highest catalytic
activity at 70 °C (Run 23). It should be noticed that complexes 2c,
2f and 2h could retain comparatively satisfactory activities even
at 90 °C (Runs 13, 24 and 34). Compared with the titanium ana-
logues [37,42,43], which reached the maximal values of catalytic
activities at about 25 °C within the studied temperature range of
0–50 °C, these zirconium catalysts exhibited more excellent ther-
mal stability. Sufficient activity at high polymerization tempera-
ture should be important especially with regard to the industrial
application aspects, because performing a solution polymerization
at high-temperature can reduce the viscosity of reaction mixture,
leading to better mass transportation and temperature control.
The effect of ethylene pressure on polymerization behavior of
these b-diketiminate zirconium complexes was also studied. For
complex 2a, the decrease of catalytic activity (from 1.83 to
0.59 Â 105 g PE/mol-Zr Á h) and molecular weight (from 4.14 to
0.89 Â 105 g/mol) of the produced polymers was significant (Run
2 versus 4) when the ethylene pressure decreased from 1.0 to
0.5 MPa. The lower ethylene pressure resulted in the reduction of
the monomer concentration in solution, which was the main rea-
son for the decrease in the catalytic activity and molecular weight
[47,49].
of the amount of MAO on the molecular weights of produced poly-
mers was also observed. In the cases of 2a, 2b and 2c, when the
[Al]/[Zr] ratio enhanced from 1000 to 2000 (Runs 5, 9 and 14),
the molecular weight of produced polyethylenes increased consid-
erably, which could not be determined by neither Mw using GPC
nor viscosity average molecular weight based on [g] due to the
rather poor solubility in o-dichlorobenzene or decahydronaphtha-
lene at 135 °C. The obtainment of ultra-high molecular weight
polyethylenes (UHMWPE) is therefore suggested [50,51].
The molecular weights of our polyethylene samples are consid-
erably higher than those afforded by other b-diketiminate group 4
complexes (Mw = 1ꢀ9 Â 105 g/mol) in published literatures under
similar conditions [28,37,42,43]. The production of UHMWPE
probably resulted from the significantly faster chain propagation
rate than b-hydride elimination or chain transfer reaction. These
polymers could represent some of those possessing highest molec-
ular weight encountered in homogenous olefin polymerization cat-
alysts, including the group 4 metallocene catalysts [52–55].
Though showing slightly lower activity than 2a–d and f, com-
plexes 2g–i exhibited different catalytic performance compared
with the smaller congeners. Complex 2g could produce polyethyl-
ene of high molecular weight (M : 17.1 Â 105 g/mol), which is even
g
higher than that afforded by 2a (M : 11.5 Â 105 g/mol) (Run 1 ver-
g
sus 28) under same conditions. It is noteworthy that complex 2h
also produced extremely high molecular weight polyethylene
(Runs 32, 33 and 35), which hardly dissolves in decahydronaphtha-
lene under the intrinsic viscosity measurement conditions. This in-
crease of molecular weight was possibly due to a decrease in the
rate of b-hydride elimination or chain transfer reaction, which
was a consequence of the increased steric congestion around the
active metal center.
The amount of cocatalyst MAO always had significant influence
on the catalytic performance of these complexes. In general, the
catalytic activity increased with the enhancement of [Al]/[Zr] ratio
from 1000 to 2000. Further increasing the [Al]/[Zr] ratio to 4000 led
to the decrease of catalytic activities of 2c, 2f and 2i (Runs 15, 26,
41). Probably the surface of active species was overlaid with the
excess MAO, which led to the inactivity of catalysts. At lower
[Al]/[Zr] ratio of 500, a sharp drop in activity was observed for
complexes 2c and 2i (Runs 10 and 36), the small amount of MAO
was not enough to activate the catalysts. Considerable influence
Complex 2b produced polymers with molecular weight distri-
bution (MWD) of 3.05 (Run 6), which was comparable to the pre-
vious report of 2a (MWD = 2.70) [28] as well as titanium analogues
[37,43] and indicated the single-site catalytic property under these
conditions. In addition, high-temperature 13C NMR analysis of the
polyethylene produced by 2b (Run 6) indicates that the polymer
Fig. 5. 13C NMR spectrum of polyethylene (sample of Run 6) prepared with complex 2b (100 MHz, 1,2-dichlorobenzene-d4, 100 °C).

S. Gong et al. / Journal of Organometallic Chemistry 693 (2008) 3509–3518
3515
N1
M
N1
M
N2
polymerization behavior of these complexes were investigated.
Experimental results showed that the substitutents at the N-aro-
matic rings displayed an important role in the catalytic perfor-
mance of these b-diketiminate zirconium complexes. Para-
trifluoromethyl substituted 2f showed the highest catalytic activity
among them due to the strong electron-withdrawing effect of CF3.
UHMWPEs could be produced by 2a–d possibly ascribed to the
considerably faster rate of chain propagation than b-hydride elim-
ination or chain transfer reaction. Complexes 2g–i and (BDI-
1j)2ZrCl2/LiCl could afford polyethylenes with broad MWD (14.8),
which most likely resulted from the introduction of bulkily unsym-
metrical b-diketiminate ligands. It was noteworthy that these zir-
conium complexes were capable of controlling the molecular
Cl
N2
Cl
Cl
N2
N1
M
Cl
N2
Cl
Cl
N1
N1
N1
N2
N2
cis-
cis-
cis-
M = Zr
Ar1
N1
N2
Cl
M
Cl
M
N1
N2
N2
N1
N1
N1
N1
N2
N2
N2
Ar2
Cl
Cl
Ar1 = bulky group
Ar2 = less bulky group
trans-
trans-
weights of resulted polymers (M : 0.45–17.7 Â 105 or even higher)
g
as well as the MWD of obtained polyethylenes from 3.05 to 14.8 by
changing the ligand structure and the polymerization conditions.
Scheme 3. The possible isomers of metal complex with general formula of
[N1,N2]2MCl2.
possesses linear structure with virtually no branching (Fig. 5).
Melting temperature measured by DSC also proves the polymer
as typical linear polyethylene (Runs 11 and 14).
4. Experimental
4.1. General
It was expected that complexes 2g–i bearing unsymmetrical li-
gand structure with the bulky ortho-isopropyls at one of the two
phenyls displayed different catalytic property from 2a–d and f.
Complex 2g produced polyethylenes with very broad MWD
(14.8) (Run 28). Although the X-ray diffraction measurement
(Fig. 2) and 1H NMR data suggested the unique structure of 2g
where it adopted the configuration of cis-I with two N2 in the
cis-positions and two N1 in the trans-positions in order to reduce
the steric congestion of the ligand (Scheme 3), the formation of
polyethylenes with broad MWD indicated the existence of multi
active species under the polymerization conditions, which was also
suggested by the authors of previously reported group 4 metal
diiminato complexes producing broad MWD polyethylenes
[30,37,39]. As depicted in Scheme 3, chelating complexes of the
form [N1,N2]2MCl2 potentially possess five isomers besides the
common cis-I isomer raised from the versatile coordination modes
of ligands in an octahederal geometry. Minor isomers are some-
times observable accompanied with the major cis-N2, trans-N1,
cis-Cl isomer by NMR spectroscopy, even though single crystals
of cis-N2, trans-N1, cis-Cl isomer characterized by X-ray diffraction
were used for the NMR experiment [44]. Li’s group [36] revealed
the existence of isomers of b-diketiminate titanium complex by
19F NMR spectroscopy. Recently Fujita and collaborators [44] found
that bis(phenoxy-imine)zirconium complex with bulky substituent
at the phenoxy ortho-position possessed at least two isomers by
15N NMR, and this complex could afford polyethylenes with multi-
modal MWD (2–24). Therefore, it is possible that the highly flux-
ional character of the complex is connected with the multimodal
MWD of polymers. Here, we tentatively regarded that the produc-
tion of broad MWD polyethylene was relevant to the introduction
of bulk unsymmetrical auxiliary ligand to the center metal, which
promoted the formation of multi active species under the polymer-
ization conditions. Furthermore, the 13C NMR spectrum reveals
that similar to that afforded by complex 2b polyethylenes pro-
duced by 2g are linear. The DSC result of polymer sample by com-
plex 2i also shows the character of linear polyethylene (Run 32).
All manipulations were carried out under a dry argon atmo-
sphere using standard Schlenk techniques or a glove-box unless
otherwise indicated. Toluene and diethyl ether was refluxed and
distilled over sodium benzophenone ketyl prior to use. Chloro-
form-d and 1-hexene were dried over calcium hydride under argon
prior to use. n-BuLi (2.5 M in n-hexane) were purchased from
Chemetall. Methylaluminoxane (MAO) of 1.53 M in toluene was
purchased from Witco. 1H NMR spectra were recorded on Bruker
AVANCE-500 spectrometers with CDCl3 as solvent. Chemical shifts
for 1H NMR spectra were referenced internally using the residual
solvent resonances and reported relative to tetramethylsilane
(TMS). Melting points were determined in sealed glass capillaries
under argon and reported without correction. Elemental analyses
were carried out on an EA-1106 type analyzer. ESI-MS spectra were
recorded on a Micromass LCT mass instrument. Polymeric grade
ethylene was directly used for polymerization without further
purification. ZrCl4 Á 2THF [56], b-diketimines 1c–g and k [45], com-
plexes 2a [28] and 2b [31] were synthesized according to the pub-
lished procedures. All other reagents were obtained from standard
commercial vendors and used as received.
13C NMR spectra of polymers were recorded on a Bruker
AVANCE-500 spectrometer in 1,2-dichlorobenzene-d4 at 100 °C.
The intrinsic viscosities of polyethylenes (PE) were measured with
an Ubbelohde viscometer in decahydronaphthalene at 135 °C. The
viscosity average molecular weights of PEs were calculated accord-
0.67
ing to the equation: [
g
] (dL/g) = 6.77 Â 10À4
M
[57]. The gel
g
permeation chromatography (GPC) performed on a Waters 150
ALC/GPC system in a 1,2-dichlorobenzene solution at 135 °C was
used to determine the weight-average molecular weights (Mw)
and the molecular weight distributions (Mw/Mn) of the polymers.
Differential scanning calorimetry (DSC) was performed on a Uni-
versal V2.3C TA instrument. The polymer sample was first equili-
brated at 0 °C and then heated to 200 °C at a rate of 10 °C/min to
remove thermal history, After being cooled to 0 °C at a rate of
10 °C/min, the second heating scan was run from 0 to 200 °C at a
rate of 10 °C/min, and the data are reported according to the
second heating scan.
3. Conclusions
4.2. Synthesis of ligands and complexes
A
series of zirconium complexes bearing symmetrical or
unsymmetrical b-diketiminate ligands were synthesized, among
which the octahedral geometric structure of complexes 2d and
2g were confirmed by X-ray diffraction study. Complexes 2g–i
probably represented the first examples that the zirconium com-
plexes bearing unsymmetrical b-diketiminate ligand. The ethylene
4.2.1. Synthesis of 2-(2,6-diisopropylphenyl)amino-4-(4-
isopropylphenyl)imino-2-pentene (1h)
The synthetic procedure was similar to that of b-diketimine
1g [45]. 9.510 g (50.00 mmol) of para-toluenesulfonic acid

3516
S. Gong et al. / Journal of Organometallic Chemistry 693 (2008) 3509–3518
monohydrate, 6.761 g (50.00 mmol) of 4-isopropylaniline, 12.97 g
(50.00 mmol) of 4-(2,6-diisopropylphenyl)amino-3-penten-2-one,
and 80 mL of toluene were combined in a round bottomed flask.
A Dean–Stark apparatus was attached and the mixture was heated
at reflux for 24 h to remove the water. The reaction mixture was
cooled to r.t. and all the volatiles were removed under reduced
pressure to give a yellow solid. The solid was treated with diethyl
ether (100 mL), water (100 mL) and sodium carbonate (10.60 g,
100 mmol), and the obtained mixture was kept stirring. After com-
plete dissolution, the aqueous phase was separated and extracted
with diethyl ether. The combined organic phase was dried over
MgSO4 and rotary evaporated to dryness under reduced pressure
to afford a yellow solid. Yellow crystals (12.80 g, 68%) were ob-
tained after recrystallization from methanol. M.p. 85–86 °C. 1H
NMR (500 MHz, CDCl3, 25 °C): d 1.13 (d, 3J = 6.8 Hz, 6H, –CH(CH3)2),
1.21 (d, 3J = 6.8 Hz, 6H, –CH(CH3)2), 1.22 (d, 3J = 6.8 Hz, 6H,
–CH(CH3)2), 1.69 (s, 3H, CH3), 2.06 (s, 3H, CH3), 2.86 (sept,
3J = 6.8 Hz, 1H, –CH(CH3)2), 2.98 (sept, 3J = 6.8 Hz, 2H, –CH(CH3)2),
4.2.4. Synthesis of complex 2c
To a solution of ligand 1c (0.798 g, 2.50 mmol) in 10 mL of tol-
uene was added dropwise n-BuLi (2.5 M, 1.0 mL, 2.5 mmol) in n-
hexane at À78 °C. The mixture was stirred and slowly allowed to
warm to ambient temperature. After being stirred for additional
12 h, the solution was added dropwise to a stirred suspension of
ZrCl4 Á 2THF (0.472 g, 1.25 mmol) in toluene at À78 °C. The reaction
mixture was allowed to warm to room temperature and stirred
overnight. Then the solvent was removed under reduced pressure
and the obtained yellow solid was extracted several times with
hot toluene. The filtrates were combined and was concentrated
to ca. 40 mL under vacuum and deposited at À20 °C overnight. Pre-
cipitate was isolated by filtration and dried in vacuum to give
0.448 g (45%) of yellow crystals. M.p. 180–185 °C (dec). 1H NMR
(500 MHz, CDCl3, 25 °C): d 1.65 (s, 12H, CH3), 5.42 (s, 2H, c-CH),
6.77–7.24 (m, 16H, Ar-H). ESI-MS (m/s): 798 (M+). Anal. Calc. for
C34H30N4Cl6Zr: C, 51.14; H, 3.79; N, 7.02. Found: C, 51.53; H,
3.91; N, 6.66%.
4.85 (s, 1H, c
-CH), 6.89 (d, 2H, 3J = 9.0 Hz, o-Ar-H), 7.07–7.13 (m,
5H, m-, p-Ar-H), 12.70 (br s, 1H, NH). 13C NMR (100 MHz, CDCl3,
25 °C): d 19.6 (CMe), 20.1 (CMe), 21.6 (CHMe2), 23.0 (CHMe2),
27.1 (CHMe2), 32.4 (CHMe2), 94.4 (CH), 121.6 (Ar-C), 121.8 (Ar-C),
123.0 (Ar-C), 126.0 (Ar-C), 139.3 (Ar-C), 140.4 (Ar-C), 142.1 (Ar-
C), 142.6 (Ar-C), 155.4 (NCMe), 162.2 (NCMe). Anal. Calc. for
C26H36N2: C, 82.93; H, 9.64; N, 7.44. Found: C, 82.94; H, 9.65; N,
7.40%.
4.2.5. Synthesis of complex 2d
Complex 2d was synthesized using the same procedure as 2c.
0.716 g (2.50 mmol) of b-diketimine 1d, 1.0 mL of n-BuLi (2.5 M,
2.50 mmol) in n-hexane, and 0.472 g (1.25 mmol) of ZrCl4 Á 2THF
were used to give 0.467 g (51%) of orange yellow crystals. M.p.
189–192 °C (dec). 1H NMR (500 MHz, CDCl3, 25 °C): d 1.65 (s,
12H, CH3), 5.43 (s, 2H, c-CH), 6.82–7.00 (m, 16H, Ar-H). ESI-MS
(m/s): 732 (M+). Anal. Calc. for C34H30N4Cl2F4Zr: C, 55.73; H, 4.13;
4.2.2. Synthesis of 2-(2,6-diisopropylphenyl)amino-4-(3-
N, 7.65. Found: C, 55.61; H, 4.15; N, 7.66%.
fluorophenyl)imino-2-pentene (1i)
b-Diketimine 1i was synthesized by the same procedure of 1h.
9.510 g (50.00 mmol) of para-toluenesulfonic acid monohydrate,
5.561 g (50.00 mmol) of 3-fluoroaniline and 12.97 g (50.00 mmol)
of 4-(2,6-diisopropylphenyl)amino-3-penten-2-one were used to
give 12.34 g (70%) of light yellow crystals. M.p. 90–91 °C. 1H
NMR (500 MHz, CDCl3, 25 °C): d 1.12 (d, 3J = 6.9 Hz, 6H, –CH(CH3)2),
1.20 (d, 3J = 6.9 Hz, 6H, –CH(CH3)2), 1.67 (s, 3H, CH3), 2.03 (s, 3H,
4.2.6. Synthesis of complex 2f
To a solution of ligand 1f (0.965 g, 2.50 mmol) in 10 mL of
diethyl ether was added dropwise n-BuLi (2.5 M, 1.0 mL,
2.50 mmol) in n-hexane at À78 °C. The mixture was stirred and
slowly allowed to warm to ambient temperature. After being stir-
red for additional 12 h, the solution was added dropwise to a stir-
red suspension of ZrCl4 Á 2THF (0.472 g, 1.25 mmol) in 10 mL of
diethyl ether at À78 °C. The reaction mixture was allowed to warm
to room temperature and stirred overnight. Then the mixture was
dried under reduced pressure and the obtained yellow solid was
extracted with 20 mL diethyl ether. The filtrate was concentrated
to ca. 6 mL under vacuum and deposited at À20 °C overnight. Pre-
cipitate was isolated by filtration and dried in vacuum to afford
0.757 g (64%) of orange yellow crystals. M.p. 135–140 °C (dec).
1H NMR (500 MHz, CDCl3, 25 °C): d 1.69 (s, 12H, CH3), 5.52 (s,
CH3), 3.00 (sept, 3J = 6.9 Hz, 2H, –CH(CH3)2), 4.87 (s, 1H,
c-CH),
6.58–6.69 (m, 3H, o-, p-Ar-H), 7.11–7.19 (m, 4H, m-, p-Ar-H),
12.54 (br s, 1H, NH). 13C NMR (100 MHz, CDCl3, 25 °C): d 19.1
(CMe), 19.8 (CMe), 21.4 (CHMe2), 23.2 (CHMe2), 27.2 (CHMe2),
94.8 (CH), 107.8 (Ar-C), 107.9 (Ar-C), 108.0 (Ar-C), 116.5 (Ar-C),
121.9 (Ar-C), 124.3 (Ar-C), 128.6 (Ar-C), 138.9 (Ar-C), 141.3 (Ar-C),
147.3 (Ar-C), 158.4 (Ar-C), 160.0 (Ar-C), 161.2 (NCMe), 163.2
(NCMe). Anal. Calc. for C23H29FN2: C, 78.37; H, 8.29; N, 7.95. Found:
C, 77.90; H, 8.36; N, 7.81%.
2H, c
-CH), 6.94–7.05 (m, 8H, o-Ar-H), 7.54 (d, 3J = 8.1 Hz, 8H, m-
Ar-H). ESI-MS (m/s): 932 (M+). Anal. Calc. for C38H30Cl2F12N4Zr: C,
4.2.3. Synthesis of 2-(2,6-diisopropylphenyl)amino-4-(3-
48.93; H, 3.24; N, 6.01. Found: C, 48.56; H, 3.47; N, 5.72%.
trifluoromethylphenyl)imino-2-pentene (1j)
b-Diketimine 1j was synthesized by the same procedure of 1h.
9.510 g (50.00 mmol) of para-toluenesulfonic acid monohydrate,
8.056 g (50.00 mmol) of 3-trifluoromethylaniline and 12.97 g
(50.00 mmol) of 4-(2,6-diisopropylphenyl)amino-3-penten-2-one
were used to give 14.69 g (73%) of yellow crystals. M.p. 85.5–
87.5 °C. 1H NMR (500 MHz, CDCl3, 25 °C): d 1.24 (d, 3J = 6.9 Hz,
6H, –CH(CH3)2), 1.33 (d, 3J = 6.9 Hz, 6H, –CH(CH3)2), 1.80 (s, 3H,
CH3), 2.12 (s, 3H, CH3), 3.15 (sept, 3J = 6.9 Hz, 2H, –CH(CH3)2),
4.2.7. Synthesis of complex 2g
Complex 2g was synthesized using the same procedure of
2c. 0.836 g (2.50 mmol) of b-diketimine 1g, 1.0 mL of n-BuLi
(2.5 M, 2.50 mmol) in n-hexane, and 0.472 g (1.25 mmol) of
ZrCl4 Á 2THF were used to give 0.571 g (55%) of orange yellow crys-
tals. M.p. 108–110 °C (dec). 1H NMR (500 MHz, CDCl3, 25 °C): d
0.70 (d, 3J = 6.5 Hz, 6H, –CH(CH3)2), 0.86 (d, 3J = 6.5 Hz, 6H,
–CH(CH3)2), 0.93 (d, 3J = 6.5 Hz, 6H, –CH(CH3)2), 1.27 (d, 3J
= 6.5 Hz, 6H, –CH(CH3)2), 1.89 (s, 6H, CH3), 1.91 (s, 6H, CH3), 3.00
5.02 (s, 1H, c
-CH), 7.13 (d, 3J = 7.8 Hz, 1H, o-Ar-H), 7.22 (s, 1H, o-
Ar-H), 7.24–7.30 (m, 3H, m-, p-Ar-H), 7.34 (d, 3J = 7.8 Hz, 1H, p-
Ar-H), 7.45 (t, 3J = 7.8 Hz, 1H, m-Ar-H), 12.58 (br s, 1H, NH). 13C
NMR (100 MHz, CDCl3, 25 °C): d 19.3 (CMe), 19.8 (CMe), 21.4
(CHMe2), 23.3 (CHMe2), 27.3 (CHMe2), 94.8 (CH), 117.4 (Ar-C),
(sept, 3J = 6.5 Hz, 4H, –CH(CH3)2), 5.86 (s, 2H,
c-CH), 6.92–7.28
(m, 16H, Ar-H). ESI-MS (m/s): 828 (M+). Anal. Calc. for
C46H58Cl2N4Zr Á 0.7(C7H8): C, 68.49; H, 7.17; N, 6.25. Found: C,
68.57; H, 7.71; N, 5.81%.
1
117.9 (Ar-C), 122.0 (Ar-C), 123.3 (q, JC-F = 270.3 Hz, CF3), 124.0
2
(Ar-C), 124.8 (Ar-C), 128.1 (Ar-C), 130.0 (q, JC-C-F = 32.2 Hz, Ar-C),
4.2.8. Synthesis of complex 2h
137.9 (Ar-C), 142.1 (Ar-C), 146.9 (Ar-C), 159.4 (NCMe), 159.9
(NCMe). Anal. Calc. for C24H29F3N2: C, 71.62; H, 7.26; N, 6.96.
Found: C, 71.82; H, 7.42; N, 6.83%.
Complex 2h was synthesized using the same procedure of 2c.
0.942 g (2.50 mmol) of b-diketimine 1h, 1.0 mL of n-BuLi (2.5 M,
2.50 mmol) in n-hexane, and 0.472 g (1.25 mmol) of ZrCl4 Á 2THF

S. Gong et al. / Journal of Organometallic Chemistry 693 (2008) 3509–3518
3517
were used to give 0.537 g (47%) of orange yellow crystals. M.p.
121.5–123.5 °C. 1H NMR (500 MHz, CDCl3, 25 °C): d 0.72 (d,
3J = 6.5 Hz, 6H, –CH(CH3)2), 0.84 (d, 3J = 6.5 Hz, 6H, –CH(CH3)2),
0.92 (d, 3J = 6.5 Hz, 6H, –CH(CH3)2), 1.28–1.32 (m, 18H, –CH(CH3)2),
1.86 (s, 6H, CH3), 1.93 (s, 6H, CH3), 2.91–3.00 (m, 6H, –CH(CH3)2),
warm to room temperature and stirred overnight. The mixture
was dried under reduced pressure and washed with toluene twice.
Then the obtained 0.916 g of yellow solid was dried in vacuum. 1H
NMR (500 MHz, CDCl3, 25 °C): d 0.80 (d, 3J = 6.4 Hz, 6H, –CH(CH3)2),
0.96 (d, 3J = 6.4 Hz, 6H, –CH(CH3)2), 1.02 (d, 3J = 6.4 Hz, 6H,
–CH(CH3)2), 1.43 (d, 3J = 6.4 Hz, 6H, –CH(CH3)2), 2.01 (s, 6H, CH3),
5.84 (s, 2H, c-CH), 6.97–7.18 (m, 14H, Ar-H). ESI-MS (m/s): 912
(M+). Anal. Calc. for C52H70Cl2N4Zr: C, 68.39; H, 7.73; N, 6.13.
Found: C, 67.91; H, 8.01; N, 5.58%.
2.03 (s, 6H, CH3), 3.13 (m, 4H, –CH(CH3)2), 6.04 (s, 2H, c-CH),
7.12–7.47 (m, 14H, Ar-H). Anal. Calc. for C48H56Cl2F6N4ZrÁ2LiCl: C,
54.91; H, 5.38; N, 5.34. Found: C, 50.51; H, 5.50; N, 4.82%.
4.2.9. Synthesis of complex 2i
Complex 2i was synthesized using the same procedure of 2c.
0.881 g (2.5 mmol) of b-diketimine 1i, 1.0 mL of n-BuLi (2.5 M,
2.5 mmol) in n-hexane, and 0.472 g (1.25 mmol) of ZrCl4 Á 2THF
were used to obtain 0.530 g (49%) of orange yellow crystals. M.p.
134–137 °C (dec). 1H NMR (500 MHz, CDCl3, 25 °C): d 0.77 (d,
3J = 6.5 Hz, 6H, –CH(CH3)2), 0.84 (d, 3J = 6.5 Hz, 6H, –CH(CH3)2),
0.95 (d, 3J = 6.5 Hz, 6H, –CH(CH3)2), 1.34 (d, 3J = 6.5 Hz, 6H,
–CH(CH3)2), 1.90 (s, 6H, CH3), 1.92 (s, 6H, CH3), 2.91 (sept,
4.3. X-ray diffraction measurements
The crystallographic data for complexes 2d and 2g were col-
lected on a Bruker AXSD8 diffractometer with graphite-monochro-
mated Mo K
20 °C using the
a
(k = 0.71073 Å) radiation. All data were collected at
-scan techniques. Details of the crystal data and
x
structure refinements are summarized in Table 4. The structures
of 2d and 2g were solved by direct methods and refined using Fou-
rier techniques. An absorption correction based on SADABS was ap-
plied [58]. All non-hydrogen atoms were refined by full-matrix
least-squares on F2 using the SHELXTL program package [59]. Hydro-
gen atoms were located and refined by the geometry method. The
cell refinement, data collection, and reduction were done by Bruker
SAINT [60]. The structure solution and refinement were performed
by SHELXTL-97 [61] and SHELXTL-97 [62], respectively.
3J = 6.5 Hz, 4H, –CH(CH3)2), 5.86 (s, 2H,
c-CH), 6.92–7.31 (m, 14H,
Ar-H). ESI-MS (m/s): 864 (M+). Anal. Calc. for C46H56Cl2F2N4Zr: C,
63.87; H, 6.52, N, 6.48. Found: C, 63.12; H, 6.63; N, 6.16%.
4.2.10. Generation of complex (BDI-1j)2ZrCl2
The mixture of complex (BDI-1j)2ZrCl2 and LiCl was prepared as
the followed procedure. To a solution of ligand 1j (1.006 g,
2.50 mmol) in 10 mL of toluene was added dropwise 1.0 mL of n-
BuLi (2.5 M, 2.50 mmol) in n-hexane at À78 °C. The mixture was
stirred and slowly allowed to warm to ambient temperature. After
being stirred for additional 12 h, the solution was added dropwise
to a stirred suspension of ZrCl4 Á 2THF (0.472 g, 1.25 mmol) in
10 mL of toluene at À78 °C. The reaction mixture was allowed to
4.4. Polymerization procedure
Ethylene polymerization was carried out in a 100 mL autoclave
equipped with a magnetic stirrer. The autoclave was heated at
100 °C under vacuum for 30 min and then cooled to the desired
temperature by immerging into a thermostatically heated bath,
then filled with ethylene. Proper amount of MAO solution and tol-
uene were added to the autoclave and was filled with ethylene for
15 min at the reaction temperature. After proper amount of tolu-
ene solution of zirconium complex was injected to the reactor, eth-
ylene at desired pressure was introduced to start the
polymerization. The reaction mixture was stirred vigorously for a
certain time and the ethylene pressure in the autoclave was slowly
vented. Then 10 mL of ethanol was added to terminate the poly-
merization. The resulting mixture was poured into 3% HCl in etha-
nol (50 mL). The polymer was collected by filtration, washed with
ethanol (30 mL Â 2), and then dried for 16 h in a vacuum oven at
60 °C to constant weight.
Table 4
Crystal data and structure refinement details for 2d and 2g
2d
2g
Empirical formula
Formula weight
Temp (K)
C34H30Cl2F4N4Zr
732.74
293(2)
C46H58Cl2N4Zr
829.08
293(2)
Wavelength (Å)
Crystal size (mm3)
Crystal system
Space group
a (Å)
0.71073
0.71073
0.400 Â 0.212 Â 0.056 0.311 Â 0.211 Â 0.047
Monoclinic
P2(1)/n
Triclinic
ꢀ
P1
12.5931(8)
15.9559(10)
17.1990(11)
90.00
8.8115(12)
12.4202(17)
21.341(3)
73.676
b (Å)
c (Å)
Acknowledgements
a
(°)
b (°)
111.3190(10)
90.00
3219.4(4)
4
82.271
78.469
2188.6(5)
2
1.258
This work is subsidized by the National Basic Research Program
of China (2005CB623801), National Natural Science Foundation of
China (NNSFC, 20604009, 20774027), the Program for New Cen-
tury Excellent Talents in University (for H. Ma, NCET-06-0413),
Shanghai Rising-Star Program (06QA14014) and the Scientific
Research Foundation for the Returned Overseas Chinese Scholars
(SRF for ROCS, SEM). All the financial supports are gratefully
acknowledged. The authors also thank the very kind donation of
a glove-box by the AvH foundation.
c
(°)
Volume (Å3)
Z
Calc. density
1.512
(Mg/m3)
Absorption coefficient (mmÀ1
F(000)
)
0.562
1488
0.408
872
h range for data collection (°)
limiting indices
1.74–27.00
À16 6 h 6 15,
À20 6 k 6 20,
À21 6 l 6 18
18748/6964
[R(int) = 0.0697]
1.00000 and 0.88551
6964/0/410
0.869
1.76–25.50
À6 6 h 6 10,
À15 6 k 6 15,
À25 6 l 6 25
11439/8016
[R(int) = 0.0698]
1.00000 and 0.77199
8016/0/490
0.879
Reflection
collected/unique
Max. and min. transmission
Data/restrains/parameters
Goodness-of-fit on F2
Appendix A. Supplementary material
CCDC 689400 and 689401 contain the supplementary crystallo-
graphic data for 2d and 2g. These data can be obtained free of
charge from the Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif. Supplementary data associ-
ated with this article can be found, in the online version, at
doi:10.1016/j.jorganchem.2008.08.024.
Final R indices [I > 2
r
(I)]
R1 = 0.0521,
wR2 = 0.0830
R1 = 0.0913,
wR2 = 0.0943
0.482 and –0.483
R1 = 0.0570,
wR2 = 0.0795
R1 = 0.1257,
wR2 = 0.0943
0.554 and –0.660
R Indices (all data)
Largest difference in peak/hole
(e ÅÀ3
)

3518
S. Gong et al. / Journal of Organometallic Chemistry 693 (2008) 3509–3518
[33] B. Qian, W.J. Scanlon, M.R. Smith, Oganometallics 18 (1999) 1693–1698.
References
[34] P.L. Franceschini, M. Morstein, H. Berke, H.W. Schmalle, Inorg. Chem. 42 (2003)
7273–7282.
[35] X. Jin, B.M. Novak, Macromolecules 33 (2000) 6205–6207.
[36] M. Zhou, S. Huang, L. Weng, W. Sun, D. Liu, J. Organomet. Chem. 665 (2003)
237–245.
[37] G. Xie, C. Qian, J. Polym. Sci. A: Polym. Chem. 46 (2008) 211–217.
[38] P.H. Budzelaar, B.V. Oorta, A.G. Orpenb, Eur. J. Inorg. Chem. (1998) 1485–
1494.
[39] W.K. Kim, M.J. Fevola, L.M. Liable-Sands, A.L. Rheingold, K.H. Theopold,
Organometallics 17 (1998) 4541–4543.
[40] G.B. Nikiforov, H.W. Roesky, J. Magull, T. Labahn, D. Vidovic, M. Noltemeyer, H.
Schmidt, N.S. Hosmane, Polyhedron 22 (2003) 2669–2681.
[41] B.C. Bailey, F. Basuli, J.C. Huffman, D.J. Mindiola, Organometallics 25 (2006)
3963–3968.
[42] Y. Li, H. Gao, Q. Wu, J. Polym. Sci. A: Polym. Chem. 46 (2008) 93–101.
[43] L. Tang, Y. Duan, X. Li, Y. Li, J. Organomet. Chem. 691 (2006) 2023–2031.
[44] Y. Tohi, H. Makio, S. Matsui, M. Onda, T. Fujita, Macromolecules 36 (2003) 523–
525.
[45] S. Gong, H. Ma, Dalton Trans. (2008) 3345–3357.
[46] M. Stender, R.J. Wright, B.E. Eichler, J. Prust, M.M. Olmstead, H.W. Roesky, P.P.
Power, J. Chem. Soc., Dalton Trans. (2001) 3465–3469.
[47] X. Yang, Y. Zhang, J. Huang, J. Mol, Catal. A: Chem. 250 (2006) 145–
152.
[48] The zirconium complex bearing meta-trifluoromethyl substituted 1j could not
be isolated as analytically pure sample according to the similar synthetic
procedure as 2c, so the mixture of complex (BDI-1j)2ZrCl2 and produced LiCl
was used directly to carry out the ethylene polymerization experiment.
[49] N. Suzuki, Y. Masubuchi, Y. Yamaguchi, T. Kase, T. Miyamoto, A. Horiuchi, T.
Mise, Macromolecules 33 (2000) 754–759.
[50] S. Ishii, J. Saito, S. Matsuura, Y. Suzuki, R. Furuyama, M. Mitani, T. Nakano, N.
Kashiwa, T. Fujita, Macromol. Rapid Commun. 23 (2002) 693–697.
[51] Y. Yoshida, S. Matsui, Y. Takagi, M. Mitani, T. Nakano, H. Tanaka, N. Kashiwa, T.
Fujita, Organometallics 20 (2001) 4793–4799.
[1] L.C. Wang, M.K. Harvey, J.C. Ng, U. Scheunemann, J. Power Sources 73 (1998)
74–77.
[2] J.C. Liu, US Patent 6635728, 2003.
[3] S. Rastogi, P. Lemstra, WO Patent 9835818, 1998.
[4] H. Knuuttila, P. Sormnnenm, WO Patent 9507305, 1995.
[5] P. Brant, J. Canich, US Patent 5444145, 1995.
[6] M.S. Weiser, M. Wesolek, R. Mulhaupt, J. Organomet. Chem. 691 (2006) 2945–
2952.
[7] Y.-X. Chen, T.J. Marks, Organometallics 16 (1997) 3649–3657.
[8] M.E. Bluhm, C. Folli, D. Pufky, M. Kroger, O. Walter, M. Doring, Organometallics
24 (2005) 4139–4152.
[9] J.P. George, M.B. Britovsek, V.C. Gibson, B.S. Kimberley, P.J. Maddox, S.
Mastroianni, S.J. McTavish, C. Redshaw, G.A. Solan, S. Stro1mberg, A.D.
Williams, J. Am. Chem. Soc. 121 (1999) 8728–8740.
[10] G. Baruzzi, C.J. Schwvierien, Eur Patent 0942011, 1999.
[11] V.C. Gibson, S.K. Spitzmesser, Chem. Rev. 103 (2003) 283–315.
[12] D.V. Vitanova, F. Hampel, K.C. Hultzsch, Dalton Trans. 5 (2005) 1565–1566.
[13] D.V. Vitanova, F. Hampel, K.C. Hultzsch, J. Organomet. Chem. 690 (2005) 5182–
5197.
[14] M.F. Pilz, C. Limberg, B. Ziemer, J. Org. Chem. 71 (2006) 4559–4564.
[15] T.J. Hebden, W.W. Brennessel, C.J. Flaschenriem, P.L. Holland, Dalton Trans.
(2006) 3855–3857.
[16] J. Vela, L.W. Zhu, C.J. Flaschenriem, W.W. Brennessel, R.J. Lachicotte, P.L.
Holland, Organometallics 26 (2007) 3416–3423.
[17] L. Merle, M.F. Lappert, J.R. Severn, Chem. Rev. 102 (2002) 3031–3065.
[18] A.P. Dove, V.C. Gibson, E.L. Marshall, J.P. White, D.J. Williams, Dalton Trans.
(2004) 570–578.
[19] J.E. Parks, R.H. Holm, Inorg. Chem. 7 (1968) 1408–1416.
[20] S.G. McGeachin, Can. J. Chem. 46 (1968) 1903–1911.
[21] C. William, J.C. Simon, K.C. Elaine, S.M. Francis, Angew. Chem. Int. Ed. 37 (1998)
796–798.
[22] Y. Yao, M. Xue, Y. Luo, Z. Zhang, R. Jiao, Y. Zhang, Q. Shen, W. Wong, K. Yu, J.
Sun, J. Organomet. Chem. 678 (2003) 108–116.
[23] K.H. Park, W.J. Marshall, J. Org. Chem. 70 (2005) 2075–2081.
[24] H. Hamaki, N. Takeda, T.Y. amasaki, T. Sasamori, N. Tokitoh, J. Organomet.
Chem. 692 (2007) 44–54.
[25] S. Fustero, M.G. Torre, B. Pina, A.S. Fuentes, J. Org. Chem. 64 (1999) 5551–5556.
[26] B. Deelman, P.B. Hitchcock, M.F. Lappert, W. Leung, H. Lee, T. Mak, J.
Organomet. Chem. 513 (1996) 281–285.
[27] B. Deelman, P.B. Hitchcock, M.F. Lappert, W. Leung, H. Lee, T. Mak,
Organometallics 18 (1999) 1444–1452.
[28] M. Rahim, N.J. Taylor, S. Xin, S. Collins, Organometallics 17 (1998) 1315–1323.
[29] R. Andrés, E.F. Jesús, J. Mata, J.C. Flores, R. Gómez, J. Organomet. Chem. 690
(2005) 939–940.
[52] J. Saito, M. Onda, S. Matsui, M. Mitani, R. Furuyama, H. Tanaka, T. Fujita,
Macromol. Rapid Commun. 23 (2002) 1118–1123.
[53] J. Saito, Y. Suzuki, H. Makio, H. Tanaka, M. Onda, T. Fujita, Macromolecules 39
(2006) 4023–4031.
[54] B. Rieger, C. Troll, J. Prouschen, Macromolecules 35 (2002) 5742–5743.
[55] M. Weiser, M. Wesolek, R. Mulhaupt, J. Organomet. Chem. 691 (2006) 2945–
2952.
[56] L.E. Manzer, Inorg. Synth. 21 (1982) 135–136.
[57] P. Francis, R. Cooke, J. Elliott, J. Polym. Sci. 31 (1958) 453–466.
[58] SADABS, Bruker Nonius area detector scaling and absorption correction-V2.05,
Bruker AXS Inc., Madison, WI, 1996.
[59] G.M. Sheldrick, SHELXTL 5.10 for windows NT, Structure Determination Software
Programs, Bruker Analytical X-ray Systems, Inc, Madison, WI, 1997.
[60] SAINT, Version 6.02; Bruker AXS Inc.: Madison, WI, 1999.
[61] G.M. Sheldrick, SHELXS-97, Program for the Solution of Crystal Structures,
University of Gottingen, Germany, 1990.
[30] R. Vollmerhaus, M. Rahim, R. Tomaszewski, S. Xin, N.J. Taylor, S. Collins,
Organometallics 19 (2000) 2161–2169.
[31] L. Kakaliou, W.J. Scanlon, B. Qian, S.W. Baek, M.R. Smith, Inorg. Chem. 38
(1999) 5964–5977.
[62] G.M. Sheldrick, SHELXL-97, Program for the Refinement of Crystal Structures,
University of Gottingen, Germany, 1997.
[32] F. Basuli, U.J. Kilgore, D. Brown, J.C. Huffman, D.J. Mindiola, Organometallics 23
(2004) 6166–6175.