Post-metallocenes: Catalytic perfomance of new bis(salicylaldiminato) zirconium complexes for ethylene polymerization

Article abstract of DOI:10.1246/cl.2000.554

New bis(salicylaldiminato) zirconium complexes were synthesized and investigated as ethylene polymerization catalysts. As a result, we demonstrated that high molecular weight (Mw: 71.6 × 10⁴) and super high activity (2096 kg-PE/mmol-Zr-h) were accomplished by changing the ligand structure.

Full text of DOI:10.1246/cl.2000.554

554
Chemistry Letters 2000
Post-Metallocenes: Catalytic Perfomance of New Bis(salicylaldiminato) Zirconium Complexes
for Ethylene Polymerization
Shigekazu Matsui, Makoto Mitani, Junji Saito, Naoto Matsukawa, Hidetsugu Tanaka, Takashi Nakano, and Terunori Fujita*
Material Science Laboratory, Mitsui Chemicals, Inc., Nagaura, Sodegaura, Chiba 299-0265
(Received January 24, 2000; CL-000072)
New bis(salicylaldiminato) zirconium complexes were syn-
thesized and investigated as ethylene polymerization catalysts.
As a result, we demonstrated that high molecular weight (Mw:
71.6 × 104) and super high activity (2096 kg-PE/mmol-Zr·h)
were accomplished by changing the ligand structure.
In recent years, olefin polymerization catalysts based on
well-defined transition metal complexes have been intensively
investigated for creating high performance catalysts. These cat-
alysts possess potential for controlling polymer structure while
displaying high polymerization activity by changing the ligand
bulkiness of the Rl substituents.
structure, as demonstrated by group 4 metallocenes.1
The increase in Mw values is rationalized as follows; name-
ly, introduced alkyl groups diminished chain transfer reaction, or
β-hydride elimination. Theoretical calculations supported this
rationalization.9 The stabilization energy (∆Eβ) as a result of β-
agostic interaction of the active species generated from complex
3 (model B; Figure 1 right) was smaller than that for complex 1
(model A; Figure 1 left), due to the steric repulsion of a β-hydro-
gen and an isopropyl group. Thus, the introduced isopropyl
group at the R1 position energetically destabilized the conforma-
tion of the polymer chain leading to β-hydride elimination.
Recently, a certain number of post-metallocenes, which dis-
play intriguing olefin polymerization properties, have been dis-
covered.2,3 It has been demonstrated that the catalytic perform-
ance of these complexes was controllable by changing the ligand
structure.3 We have already reported that post-metallocenes
based on group 4 transition metals, having salicylaldimine lig-
ands, exhibit high ethylene polymerization activity.4 As a contin-
uation of this work, we synthesized new bis(salicylaldiminato)
zirconium complexes and studied their catalytic performance for
ethylene polymerization using MAO as a cocatalyst.
Concerning the ligand structure of bis[N-(3-t-butylsalicyli-
dene)phenylaminato]zirconium(IV) dichloride (1),4b we changed
the substituent, especially changing the substituents R1 and R3,
both near the polymerization reaction center, and investigated
what happened to the catalytic performance (Scheme 1).5-7
Regarding polymerization activity, complex 3 displayed 58
kg-PE/mmol-Zr·h, which was higher than that of Cp2ZrCl2
under the same polymerization conditions.
Table 2 summarizes the catalytic performance of bis(sali-
cylaldiminato) zirconium complexes as a result of introducing
an alkyl group at the R2 position and changing the alkyl sub-
stituent at the R3 position.10
Table 1 summarizes the catalytic performance of bis(salicyl-
aldiminato) zirconium complexes as a result of introducing an
alkyl group at the Rl position.8 Complex 1 (R1 = H) furnished an
Mw value of 0.8 × 104, based on GPC measurement. The intro-
duction of a methyl group at the R1 position (complex 2) dramati-
cally boosted the Mw value to 23.0 × 104. Moreover, the intro-
duction of a larger alkyl group further enhanced the molecular
Introduction of a methyl group at the R2 position resulted
in no significant change with respect to catalytic performance
though it somewhat lowered polymerization activity. Complex
4, R2 = Me, afforded 331 kg-PE/mmol-Zr·h of activity with an
Mw value of 0.8 × 104 (Table 2, Entry 1, 2).
i
weight. Thus, complex 3 (R1 = Pr) provided an Mw value of
71.6 × 104, an Mw value comparable to that of Cp2ZrC12. These
results indicate that the molecular weight values depend on the
Copyright © 2000 The Chemical Society of Japan

Chemistry Letters 2000
555
Maddox, S. J. McTavish, G. A. Solan, A. P. White, and D. J.
Williams, Chem. Commun., 1998, 849.
a) S. Matsui, Y. Tohi, M. Mitani, J. Saito, H. Makio, H. Tanaka,
M. Nitabaru, T. Nakano, and T. Fujita, Chem. Lett., 1999, 1065.
b) S. Matsui, M. Mitani, J. Saito, Y. Tohi, H. Makio, H. Tanaka,
and T. Fujita, Chem. Lett., 1999, 1263.
General synthesis procedure: Treatment of an o-substituted phe-
nol derivative with paraformaldehyde in the presence of a base
produced a salicylaldehyde derivative in about 80% yield. This
compound reacted with aniline, via Schiff base condensation, to
afford a salicylaldimine ligand in approximately 90% yield.
Complexation of ZrC14 with 2 equiv lithium salt of the ligand
thus obtained furnished a corresponding zirconium complex in
around 70% yield.
4
5
6
7
General polymerization procedure: Flow of ethylene gas (100
L/h) was charged into toluene (250mL). To this solution, a
solution of a complex and MAO (produced by Albemarle) was
added at 25 °C; see Ref. 4.
Changing the alkyl substituent at the R3 position influenced
polymerization activity. Attachment of a methyl or an isopropyl
group, which were sterically smaller than a t-butyl group, at the
R3 position, dramatically decreased polymerization activity,
whereas it slightly decreased Mw, compared to complex 1.
Complex 5 (R3 = Me) and complex 6 (R3 = iPr) displayed activi-
ties of 0.4 kg-PE/mmol-Zr·h (Mw = 0.4 × 104) and 0.9 kg-
PE/mmol-Zr·h (Mw = 0.5 × 104), respectively (Table 2, Entry 3,
4). Alternatively, attachment of a cumyl group, being sterically
larger than a t-butyl group, at the R3 position, enhanced polymer-
ization activity whereas it slightly increased Mw, compared to
complex 4. Thus, complex 7 (R3 = cumyl) displayed an activity
of 2096 kg-PE/mmol-Zr·h (Mw = 2.6 × 104), (Table 2, Entry 5).
This activity value corresponds to a catalyst turn over frequency
(TOF) value of 7.5 × 107 /h/atm. These results suggest that activ-
ity values depend on the bulkiness of the substituent at the R3
position. Regarding this activity enhancement, our speculation is
as follows; (1) the sterically large substituent protected phenoxy
oxygen from the coordination of the cocatalyst, which will
reduce space for incoming ethylene to coordinate to the metal
center and insertion of the ethylene into the carbon metal bond,
and (2) the large substituent may effectively separate the cationic
species and the anionic cocatalyst, and the ion separation will
increase unsaturation degree of the active species.11
In summary, we have demonstrated that enhancing the cat-
alytic performance of bis [N-(3-t-butylsalicylidene)phenylami-
nato]zirconium(IV) dichloride (1) was feasible by changing the
ligand structure. Thus, complex 3 (R1 = iPr, R2 = H, R3 = tBu)
provided a high Mw value of 71.6 × 104 with 58 kg-PE/mmol-
Zr·h activity. Furthermore, complex 7 (R1 = H, R2 = Me, R3 =
cumyl) afforded an unprecedented activity, 2096 kg-PE/mmol-
Zr·h with an Mw value of 2.6 × 104, even under atmospheric
ethylene pressure. To the best of our knowledge, this is the
highest olefin polymerization activity to date.12
Spectral data of the complexes: Complex 1; see Ref. 4b.
Complex 2 (C36H40N2O2ZrCl2); 1H-NMR (CDCl3); δ 1.08–1.71
(m, 18H), 2.33–2.45 (m, 6H), 6.44–7.70 (m, 14H), 8.08–8.28
(m, 2H). Reasonable elemental analysis data was not obtained
since complex 2 was unstable and decomposed on standing.
1
FD-mass, 694 (M+). Complex 3 (C40H48N2O2ZrCl2); H-NMR
(CDCl3); δ 0.90–1.35 (m, 18H + 12H), 3.10–3.31 (m, 2H),
6.36–7.70 (m, 14H), 8.08–8.29 (m, 2H). Anal. Found; C, 63.57;
H, 6.41; N, 3.34; Zr, 11.89%. Calcd for C, 63.98; H, 6.44;
N, 3.73; Zr, 12.15%. FD-mass, 750 (M+). Complex 4
1
(C36H40N2O2ZrCl2); H-NMR (CDCl3), δ 1.22–1.61 (m, 18H),
2.21–2.36 (m, 6H), 6.78–7.45 (m, 14H), 7.90–8.11 (m, 2H).
Anal. Found; C, 61.87; H, 5.57; N, 3.81; Zr, 13.06%. Calcd for
C, 62.23; H, 5.80; N, 4.03; Zr, 13.13%. FD-mass, 694 (M+).
Complex 5 (C28H24N2O2ZrCl2); 1H-NMR (CDCl3); δ 1.50–2.50
(m, 6H), 6.50–7.62 (m, 16H), 8.00–8.12 (m, 2H). Anal. Found;
C, 58.90; H, 4.47; N, 4.74; Zr, 15.37%. Calcd for C, 57.72; H,
4.15; N, 4.81; Zr, 15.66 %. FD-mass, 580 (M+). Complex 6
1
(C32H32N2O2ZrCl2); H-NMR (CDCl3), δ 0.80–1.43 (m, 12H),
2.74–3.32 (m, 2H), 6.13–7.37 (m, 16H), 7.89–8.17 (m, 2H).
Anal. Found; C, 59.86; H, 4.88; N, 4.46; Zr, 14.12%. Calcd for
C, 60.17; H, 5.05; N, 4.39; Zr, 14.28%. FD-mass, 638 (M+).
Complex 7 (C46H44N2O2ZrCl2): 1H-NMR (CDCl3), δ 1.35–1.65
(m, 12H), 1.70–2.11 (m, 6H), 6.25–8.05 (m, 24H + 2H). Anal.
Found; C, 67.61; H, 5.49; N, 3.18; Zr, 10.79%. Caled for C,
67.46; H, 5.42; N, 3.42; Zr, 11.14%. FD-mass, 818 (M+).
The molecular weight distribution (Mw/Mn) for complexes 1–3
were as follows; complex 1; 2.06, complex 2; 2.13, complex 3;
2.61.
8
9
DFT calculation has been widely used for theoretical studies of
transition metal complexes, cf.; L. Deng, T. Zieglar, T. K. Woo,
P. Margl, and L. Fan, Organometallics, 17, 3240 (1998). All
calculation were performed at the gradient corrected density
functional BLYP level by Amsterdam Density Functional
(ADF) program; C. Fonseca Guerra, J. G. Snijders, G. te Velde,
and E. J. Baerends, Theor. Chem. Acc., 99, 391 (1998). We used
the triple ζ STO basis set on the Zr and the double ζ STO basis
n
set on the N, O and Pr as a model of a polymer chain, and the
single ζ STO basis set on the other atoms to calculate the opti-
mized geometries. For energy calculations, the triple ζ STO
basis set on the Zr and the double ζ plus polarization STO basis
set on the other atoms are used and the quasi-relativistic correc-
tion is also added.
References and Notes
1
Review; H. H. Brinzinger, D. Fischer, R. Mülhaupt, B. Rieger,
and R. M. Waymouth, Angew. Chem., Int. Ed. Engl., 34, 1143
(1995).
2
3
Review: G. J. P. Biritovsek, V. C. Gibson, and D. F. Wass,
Angew. Chem., Int. Ed. Engl., 38, 428 (1999).
10 The molecular weight distribution (Mw/Mn) for complexes 4-7
were as follows; complex 4; 2.00, complex 5; 2.31, complex 6;
2.48, complex 7; 7.20.
11 P. A. Deck, C. L. Beswick, and T. J. Marks, J. Am. Chem. Soc.,
120, 1772 (1998).
12 a) H. G. Alt, W. Milius, and S. J. Palackal, J. Organomet.
Chem., 472, 113 (1994). b) W. Spaleck, F. Ktiber, A. Winter, J.
Rohrmmann, B. Bachmann, M. Antberg, V. Dolle, and E. F.
Paulus, Organometallics, 13, 954 (1994).
For example; a) L. K. Johnson, C. M. Killian, and M. S.
Brookhart, J. Am. Chem. Soc., 117, 6414 (1995). b) C. M.
Killian, D. J. Temple, L. K. Johnson, and M. S. Brookhart, J.
Am. Chem. Soc., 118, 11664 (1996). c) C. Wang, A. Friedrich,
T. R. Younkin, R. T. Li, R. H. Grubbs, A. Bansleben, and M.
W. Day, Organometallics, 17, 3149 (1998). d) B.L. Small, M.
S. Brookhart, and A. A. Bennett, J. Am. Chem. Soc., 120, 4049
(1998). e) G. J. P. Britovsek, V. C. Gibson, B. S. Kimberley, J.