Imido titanium ethylene polymerization catalysts containing triazacyclic ligands

3888

Organometallics 2006, 25, 3888-3903

Imido Titanium Ethylene Polymerization Catalysts Containing

Triazacyclic Ligands

Nico Adams,^†Henricus J. Arts,^‡Paul D. Bolton,^†Dan Cowell,^§Stuart R. Dubberley,^†

Nic. Friederichs, Craig M. Grant,^§Mirko Kranenburg,^|Andrew J. Sealey,^†Bing Wang,

Paul J. Wilson,^#Martin Zuideveld,^‡Alexander J. Blake,^#Martin Schro¨der,^#and

Philip Mountford*^,†

Chemistry Research Laboratory, UniVersity of Oxford, Mansfield Road, Oxford OX1 3TA, U.K.,

DSM Research, P.O. Box 18, 6160 MD Geleen, The Netherlands, Millennium Pharmaceuticals Research

and DeVelopment Ltd., Granta Park, Great Abington, Cambridge CB1 6ET, U.K., SABIC Europe,

Research & DeVelopment, P.O. Box 319, 6160AH Geleen, The Netherlands, DSM Elastomers,

Global R&D, P.O. Box 1130, 6160 BC Geleen, The Netherlands, and School of Chemistry,

UniVersity of Nottingham, UniVersity Park, Nottingham NG7 2RD, U.K.

ReceiVed April 13, 2006

A comprehensive account of the synthesis, properties, and evaluation of a wide range of ethylene

homopolymerization catalysts derived from imido titanium compounds supported by the triazacyclic ligands

Me₃[9]aneN₃and R₃[6]aneN₃is described (Me₃[9]aneN₃) 1,4,7-trimethyltriazacyclononane; R₃[6]aneN₃

) 1,3,5-trimethyl- or 1,3,5-tris(n-dodecyl)triazacyclohexane). Conventional preparative-scale reactions

t

afforded the triazacycle-supported imido titanium compounds Ti(NR)(Me₃[9]aneN₃)Cl₂(R ) Bu (1),

i

2,6-C₆H₃Me₂, 2,6-C₆H₃Pr₂, Ph, C₆F₅, or CH₂Ph (6)). Solid phase-supported analogues of 1 and 6 (linked

by either the macrocycle or imido ligand to a 1% cross-linked polystyrene support) and representative

Me₃[6]aneN₃solution phase systems Ti(NR)(R₃[6]aneN₃)Cl₂(R ) Me or n-dodecyl) were also synthesized.

At ambient temperature, solution phase Me₃[9]aneN₃catalyst systems were more active for ethylene

polymerization (methyl aluminoxane (MAO) cocatalyst) than their solid phase-supported or Me₃[6]aneN₃

analogues. A library of 41 other triazacyclononane-supported catalysts was prepared by the semiautomated,

sequential treatment of Ti(NMe₂)₂Cl₂with RNH₂and Me₃[9]aneN₃. The ethylene polymerization cap-

abilities of 46 compounds of the type Ti(NR)(Me₃[9]aneN₃)Cl₂were evaluated at 100 °C (MAO cocatalyst)

and compared in representative cases to the corresponding productivities at ambient temperature. Whereas

either bulky N-alkyl or N-aryl imido substituents in the compounds Ti(NR)(Me₃[9]aneN₃)Cl₂were suffi-

cient to give highly active catalysts at ambient temperature, only those with bulky N-alkyl groups excelled

at 100 °C. Polymer end group analysis indicated that polymeryl chain transfer to both AlMe₃and ethylene

monomer is an active mechanism in these systems. The use of MAO pretreated with BHT-H (BHT-H )

2,6-di-tert-butyl-4-methylphenol) led to higher productivites, increased polymer molecular weights, and

more polymer chain unsaturations, but productivity decreased when a large excess of BHT-H was used.

The reactions of the well-defined alkyl species Ti(N^tBu)(Me₃[9]aneN₃)Me₂, [Ti(N^tBu)(Me₃[9]aneN₃)-

(µ-Me)₂AlMe₂]⁺, and [Ti(N^tBu)(Me₃[9]aneN₃)Me]⁺with BHT-H were examined, and the aryloxide

compound [Ti(N^tBu)(Me₃[9]aneN₃)(BHT)][BAr^F] was isolated (Ar^F) C₆F₅). The X-ray structures of

4

i

Ti(NR)(Me₃[9]aneN₃)Cl₂(R ) ^tBu, 2,6-C₆H₃Me₂, 2,6-C₆H₃Pr₂, Ph, C₆F₅) and Ti(NR)(Me₃[6]aneN₃)Cl₂

i

(R ) 2,6-C₆H₃Pr₂, Ph, C₆F₅) are reported. The perfluorophenyl imido titanium compounds both exhibit

well-defined supramolecular structures based on C‚‚‚F intermolecular interactions.

type L_nMdNR) to act as Ziegler-type olefin polymerization

catalysts. The first systems of this type were derived from bis-

(imido)chromium dichloride and dialkyl compounds Cr(NR)₂X₂

Introduction

Homogeneous olefin polymerization catalysts have been a

continuing focus of academic and industrial interest, and a

diverse number of systems have been investigated with the pur-

pose of optimizing catalyst performance and polymer prop-

erties.^1-23Of particular relevance to our present contribution is

the ability of certain transition metal imido compounds (general

t

i

(R ) Bu or 2,6-C₆H₃Pr₂; X ) Cl, Me, or CH₂Ph),^24,25and a

number of other group 4, 5, and 6 systems have since been

(2) Brintzinger, H. H.; Fischer, D.; Mu¨lhaupt, R.; Rieger, B.; Waymouth,

R. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1143.

(3) McKnight, A. L.; Waymouth, R. M. Chem. ReV. 1998, 98, 2587.

(4) Kaminsky, W. J. Chem. Soc., Dalton Trans. 1998, 1413.

(5) Okuda, J.; Eberle, T. In Metallocenes: synthesis, reactiVity, applica-

tions; Togni, A., Halterman, R. L., Eds.; Wiley-VCH: New York, 1998;

Vol. 1.

* To whom correspondence should be addressed. E-mail:

philip.mountford@chem.ox.ac.uk.

^†University of Oxford.

^‡DSM Research.

(6) Britovsek, G. J. P.; Gibson, V. C.; Wass, D. F. Angew. Chem., Int.

Ed. 1999, 38, 429.

^§Millennium Pharmaceuticals Research and Development Ltd.

SABIC Europe, Research & Development.

^|DSM Elastomers.

(7) Ittel, S. D.; Johnson, L. K.; Brookhart, M. Chem. ReV. 2000, 100,

1169.

^#University of Nottingham.

(8) Coates, G. W. J. Chem. Soc., Dalton Trans. 2002, 467.

(9) Hou, Z.; Wakatsuki, Y. Coord. Chem. ReV. 2002, 231, 1.

(1) Okuda, J. Angew. Chem., Int. Ed. Engl. 1992, 31, 47.

Publication on Web 07/01/2006

Imido Titanium Ethylene Polymerization Catalysts

Organometallics, Vol. 25, No. 16, 2006 3889

reported.²³A notable feature of group 6 bis(imido) and group

5 cyclopentadienyl-imido catalysts of the type M(NR)₂X₂and

CpM(NR)X₂, respectively, is their isolobal relationship²⁶(dis-

cussed and documented elsewhere^27-30) with the quintessential

group 4 metallocene-type systems Cp₂MX₂. As part of our broad

research program in titanium imido chemistry^31-33we have

targeted compounds of the type Ti(NR)(fac-L₃)Cl₂(fac-L₃)

triazacyclohexane,³⁴triazacyclononane,^35-37or related^35,38tris-

(3,5-dimethylpyrazolyl)methane^39,40) as potential polymerization

precatalysts. These systems are also isolobal analogues of group

4 metallocenes, but their properties and reactivity depend

significantly on the identity of the particular fac-L₃ligand and

imido N-substituent, R. Very recently, we described the

synthesis, properties, and DFT bonding analysis of well-defined

macrocycle-supported tert-butyl imido titanium monoalkyl

cations [Ti(N^tBu)(Me₃[9]aneN₃)R]⁺(R ) Me or CH₂SiMe₃;

Me₃[9]aneN₃) 1,4,7-trimethyltriazacyclononane), along with

stoichiometric C-H bond and solvent activation reactions,

insertion reactions, and addition product with AlMe₃.^41,42In this

contribution we describe in full our studies of the ethylene

Scheme 1. Synthetic Routes to Complexes of the Type

Ti(NR)(Me₃[9]aneN₃)Cl₂

(10) Murphy, V.; Bei, X.; Boussie, T. R.; Bru¨mmer, O.; Diamond, G.

M.; Goh, C.; Hall, K. A.; LaPointe, A. M.; Leclerc, M. K.; Longmire, J.

M.; Schoemaker, J. A. W.; Turner, H. W.; Weinberg, H. Chem. Rec. 2002,

2, 278.

(11) Hagen, H.; Boersma, J.; van Koten, G. Chem. Soc. ReV. 2002, 31,

357.

(12) Erker, G. Chem. Commun. 2003, 1469.

(13) Suzuki, Y.; Terao, H.; Fujita, T. Bull. Chem. Soc. Jpn. 2003, 76,

1493.

(14) Gambarotta, S. Coord. Chem. ReV. 2003, 237, 229.

(15) Gibson, V. C.; Spitzmesser, S. K. Chem. ReV. 2003, 103, 283.

(16) Leung, W.-H. Eur. J. Inorg. Chem. 2003, 583.

(17) Mitani, M.; Saito, J.; Ishii, S.; Nakayama, Y.; Makio, H.; Matsukawa,

N.; Matsui, S.; Mohri, J.; Furuyama, R.; Terao, H.; Bando, H.; H., T.; Fujita,

T. Chem. Rec. 2004, 4, 137.

(18) Bochmann, M. J. Organomet. Chem. 2004, 689, 3982.

(19) Gromada, J.; Carpentier, J.-F.; Mortreux, A. Coord. Chem. ReV.

2004, 248, 397.

(20) Nakayama, Y.; Yasuda, H. J. Organomet. Chem. 2004, 689, 4489.

(21) Stephan, D. W. Organometallics 2005, 24, 2548.

(22) Speiser, F.; Braunstein, P.; Saussine, L. Acc. Chem. Res. 2005, 38,

784.

(23) Bolton, P. D.; Mountford, P. AdV. Synth. Catal. 2005, 347, 355.

(24) Coles, M. P.; Dalby, C. I.; Gibson, V. C.; Clegg, W.; Elsegood, M.

R. J. J. Chem. Soc., Chem. Commun. 1995, 1709.

(25) Coles, M. P.; Dalby, C. I.; Gibson, V. C.; Little, I. R.; Marshall, E.

L.; Ribeiro da Costa, M. H.; Mastroianni, S. J. Organomet. Chem. 1999,

591, 78.

polymerization capabilities of two families of macrocycle-

supported Ti(NR)(fac-L₃)Cl₂systems. Part of this work has been

communicated.^35,43

Results and Discussion

Synthesis of Selected Precatalysts Ti(NR)(Me₃[9]aneN₃)-

Cl₂. The available synthetic routes to the compounds Ti(NR)-

(Me₃[9]aneN₃)Cl₂(1-6) are summarized in Scheme 1. Prelimi-

nary work focused on the tris(pyridine) complexes Ti(NR)-

Cl₂(py)₃(R ) ^tBu, 2,6-C₆H₃Me₂, 2,6-C₆H₃Pr₂) reported by us

i

previously⁴⁴and shown to be very versatile precursors to a range

of new imido titanium compounds by pyridine and/or chloride

ligand substitution.^31,33Reaction of these compounds with Me₃-

[9]aneN₃in CH₂Cl₂at room temperature gave deep orange- or

brown-colored solutions from which the target compounds Ti-

(26) Hoffmann, R. Angew. Chem., Int. Ed. Engl. 1992, 21, 711-800.

(27) Williams, D. S.; Schofield, M. H.; Anhaus, J. T.; Schrock, R. R. J.

Am. Chem. Soc. 1990, 112, 6728.

(28) Glueck, D. S.; Green, J. C.; Michelman, R. I.; Wright, I. N.

Organometallics 1992, 11, 4221.

t

(NR)(Me₃[9]aneN₃)Cl₂(R ) Bu (1), 2,6-C₆H₃Me₂(2), 2,6-

i

C₆H₃Pr₂(3)) were isolated in 85-87% yield. The analytical

and spectroscopic data for these compounds, each of which has

been structurally characterized, were consistent with the C_s

symmetric structures proposed in Scheme 1. Full details are

given in the Experimental Section. Encouraged by the prelimi-

nary ethylene polymerization capabilities of some of these

compounds (see below), we decided to target a wider range of

such complexes, with a particular emphasis on developing routes

suitable for varying the imido N-substituent using semiauto-

mated methods for the synthesis of a precatalyst library.

Although successful for the synthesis of compounds 1-3,

tris(pyridine) complexes of the general type Ti(NR)Cl₂(py)₃were

considered to be less suitable for the preparation of a wide range

(29) Williams, D. S.; Schofield, M. H.; Schrock, R. R. Organometallics

1993, 12, 4560.

(30) Gibson, V. C. J. Chem. Soc., Dalton Trans. 1994, 1607.

(31) Mountford, P. Chem. Commun. 1997, 2127.

(32) Gade, L. H.; Mountford, P. Coord. Chem. ReV. 2001, 216-217,

65.

(33) Hazari, N.; Mountford, P. Acc. Chem. Res. 2005, 38, 839.

(34) Wilson, P. J.; Blake, A. J.; Mountford, P.; Schro¨der, M. J.

Organomet. Chem. 2000, 600, 71.

(35) Wilson, P. J.; Blake, A. J.; Mountford, P.; Schro¨der, M. Chem.

Commun. 1998, 1007.

(36) Male, N. A. H.; Skinner, M. E. G.; Bylikin, S. Y.; Wilson, P. J.;

Mountford, P.; Schro¨der, M. Inorg. Chem. 2000, 39, 5483.

(37) Gardner, J. D.; Robson, D. A.; Rees, L. H.; Mountford, P. Inorg.

Chem. 2001, 40, 820.

(38) Wilson, P. J.; Blake, A. J.; Mountford, P.; Schro¨der, M. Inorg. Chim.

Acta 2003, 345, 44.

(39) Lawrence, S. C.; Skinner, M. E. G.; Green, J. C.; Mountford, P.

Chem. Commun. 2001, 705.

(40) Bigmore, H. R.; Dubberley, S. R.; Kranenburg, M.; Lawrence, S.

C.; Sealey, A. J.; Selby, J. D.; Zuideveld, M.; Cowley, A. R.; Mountford,

P. Chem. Commun. 2006, 436.

(42) Bolton, P. D.; Clot, E.; Adams, N.; Dubberley, S. R.; Cowley, A.

R.; Mountford, P. Organometallics 2006, 25, 2806.

(43) Adams, N.; Arts, H. J.; Bolton, P. D.; Cowell, D.; Dubberley, S.

R.; Friederichs, N.; Grant, C.; Kranenburg, M.; Sealey, A. J.; Wang, B.;

Wilson, P. J.; Cowley, A. R.; Mountford, P.; Schro¨der, M. Chem. Commun.

2004, 434.

(41) Bolton, P. D.; Clot, E.; Cowley, A. R.; Mountford, P. Chem.

Commun. 2005, 3313.

(44) Blake, A. J.; Collier, P. E.; Dunn, S. C.; Li, W.-S.; Mountford, P.;

Shishkin., O. V. J. Chem. Soc., Dalton Trans. 1997, 1549.

3890 Organometallics, Vol. 25, No. 16, 2006

Adams et al.

of triazacyclononane-supported precatalysts for the following

reasons. Only alkyl imido homologues (in particular tert-butyl

derivatives) are readily accessible from TiCl₄and an excess of

RNH₂since the side-products RNH₃Cl cannot be readily

separated when R is an aryl group.⁴⁴Although aryl imides

Ti(NAr)Cl₂(py)₃(Ar ) aryl) can be prepared from Ti(N^tBu)-

Cl₂(py)₃by treatment with ArNH₂, the products can be rather

insoluble except for bulky Ar groups (e.g., 2,6-disubstituted

phenyl rings) and also sometimes retain the displaced ^tBuNH₂,

thereby forcing further recrystallizations in the presence of an

excess of pyridine. The pyridine trans to TidNR is labile and

can be readily lost to form rather insoluble µ-imido-bridged

dimers Ti₂(µ-NR)₂Cl₄(py)₄.^44-46To avoid all of these issues and

to provide an alternative set of potential synthons in imido

titanium chemistry, we recently reported a family of monomeric,

five-coordinate compounds Ti(NR)Cl₂(NHMe₂)₂(R ) alkyl,

aryl).⁴⁷All of the 13 new compounds described (6 of which

were structurally characterized) had good solubilities, and each

was formed in excellent yield by the reaction of Ti(NMe₂)₂Cl₂

with the appropriate primary amine RNH₂. The solid phase-

supported synthon Ti(NCH₂PS)Cl₂(NHMe₂)₂(“PS” represents

1% cross-linked polystyrene resin) was also prepared.

(Ti(1)-N(1)-C(10) ) 166.3(2)-171.22(12)°) suggest that these

groups are acting as four-electron donors to the metal.^50,51

The Ti(1)-N(9) distances are all consistently longer than

Ti(1)-N(3) and Ti(1)-N(6) due to the strong trans influence

of imido ligands. The five structures form a useful homologous

series for the comparison of the ground-state structural effects

of the different types of imido ligands, which vary in their

electron-releasing/withdrawing and steric properties. In general,

the observed structural variations are analogous to those found

in other homologous series of imido titanium dichloride

compounds, namely, Ti(NR)Cl₂(py)₃(R ) ^tBu, Ph, 4-C₆H₄Me,

i

4-C₆H₄NO₂)⁴⁴and Ti(NR)Cl₂(NHMe₂)₂(R ) Pr, Ph, C₆F₅,

t

2-C₆H₄R′ (R′ ) CF₃, Bu)).⁴⁷

The orientation of the 2,6-disubstituted aryl substituents in 2

and 3 are somewhat different than those in 4 and 5 (Supporting

Information, Figure S1). In the former compounds the ring

carbons lie approximately in the same plane as Ti(1), N(9), and

the midpoints between Cl(1) and Cl(2). In 4 and 5 the ring

carbons are best considered to be approximately perpendicular

to this plane. Inspection of the structures suggests that the

observed orientation of the 2,6-C₆H₃R₂(R ) Me, ⁱPr) rings best

minimizes nonbonded contact between the ortho R-groups and

the macrocycle N-methyl groups cis to the imido ligand (i.e.,

C(31) and C(61), Figure 1).

Scheme 1 shows that both 1 and the new compounds

Ti(NR)(Me₃[9]aneN₃)Cl₂(R ) Ph (4), C₆F₅(5), CH₂Ph (6))

can be prepared either from already-isolated Ti(NR)Cl₂-

(NHMe₂)₂(R ) Ph or C₆F₅, giving 4 (99% yield) or 5 (79%))

or in a “one-pot” reaction from Ti(NMe₂)₂Cl₂and RNH₂, first

generating Ti(NR)Cl₂(NHMe₂)₂in situ prior to addition of the

The tert-butyl imido ligand in 1 has the shortest Ti(1)-N(1)

bond, while the longest are found for the electron-withdrawing

C₆F₅or bulkier 2,6-disubstituted phenyl ligands. The electron-

withdrawing effect of C₆F₅is also presumably responsible for

the comparatively short N(1)-C(10) distance compared to those

in 2-4. The longest and shortest Ti-Cl distances are found

for the tert-butyl- and C₆F₅-substituted systems, respectively,

consistent with their relative electron-releasing properties, but

there is no particular trend in Ti-Cl distances for compounds

2-4. Similarly, while the trans influences (see Table 1 for

definition) of the imido ligands appear to be smallest for the

C₆H₅- and C₆F₅-substituted imido ligands, there is no obvious

trend among compounds having the bulkier substituents

(1-3), and in fact it is the comparatively shorter Ti(1)-N(9)

distances in particular for 4 and 5 that appear to be mainly

responsible for the smaller trans influences of the imido ligands

in these two cases. The origins of the trans influence in

octahedral imido titanium complexes have been described

previously,⁵²and smaller influences are expected for less

electron-releasing imido substituents. There appear to be no

systematic trends in the angles subtended at the titanium cen-

ters. For instance the Cl-Ti-Cl angles in 1 (95.75(4)°) and 5

(95.77(2)°) are experimentally identical despite the significantly

different Ti-Cl distances (av 2.393 and 2.357 Å, respectively).

Notwithstanding any possible influences of intermolecular/

crystal packing forces on the intramolecular distances and angles

discussed, it appears that the different imido substituents have

significant but comparatively small structural effects on the

precatalysts.

t

macrocycle (R ) Bu or CH₂Ph, giving 1 (85%) or 6 (96%)).

The new compounds were fully characterized, and the data for

samples of 1 prepared via the “one-pot” route were identical to

those for the sample prepared from Ti(N^tBu)Cl₂(py)₃. Conduct-

ing the reactions in C₆H₆allowed for facile isolation of 1 and

4-6 since the solubilities of the compounds Ti(NR)(Me₃[9]-

aneN₃)Cl₂in this solvent are rather low and they tend to

precipitate in analytically pure form. As discussed further below,

the “one-pot” synthetic route was eventually adopted for the

synthesis of a library of such catalysts.

X-ray Crystal Structures of Ti(NR)(Me₃[9]aneN₃)Cl₂. The

solid-state structures of the compounds Ti(NR)(Me₃[9]aneN₃)Cl₂

t

i

(R ) Bu (1), 2,6-C₆H₃Me₂(2), 2,6-C₆H₃Pr₂(3), Ph (4), C₆F₅

(5)) have been determined. The molecular structures of 1-4

are given in Figure 1, and a view of 5 is provided in the

Supporting Information. The supramolecular structure of 5 and

selected intermolecular C‚‚‚F contacts are shown in Figure 2,

and selected intramolecular bond distances and angles for all

five structures are listed in Table 1. All five molecular structures

feature a κ³-coordinated Me₃[9]aneN₃ligand bound to an

approximately octahedral titanium center, the coordination

sphere of which is completed by two mutually cis chloride

ligands and an imido group. The solid-state structures are

consistent with the solution NMR data assuming free rotation

about the N(1)-C(10) bonds of 2 and 3 in solution.

While molecules of Ti(NPh)(Me₃[9]aneN₃)Cl₂(4) do not

exhibit any particular supramolecular arrangement, those of

Ti(NC₆F₅)(Me₃[9]aneN₃)Cl₂(5) form infinite stacks linked by

C(δ+)‚‚‚F(δ-) contacts in the range 3.078(2)-3.212(2) Å with

four carbons (C(11), C(12), C(13), C(14)) and two fluorines

(F(2), F(3)) of each C₆F₅ring contributing (Figure 2). Inter-

molecular interactions between perfluorophenyl rings have

In general terms, the distances and angles around the metal

centers and within the ligands are within previously re-

ported ranges,^48,49and the approximately linear imido linkages

(45) Hazari, N.; Cowley, A. R.; Mountford, P. Acta Crystallogr. 2004,

E60, m1844.

(46) Nielson, A. J.; Glenny, M. W.; Rickard, C. E. F. J. Chem. Soc.,

Dalton Trans. 2001, 232.

(47) Adams, N.; Bigmore, H. R.; Blundell, T. L.; Boyd, C. L.; Dubberley,

S. R.; Sealey, A. J.; Cowley, A. R.; Skinner, M. E. G.; Mountford, P. Inorg.

Chem. 2005, 44, 2882.

(48) Fletcher, D. A.; McMeeking, R. F.; Parkin, D. J. Chem. Inf. Comput.

Sci. 1996, 36, 746 (The United Kingdom Chemical Database Service).

(49) Allen, F. H.; Kennard, O. Chem. Des. Autom. News 1993, 8, 1&31.

(50) Cundari, T. R. Chem. ReV. 2000, 100, 807.

(51) Parsons, T. B.; Hazari, N.; Cowley, A. R.; Green, J. C.; Mountford,

P. Inorg. Chem. 2005, 8442.

(52) Kaltsoyannis, N.; Mountford, P. J. Chem. Soc., Dalton Trans. 1999,

781.

Imido Titanium Ethylene Polymerization Catalysts

Organometallics, Vol. 25, No. 16, 2006 3891

i

Figure 1. Displacement ellipsoid plots (20% probability) of Ti(NR)(Me₃[9]aneN₃)Cl₂(R ) ^tBu (1), 2,6-C₆H₃Me₂(2), 2,6-C₆H₃Pr₂(3), Ph

(4)). H atoms are omitted.

recently received increased attention both computationally and

experimentally.^53-58It has been proposed that intermolecular

interactions between perfluorinated aromatic rings are slightly

more attractive than the very well established ones between their

hydro analogues (due mainly to an increased van der Waals

component),⁵³and the structures of 4 and 5 appear to support

this. Note that in 5 the adjacent perfluorophenyl rings are

considerably offset from each other (see Figures S2 and S3 in

the Supporting Information).

this technology and the control of polymer morphology that

can be achieved. A wide range of inorganic materials (e.g., silica,

alumina, zeolites, magnesium chloride) and polymeric supports

(e.g., polysiloxanes, polystyrene, polyolefins, dendrimers) has

been used.^59,60Among these latter supports, polystyrene has

attracted much attention due to its relative inertness, ease of

functionalization, and the control afforded of the bead size

and porosity. Polystyrene-supported zirconocene,⁶¹constrained

geometry,⁶²other group 4 half-sandwich,^63-66and iron or nickel

Polystyrene-Supported Precatalysts. Immobilized single-

site olefin polymerization catalysts have been of much interest

because of their relevance to the industrial implementation of

(58) Hair, G. S.; Cowley, A. H.; Gorden, J. D.; Jones, J. N.; Jones, R.

A.; Macdonald, C. L. B. Chem. Commun. 2003, 424.

(59) Hlatky, G. G. Chem. ReV. 2000, 100, 1347.

(60) Severn, J. R.; Chadwick, J. C.; Duchateau, R.; Friederichs, N. Chem.

ReV. 2005, 105, 4073.

(61) Stork, M.; Koch, M.; Klapper, M.; Mullen, K.; Gregorius, H.; Rief,

U. Macromol. Rapid Commun. 1999, 20, 210.

(62) Kasi, R. M.; Coughlin, E. B. Organometallics 2003, 22, 1534.

(63) Kitawaga, T.; Uozumi, T.; Soga, K.; Takata, T. Polymer 1997, 38,

615.

(64) Zhu, H.; Jin, G.-X.; Hu, N. J. Organomet. Chem. 2002, 655,

167.

(65) Barrett, A. G. M.; de Miguel, Y. R. Tetrahedron 2002, 58,

3785.

(53) Lorenzo, S.; Lewis, G. R.; Dance, I. New J. Chem. 2000, 24,

295.

(54) Blanchard, M. D.; Hughes, R. P.; Concolino, T. E.; Rheingold, A.

L. Chem. Mater. 2000, 12, 1604.

(55) Deck, P. A.; Kroll, C. E.; Hollis, W. G.; Fronczek, F. R. J.

Organomet. Chem. 2001, 637-639, 107.

(56) Deck, P. A.; Lane, M. J.; Montgomery, J. L.; Slebodnick, C.;

Fronczek, F. R. Organometallics 2000, 19, 1013.

(57) Thornberry, M. P.; Slebodnick, C.; Deck, P. A.; Fronczek, F. R.

Organometallics 2000, 19, 5352.

3892 Organometallics, Vol. 25, No. 16, 2006

Adams et al.

Figure 2. Partial packing diagram for Ti(NC₆F₅)(Me₃[9]aneN₃)Cl₂(5). H atoms are omitted and all atoms are drawn as spheres of arbitrary

radius. Symmetry operators: B, [2-x, -y, 1-z]; C, [2-x, -1-y, 1-z]; D, [x, y-1, z]. The dashed lines indicate the closest points

of intermolecular contact: C(11)‚‚‚F(2B) ) 3.198(2) Å; C(12)‚‚‚F(2B) ) 3.114(2) Å; C(13)‚‚‚F(3C) ) 3.078(2) Å; C(14)‚‚‚F(3C) )

3.212(2) Å.

Table 1. Selected Bond Lengths (Å) and Angles (deg) for

t

Ti(NR)(Me₃[9]aneN₃)Cl₂(R ) Bu (1), 2,6-C₆H₃Me₂(2),

i

2,6-C₆H₃Pr₂(3), Ph (4), C₆F₅(5))

parameter

Ti(1)-N(1)

1

2

3

4

5

1.694(2) 1.736(2) 1.735(2)

2.265(3) 2.281(2) 2.271(2)

2.270(3) 2.252(2) 2.267(2)

2.437(3) 2.426(2) 2.441(2)

2.394(1) 2.3889(8) 2.3613(10) 2.3832(5) 2.3565(4)

2.3915(9) 2.3787(9) 2.3904(10) 2.3859(5) 2.3581(4)

1.454(4) 1.387(3) 1.396(3)

0.172(4) 0.145(3) 0.170(3)

0.167(4) 0.174(3) 0.174(3)

1.717(2)

1.7420(13)

Ti(1)-N(3)

Ti(1)-N(6)

Ti(1)-N(9)

Ti(1)-Cl(1)

Ti(1)-Cl(2)

N(1)-C(10)

trans

2.2651(14) 2.2723(13)

2.2787(14) 2.2532(12)

2.3921(14) 2.3868(12)

Catalysts of the type Ti(NR)(Me₃[9]aneN₃)Cl₂can in principle

be immobilized via either the imido ligand R-group or the

macrocycle. We have established “proof of concept” results for

both approaches and prepared Ti(NCH₂PS)(Me₃[9]aneN₃)Cl₂

(7), the new supported macrocyclic ligand PSCH₂Me₂[9]aneN₃

(8), and its derivative Ti(N^tBu)(PSCH₂Me₂[9]aneN₃)Cl₂(9)

according to the further details given in the Supporting

Information (“PS” represents 1% cross-linked polystyrene).

Triazacyclohexane-Supported Imido Titanium Precata-

lysts. 1,3,5-Triazacyclohexane ligands (R₃[6]aneN₃) have smaller

ring sizes than their R₃[9]aneN₃homologues, and their coor-

dination chemistry in general is less extensively developed in

comparison.^34,72-79Recently, Ko¨hn and co-workers reported that

MAO-activated chromium complexes of the type Cr(R₃[6]-

1.387(2)

0.127(2)

0.113(2)

92.35(6)

99.25(6)

1.365(2)

0.114(2)

0.134(2)

93.38(5)

96.39(5)

influence (Å)^a

N(1)-Ti(1)-N(3) 97.3(1)

N(1)-Ti(1)-N(6) 96.6(1)

N(1)-Ti(1)-N(9) 169.1(1) 172.20(9) 171.78(8) 167.21(6) 167.41(5)

N(1)-Ti(1)-Cl(1) 100.34(9) 98.49(8) 99.60(7)

N(1)-Ti(1)-Cl(2) 98.49(9) 100.17(8) 98.94(7)

Cl(1)-Ti(1)-Cl(2) 95.75(4) 97.78(3) 97.36(4)

Ti(1)-N(1)-C(10) 171.0(2) 166.3(2) 166.5(2)

98.57(9) 98.77(9)

99.5(1) 99.47(9)

97.40(5)

101.07(5) 99.41(4)

96.32(2) 95.77(2)

168.18(14) 171.22(12)

101.46(4)

^aDefined as the difference between the Ti(1)-N(9) distance (i.e., trans

to TidNR) and Ti(1)-N(3) and Ti(1)-N(6) distances.

R-diimine^67-70catalysts have been reported. Of particular

relevance to our work are Gibson’s⁷¹and Nielson’s⁴⁶studies

of polystyrene-supported imido vanadium⁷¹and titanium⁴⁶

ethylene polymerization catalysts, respectively.

(72) Ko¨hn, R.; Haufe, M.; Mihan, S.; Lilge, D. Chem. Commun. 2000,

1927.

(66) Zhang, Y.; Sita, L. R. Chem. Commun. 2003, 2358.

(67) Zhang, D.; Jin, G.-X. Appl. Catal. A 2004, 262, 13.

(68) Liu, C.; Jin, G.-X. New J. Chem. 2002, 10, 1485.

(69) Boussie, T. R.; Coutard, C.; Turner, H.; Murphy, V.; Powers, T. S.

Angew. Chem., Int. Ed 1998, 37, 3272.

(70) Boussie, T. R.; Murphy, V.; Hall, K. A.; Coutard, C.; Dales, C.;

Petro, M.; Carlson, E.; Turner, H. W.; Powers, T. S. Tetrahedron 1999,

55, 11699.

(73) Ko¨hn, R. D.; Haufe, M.; Kociok-Ko¨hn, G.; Filippou, A. C. Inorg.

Chem. 1997, 36, 6064.

(74) Ko¨hn, R. D.; Kociok-Ko¨hn, G. Angew. Chem., Int. Ed. Engl. 1994,

33, 1877.

(75) Ko¨hn, R. D.; Pan, Z.; Kociok-Ko¨hn, G.; Mahon, M. F. J. Chem.

Soc., Dalton Trans. 2002, 2344.

(76) Tredget, C. S.; Lawrence, S. C.; Ward, B. D.; Howe, R. G.; Cowley,

A. R.; Mountford, P. Organometallics 2005, 24, 3136.

(77) Armanasco, N. L.; Baker, M. V.; North, M. R.; Skelton, B. R.;

White, A. H. J. Chem. Soc., Dalton Trans. 1997, 1363.

(71) Chan, M. C. W.; Chew, K. C.; Dalby, C. I.; Gibson, V. C.;

Kohlmann, A.; Little, I. R.; Reed, W. Chem. Commun. 1998, 1673.

Imido Titanium Ethylene Polymerization Catalysts

Organometallics, Vol. 25, No. 16, 2006 3893

Table 2. Ethylene Polymerization Data for Triazacycle-Supported Imido Titanium Catalysts (ambient temperature)^a

precatalyst

loading (µmol)

yield of

PE (g)

productivity (kg

entry

precatalyst

mol^-1h^-1bar^-1

)

M_w(g mol^-1

)

M_n(g mol^-1

)

M_w/M_n

1

2

3

4

5

6

7

8

Ti(N^tBu)(Me₃[9]aneN₃)Cl₂(1)

2.5

20

2.5

20

2.5

20

9.5

0.92

1.8

630

8

120

<1

5

13

3

<1

2

5

3

2

1

220

580

3

1 140 000

736 000

440 000

c

264 000

12 900

55 500

c

4.3

57

7.9

Ti(N-2,6-C₆H₃Me₂)(Me₃[9]aneN₃)Cl₂(2)

i

Ti(N-2,6-C₆H₃Pr₂)(Me₃[9]aneN₃)Cl₂(3)

Ti(NPh)(Me₃[9]aneN₃)Cl₂(4)

0.04

0.62

1.54

0.44^b

0.03^b

0.25

0.60

0.37

0.27

0.12

3.2

Ti(NC₆F₅)(Me₃[9]aneN₃)Cl₂(5)

Ti(NCH₂Ph)(Me₃[9]aneN₃)Cl₂(6)

Ti(NCH₂PS)(Me₃[9]aneN₃)Cl₂(7)

Ti(N^tBu)(PSCH₂Me₃[9]aneN₃)Cl₂(9)

Ti(N^tBu)(Me₃[6]aneN₃)Cl₂(10)

c

874 000

1 300 000

624 000

803 000

977 000

1 040 000

812 500

647 000

2 245 000

532 000

372 500

78 000

117 000

19 200

12 250

19 650

16 750

16 450

11 050

937 000

165 000

25 800

11

33

31

54

64

50

59

2.4

3.3

14

9

i

10

11

12

13

14

15

16

Ti(N-2,6-C₆H₃Pr₂)(Me₃[6]aneN₃)Cl₂(11)

Ti(NC₆F₅)(Me₃[6]aneN₃)Cl₂(13)

Ti(N^tBu)(dd₃[6]aneN₃)Cl₂(14)

i

Ti(N-2,6-C₆H₃Pr₂)(dd₃[6]aneN₃)Cl₂(15)

Ti(N-2-C₆H₄CF₃)(Me₃[9]aneN₃)Cl₂(57)

t

Ti(N-2-C₆H₄Bu)(Me₃[9]aneN₃)Cl₂(58)

8.6

0.41

t

Ti(N-2-C₆H₄Bu)(Me₃[6]aneN₃)Cl₂(59)

^aConditions: 6 bar C₂H₄on demand, Al:Ti ) 1 500, 250 mL of toluene, 1 L reactor, 1 h, ambient temperature. ^bCorrected for weight of polystyrene

support. ^cPolymer was very insoluble and gave a poor refractive index detector response.

aneN₃)Cl₃are highly active homogeneous models of the Phillips-

type catalysts.⁷²Therefore it was of interest to evaluate their

effectiveness as supporting ligands in imido titanium polym-

erization catalysis.

is by far the most active, producing high M_wpolyethylene (PE)

with a polydispersity index (PDI) of 4.3. Single-site MAO-

activated catalysts typically produce PE with PDIs of 2-2.5.

The larger value for 1 may indicate changing reaction conditions

(e.g., temperature rise as the polymerization proceeds, mass

transport limitations, precipitation of insoluble PE) or the

presence of more than one active species.¹⁸Of the aryl imido

systems 2-5, it is again the compound with the bulkier imido

i

R-group, namely, Ti(N-2,6-C₆H₃Pr₂)(Me₃[9]aneN₃)Cl₂(3), that

affords the higher productivity. This catalyst system forms PE

with a lower M_wthan that formed with 1 and with a somewhat

broader PDI.

The polystyrene-supported systems Ti(NCH₂PS)(Me₃[9]-

aneN₃)Cl₂(7) and Ti(N^tBu)(PSCH₂Me₃[9]aneN₃)Cl₂(9) had

lower activities than their solution phase analogues 6 and 1,

respectively. The PE formed with 7 had a higher M_wthan that

formed with 6, suggesting that chain transfer is inhibited relative

to chain propagation in this system. In contrast, the PE formed

with 9 had a lower M_wthan that obtained with 1. However,

given the very low productivity and mass of PE isolated in the

case of 9, it is unsafe to draw firm conclusions from this

observation.

As part of a general survey of imido titanium coordination

chemistry, we previously reported³⁴the compounds Ti(NR)-

t

i

(Me₃[6]aneN₃)Cl₂(R ) Bu (10), 2,6-C₆H₃Pr₂(11)) together

with the X-ray structure of the tert-butyl derivative. However,

no polymerization studies using these compounds have been

reported, and so we included them in the current study. For

comparison with the aryl imido triazacyclononane systems 4

and 5 we also prepared Ti(NR)(Me₃[6]aneN₃)Cl₂(R ) Ph (12),

C₆F₅(13)). Finally, since Ko¨hn showed that the ethylene

polymerization productivity of the tris(n-dodecyl)-1,3,5-triaza-

cyclohexane complex Cr(dd₃[6]aneN₃)Cl₃is ca. 60% higher than

that of Cr(Me₃[6]aneN₃)Cl₃, we also prepared Ti(NR)(dd₃[6]-

There is literature precedent for both enhanced and diminished

catalyst productivities on going from solution phase systems to

their supported analogues.⁶⁰Of particular relevance to our work

are contrasting observations made for two previous supported

imido-based catalyst systems. Nielson found that polystyrene-

supported imido titanium catalysts of the type Ti(NR)Cl₂(tmeda)

had almost negligible activity compared to their solution phase

analogues,⁴⁶whereas Gibson found the opposite was true for

cyclopentadienyl vanadium (+5) systems. This latter result was

rationalized in terms of decreased bimolecular catalyst deactiva-

tion in the supported (site-isolated) system.⁷¹

t

i

aneN₃)Cl₂(R ) Bu (14), 2,6-C₆H₃Pr₂(15); dd ) n-dodecyl).

For further details of the syntheses and the X-ray structures see

the Supporting Information.

Preliminary Ethylene Polymerization Studies. Ambient

Temperature Screening of Triazacyclononane Systems. The

solution phase precatalysts 1-6, the solid-phase analogues 7

and 9, and the R₃[6]aneN₃systems 10-15 were screened for

their ethylene homopolymerization capability at ambient tem-

perature in toluene using methyl aluminoxane (MAO) cocatalyst

(ratio Al:Ti ) 1500:1). On the basis of the results of preliminary

experiments, the more active systems 1 and 3 were reevaluated

at a lower catalyst loading to minimize mass transport limitation

effects. The results are summarized in Table 2.

There may be several reasons for the comparatively low

productivities of the supported systems 7 and 9. One very likely

contributing factor is the much lower apparent swelling of the

resins in toluene when they contain triazacyclononane-titanium

dichloride moieties. This would doubtless block cocatalyst

and/or monomer access to the metal centers. Higher produc-

tivities of supported imido titanium catalysts could potentially

be achieved with a more rigid support. However, such studies

were beyond the scope of the present study.

Of the two alkyl imido (solution phase) systems Ti(NR)(Me₃-

[9]aneN₃)Cl₂(R ) ^tBu (1), CH₂Ph (6)), the tert-butyl analogue

(78) Armanasco, N. L.; Baker, M. V.; North, M. R.; Skelton, B. W.;

White, A. H. J. Chem. Soc., Dalton Trans. 1998, 1145.

(79) Armanasco, N. L.; Baker, M. V.; North, M. R.; Skelton, B. W.;

White, A. H. J. Chem. Soc., Dalton Trans. 1997, 1363.

Ambient-Temperature Screening of Triazacyclohexane

Systems. The performance of the triazacyclohexane systems

10-15 was disappointing in all cases assessed (given the very

3894 Organometallics, Vol. 25, No. 16, 2006

Adams et al.

Table 3. Ethylene Polymerization Data for Triazacyclononane-Supported Imido Titanium Catalysts (100 °C)

Al:Ti productivity (kg

M_w

M_n

vinyl end groups

per chain^b

entry

precatalyst

conditions^a

ratio

mol^-1h^-1bar^-1

)

(g mol^-1

)

(g mol^-1

)

M_w/M_n

1

2

3

4

5

6

7

8

Ti(N^tBu)(Me₃[9]aneN₃)Cl₂(1)

Ti(N-2,6-C₆H₃Pr₂)(Me₃[9]aneN₃)Cl₂(3)

A

B

A

660

23 800

220

c

0.37

0.11

0.59

1.5^d

0.63

0.56

0.85

0.68

f

i

Ti(NPh)(Me₃[9]aneN₃)Cl₂(4)

Ti(NC₆F₅)(Me₃[9]aneN₃)Cl₂(5)

Ti(N^tBu)(Me₃[9]aneN₃)Cl₂(1)

Ti(NCH₂C₆H₁₁)(Me₃[9]aneN₃)Cl₂(19)

Ti(NC₅H₉)(Me₃[9]aneN₃)Cl₂(21)

Ti(N^tBu)(Me₃[9]aneN₃)Cl₂(1)

70

660

c

1600

6500

1200

1600

96 600

13 100

5890

4820

10 070

8730

10 340

3370

1230

690

100 000

11 000

152 700

1 501 900

273 850

190 700

152 900

50 300

863 000

667 000^e

37 000

5100

2.7

2.2

10

3.8

7.0

5.8

6.7

4.8

4.4

26^e

15 200

391 200

39 500

33 000

22 900

10 400

195 500

25 700^e

9

t

10

11

12

13

14

Ti(NCMe₂CH₂Bu)(Me₃[9]aneN₃)Cl₂(23)

Ti(NAd)(Me₃[9]aneN₃)Cl₂(24)

Ti{N-3,5-C₆H₃(CF₃)₂}(Me₃[9]aneN₃)Cl₂(39)

Ti(N-2-C₆H₄CF₃)(Me₃[9]aneN₃)Cl₂(57)

t

Ti(N-2-C₆H₄Bu)(Me₃[9]aneN₃)Cl₂(58)

f

^aConditions: (A) 7 bar C₂H₄on demand, 750 mL of PMH (or 1000 mL of heptanes for entries 5 and 6), 2 L autoclave, 0.05-5 µmol of precatalyst, 10

min, 100 °C; (B) 7 bar C₂H₄on demand, 62 mL of PMH, parallel microreactor array, 0.16 µmol of precatalyst, 10 min, 100 °C. ^bDetermined by NMR

and/or FTIR spectroscopy (see the Supporting Information for further details). ^cNot recorded. ^dEstimate of ∼1.5 RCHdCH₂per chain may be inaccurate

due to low number of unsaturations per 100 000 C and high M_w. ^eBimodal. ^fNot determined (additional unknown absorptions obscured necessary spectral

region).

poor performance of Ti(NPh)(Me₃[9]aneN₃)Cl₂, compound 12

was not tested). The low productivities in each instance were

accompanied by very broad and complex molecular weight

distributions, indicating extensive catalyst degradation to several

species of limited activity. It is possible that the more open metal

centers afforded by the smaller macrocycles lead more easily

to bimolecular decomposition pathways, for instance via het-

erobimetallic or homobimetallic species, possibly with bridging

imido ligands. Note, for example, that CpTi(N^tBu)Cl(py) readily

loses pyridine to form the µ-imido-bridged dimer Cp₂Ti₂(µ-

N^tBu)₂Cl₂, whereas the bulkier Cp*Ti(N^tBu)Cl(py) is perfectly

stable in this regard.⁸⁰Alternatively (or additionally) it is

possible that the R₃[6]aneN₃ligands are lost more readily in

the triazacyclohexane systems. We note that Ko¨hn has advanced

evidence⁸¹for transfer of up to 50% of the triazacyclic ligand

from chromium to aluminum in certain Cr(R₃[6]aneN₃)Cl₃-based

catalyst systems.

w/w in the MAO used) and â-hydrogen transfer to ethylene

monomer leading to RCHdCH₂end groups (R ) polymeryl).

The PE produced had a PDI of only 2.7, indicative of a single

site type catalyst.

Increasing the MAO:1 ratio to 6500:1 (Table 3, entry 6) gave

a substantial decrease in productivity and PE with a much lower

M_wand an average of only 0.11 vinyl groups per chain. Since

increasing the amount of MAO also increases the absolute

amount of AlMe₃in the system, this would lead to both an

increased rate of chain transfer to AlMe₃and suppression of

catalyst activity due to increased equilibrium concentrations of

AlMe₃adducts of the type [Ti(N^tBu)(Me₃[9]aneN₃)(µ-Me)(µ-

R)AlMe₂]⁺(R ) polymeryl or Me), which has been character-

ized crystallographically for R ) Me.⁴¹

Regrettably, none of the complexes evaluated showed an

affinity for propene polymerization or incorporation as co-

monomer.

Synthesis and Evaluation of a Library of Precatalysts

Ti(NR)(Me₃[9]aneN₃)Cl₂. Ti(N^tBu)(Me₃[9]aneN₃)Cl₂/MAO

(1/MAO) is the most productive imido-based Ziegler ethylene

polymerization catalyst reported to date²³and operates under

commercially relevant conditions. Tables 2 and 3 also show

clear evidence of dependence of productivity on the imido

Higher Temperature Screening of Triazacyclononane

Systems (Conventional Semibatch Reactors). While the

triazacyclohexane systems are not promising candidates for

development, the ambient temperature data (Table 2) for certain

triazacyclononane precatalysts suggested that these could be

viable systems for development under more commercially

relevant conditions, in particular at higher operating tempera-

tures. Preliminary experiments at 100 °C are shown in Table 3,

entries 1-6.

t

R-group (for example, precatalyst 1 (R ) Bu) has a signifi-

i

cantly higher productivity than 3 (R ) 2,6-C₆H₃Pr₂), especially

at higher temperatures). The ambient temperature data indi-

cated that polymer properties are influenced, and perhaps could

eventually be controlled, by appropriate choice of imido

N-substituent. Over the last 5 years in particular, high-throughput

catalyst synthesis and screening techniques^10,82,83have started

to become well established for the discovery and optimization

of polymerization catalyst systems.^84-92Because the compounds

Ti(NR)(Me₃[9]aneN₃)Cl₂are easily prepared from our imido-

dichloride synthons Ti(NR)Cl₂(NHMe₂)₂, we decided to target

a library of such precatalysts and evaluate their polymerization

capabilities using a high-throughput screening approach. No

systematic survey of structure-activity relationships in imido-

based Ziegler catalyst systems²³has so far been described

beyond our own preliminary communications.^40,43

The catalyst systems Ti(NR)(Me₃[9]aneN₃)Cl₂/MAO (R )

^tBu (1), 2,6-C₆H₃Pr₂(3), Ph (4), C₆F₅(5)) were evaluated for

i

ethylene polymerization in a conventional semibatch autoclave

reactor at 100 °C with an initial Al:Ti ratio of 660:1 (Table 3,

entries 1-4). While the tert-butyl imide 1 showed an excellent

productivity of 23 800 kg mol^-1h^-1bar^-1, the three aryl imido

systems were very disappointing. Focusing on the 1/MAO

system, it was found that increasing the relative amount of MAO

(entry 5, Al:Ti ratio 1600:1) gave a 4-fold increase in productiv-

ity to 96 600 kg mol^-1h^-1bar^-1. NMR analysis of the linear

PE produced showed an average of 0.37 vinyl end group per

chain. This suggests¹⁸that two predominant chain transfer

mechanisms are operative in the 1/MAO catalyst system,

namely, chain transfer to AlMe₃(present in ca. 20% amounts

Library Synthesis. A “one-pot” synthesis using the sequen-

tial addition of RNH₂and then Me₃[9]aneN₃to Ti(NMe₂)₂Cl₂

(80) Dunn, S. C.; Mountford, P.; Robson, D. A. J. Chem. Soc., Dalton

Trans. 1997, 293.

(81) Ko¨hn, R. D.; Smith, D.; Mahon, M. F.; Prinz, M.; Mihan, S.; Kociok-

Ko¨hn, G. J. Organomet. Chem. 2003, 683, 200.

(82) Murphy, V.; Volpe, A. F.; Weinberg, W. H. Curr. Op. Chem. Biol.

2003, 7, 427.

(83) Stambuli, J. M.; Hartwig, J. F. Curr. Op. Chem. Biol. 2003, 7, 420.

Imido Titanium Ethylene Polymerization Catalysts

Organometallics, Vol. 25, No. 16, 2006 3895

Figure 3. Summary of ethylene polymerization productivities at 100 °C for the library of catalysts Ti(NR)(Me₃[9]aneN₃)Cl₂. Selected

N-R groups are highlighted. See Table 3 and the Supporting Information for further details.

was selected since this route had already been used successfully

route, and the isolated yields ranged from ca. 30 to 95 (av 61)%.

These included the five previously prepared compounds 1-3,

5, and 6 for comparison. The NMR data and semibatch reactor

polymerization capability of the compounds prepared in the

semiautomated manner were identical to those of the conven-

tionally prepared samples. The 41 new compounds (16-56)

t

for the preparation of Ti(NR)(Me₃[9]aneN₃)Cl₂(R ) Bu or

CH₂Ph, Scheme 1) and Ti(NR)(R′₃[6]aneN₃)Cl₂(R′ ) Me, R

i

) Ph, C₆F₅; R′ ) dd, R ) 2,6-C₆H₃Pr₂). From the point of

view of using semiautomated technology, this method has the

great advantage that all the reagents (Ti(NMe₂)₂Cl₂, RNH₂, and

Me₃[9]aneN₃) and also the intermediates Ti(NR)Cl₂(NHMe₂)₂

are very benzene-soluble, whereas the target dichlorides

Ti(NR)(Me₃[9]aneN₃)Cl₂have quite limited solubility in this

solvent and should precipitate, at least in sufficient quantities

for characterization and subsequent use in polymerization

screening. For otherwise “clean” reactions this therefore allows

for ready separation and washing of the desired products and

the use, where necessary, of an excess of certain reagents

because they can be readily washed away. Full details of the

method for library synthesis are provided in the Experimental

Section, and the characterizing data and yields for all the

compounds are given in the Supporting Information.

1

were characterized by satisfactory and clean H NMR spectra

in every case, and 39 of them gave correct molecular ions (or

conclusive fragments arising from Cl, Me, or similar simple

radical loss) in their high-resolution EI mass spectra (see the

Supporting Information). None of the Ti(NR)Cl₂(NHMe₂)₂

intermediates was isolated, but in a previous paper we pre-

pared and fully characterized 13 such species (including six

X-ray structures), and we have also shown explicitly above that

the compounds Ti(NR)(Me₃[9]aneN₃)Cl₂(R ) Ph, C₆F₅) and

Ti(N^tBu)(dd₃[6]aneN₃)Cl₂are formed by addition of R′₃[n]aneN₃

to such precursors (R′ ) Me or dd; n ) 6 or 9).

Polymerization Testing and Polymer Properties. The

precatalysts 1-3, 5, 6, and 16-56 were screened for their

ethylene polymerization capability at 100 °C using MAO co-

catalyst (Al:Ti ) 1200:1) in high-throughput, parallel micro-

reactors. The productivities are displayed graphically in Figure

3. Further details of the productivities and PE properties for

the most productive precatalysts, namely, 1, 19, 21, 23, 24, and

39 (Figure 4), are given in Table 3. Full details of these PE

analyses and all the catalyst productivities are given in the

Supporting Information.

For ease of reference, the imido R-groups in Figure 3 are

arranged as follows (from left to right): primary, secondary,

and tertiary alkyl; benzyl; para-, meta-, mono-ortho-, and di-

ortho-phenyl. It is clear that the productivity of Ti(N^tBu)(Me₃-

[9]aneN₃)Cl₂(1, 10 070 kg mol^-1h^-1bar^-1, Table 3, entry 9)

using the microreactors is lower than that found on the 2 L

conventional semibatch reactors at Al:Ti ratios of either 660

(23 800 kg mol^-1h^-1bar^-1, Table 3, entry 1) or 1600 (96 600

kg mol^-1h^-1bar^-1, Table 3, entry 5). This decrease in

productivity is not an uncommon feature of this microreactor

A library of 46 compounds of the type Ti(NR)(Me₃[9]-

aneN₃)Cl₂was prepared using this “one-pot” semiautomated

(84) Tian, J.; Coates, G. W. Angew. Chem., Int. Ed. 2000, 39, 3626.

(85) Diamond, G. M.; Goh, C.; Leclerc, M. K.; Murphy, V.; Turner, H.

W. Symyx Technologies Inc., 2001.

(86) Hagemeyer, A.; Jandeleit, B.; Liu, Y.; Poojary, D. M.; Turner, H.

W.; Volpe, J.; Anthony F.; Henry Weinberg, W. Appl. Catal. A: Gen. 2001,

221, 23.

(87) Jones, D. J.; Gibson, V. C.; Green, S. M.; Maddox, P. J. Chem.

Commun. 2002, 1038.

(88) Boussie, T. R.; Diamond, G. M.; Goh, C.; Hall, K. A.; LaPointe,

A. M.; Leclerc, M. K.; Lund, C.; Murphy, V.; Shoemaker, J. A. W.; Tracht,

U.; Turner, H.; Zhang, J.; Uno, T.; Rosen, R. K.; Stevens, J. C. J. Am.

Chem. Soc. 2003, 125, 4306.

(89) Kolb, P.; Demuth, D.; Newsam, J. M.; Smith, M. A.; Sundermann,

A.; Schunk, S. A.; Bettonville, S.; Breulet, J.; Francois, P. Macromol. Rapid

Commun. 2004, 25, 280.

(90) Mason, A. F.; Coates, G. W. J. Am. Chem. Soc. 2004, 10798.

(91) Kerton, F. M.; Whitwood, A. C.; Willans, C. E. Dalton Trans. 2004,

2237.

(92) Jones, D. J.; Gibson, V. C.; Green, S. M.; Maddox, P. J.; White, A.

J. P.; Williams, D. J. J. Am. Chem. Soc. 2005, 127, 11037.

3896 Organometallics, Vol. 25, No. 16, 2006

Adams et al.

Figure 4. The most productive ethylene polymerization catalysts as assessed using high-throughput microreactors at 100 °C (see Figure

3 and Table 3 for further details).

system and may reflect differences in precatalyst and cocatalyst

concentration and gas mixing. However, the general trends in

the productivities as a function of imido N-R group determined

using the high-throughput apparatus parallel those found in the

100 °C conventional semibatch reactor tests for 1, Ti(N-2,6-

somewhat broad for single-site catalysts. This might again reflect

the different conditions in the microreactors compared to the

conventional semibatch reactors (compare Table 3, entries 5

and 9, for the catalyst system 1/MAO). The PEs generally

contain between 0.56 and 0.85 vinyl end group per polymer

chain, so it is also clear that at least two different chain transfer

mechanisms are operating. Except for that from 21/MAO (Table

3, entry 8), the PE analysis indicates the formation of between

ca. 300 and 500 polymer chains per precatalyst Ti center

(assuming that all are activated and productive). The catalyst

systems with the highest numbers of chains per Ti (namely,

39/MAO (ca. 380) and 24/MAO (ca. 530) also gave the higher

vinyl end group content. The PE molecular weights for all

samples except that from 21 were in the range ca. 50 000-

i

C₆H₃Pr₂)(Me₃[9]aneN₃)Cl₂(3), and Ti(NC₆F₅)(Me₃[9]aneN₃)-

Cl₂(5), with 3/MAO and 5/MAO giving zero or negligible

yields of PE compared to 1 by both methods (Table 3 and Table

S3 of the Supporting Information).

The structure-productivity profile described by Figure 3

shows clearly that (at 100 °C with MAO) bulky tertiary alkyl-

substituted compounds Ti(NR)(Me₃[9]aneN₃)Cl₂are consistently

the most productive (i.e., 1, 23, and 24). Furthermore, this self-

consistency further validates the library screening technique (in

addition to the identical structure-activity trends found for 1,

3, and 5 under both semibatch and microreactor screening). With

the exception of 19 and 21 (Figure 4), all other primary and

secondary N-alkyl-substituted precatalysts 16-22 give little or

no PE. We note that compounds 19 and 21 contain a cycloalkyl

group as or part of their substituents, although it is not clear

why these would confer a special advantage. The benzyl-

substituted imido precatalysts 6 and 25-30 all showed poor

activities. It is interesting to note that Ti(NCH₂Ph)(Me₃[9]-

aneN₃)Cl₂(6) is the unsaturated analogue of reasonably active

19. None of the para-substituted phenyl imides 31-37 showed

productivities of any merit. This is consistent with the conven-

tional semibatch reactor results for Ti(NPh)(Me₃[9]aneN₃)Cl₂

(Table 3, entry 3). Compound 39 is interesting, as it appears to

show that meta-positioned electron-withdrawing groups may

assist catalyst productivity. The adjacent compound Ti{N-3,4,5-

C₆H₂(OMe)₃}(Me₃[9]aneN₃)Cl₂(38), with electron-releasing

groups in the meta- and para-positions, has a productivity that

is 10% of that of 39. On the other hand, compounds Ti(N-4-

C₆F₄CF₃)(Me₃[9]aneN₃)Cl₂(54) and Ti(NC₆F₅)(Me₃[9]aneN₃)-

Cl₂(5), for example, gave even lower productivities than 38,

and so simply using electron-withdrawing aryl groups in general

appears to give no advantage. None of the mono-ortho- and

di-ortho-phenyl-substituted imides had significant productivities.

270 000 kg mol^-1

.

The PE formed with 21/MAO was very different from that

produced by the other catalyst systems. This PE has consider-

ably higher M_wand only ca. 15 PE chains per precatalyst Ti

center. At this time it is not possible to speculate further on

this result.

High-Temperature versus Low-Temperature Structure-

Productivity Relationships and Precatalysts with 2-C₆H₄R

t

(R ) CF₃, Bu) Substituents. The ambient-temperature pro-

ductivities for the precatalysts 1-6 (Table 2, entries 1-6) sug-

gest that bulky aryl imido (e.g., 3) as well as alkyl imido sys-

tems (e.g., 1) should have reasonable productivities on moving

to higher temperature testing. However, the productivities for

i

Ti(N^tBu)(Me₃[9]aneN₃)Cl₂(1) and Ti(N-2,6-C₆H₃Pr₂)(Me₃[9]-

aneN₃)Cl₂(3) at 100 °C (Table 3, entries 1, 2; Figure 3) show

that this expectation is unfulfilled. In this context it is in-

teresting to note our very recent results for the MAO-activated

tris(pyrazolyl)methane-supported titanium imido compounds

t

i

Ti(NR){HC(Me₂pz)₃}Cl₂(I, R ) Bu (a), 2,6-C₆H₃Pr₂(b),

2-C₆H₄CF₃(c), 2-C₆H₄Bu (d)).⁴⁰Under identical ambient

t

temperature conditions to those used for 1 and 3 (Table 2, entries

1, 3) the productivities of Ia-Id were 230, 810, 1170, and 930

kg mol^-1h^-1bar^-1, respectively. However, at 100 °C both Ia

and Ib were virtually inactive, whereas Ic and Id had produc-

tivities of 133 700 (PDI ) 3.8) and 152 500 (PDI ) 2.4) kg

mol^-1h^-1bar^-1. It therefore appears that dramatic changes in

relative productivities (i.e., compounds with different N-R

The PEs produced by compounds 1, 19, 21, 23, 24, and 39

were predominantly linear with little branching (Table S4,

Supporting Information). The PDIs (Table 3, entries 7-12) were

Imido Titanium Ethylene Polymerization Catalysts

Organometallics, Vol. 25, No. 16, 2006 3897

groups) on going from ambient to higher temperature conditions

are not unexpected in these classes of imido titanium catalyst

systems.

active species in polymerization,^93-99typically a monoalkyl

cation “[(L)MR]⁺”.¹⁸In certain catalyst systems the AlMe₃can

also degrade or otherwise modify the supporting ligand

set.^15,21As mentioned above, several experimental observations

point to an active role of AlMe₃in catalyst systems based on

Ti(NR)(Me₃[9]aneN₃)Cl₂/MAO. First the PE formed has, on

average, significantly fewer vinyl end groups per chain (Table

3, entries 5-7 and 9-12) than would be expected if â-H

transfer to monomer were the dominant chain transfer process,

and the use of a very large excess of MAO (Al:Ti ) 6500;

Table 3 entry 6) gives PE having a drastically reduced M_w

and even fewer vinyl groups per chain. Second, we have re-

cently structurally characterized the AlMe₃adduct [Ti(N^tBu)-

(Me₃[9]ane)(µ-Me)₂AlMe₂]⁺(II⁺) as its [BAr^F]^-salt (Ar^F)

4

C₆F₅), showing that even with the bulky tert-butyl imido ligand

the 14 valence electron cation [Ti(N^tBu)(Me₃[9]ane)Me]⁺can

interact strongly with AlMe₃.⁴¹Third, we found that addition

of an excess of AlMe₃to the highly active catalyst system Ti-

The “value-added” benefit (relative to tert-butyl) of the bulky

ortho-aryl imido N-substituents in the tris(pyrazolyl)meth-

ane precatalyst family I was attributed to fundamental differ-

ences in topology between the two different fac-N₃lig-

ands, HC(Me₂pz)₃and Me₃[9]aneN₃.⁴⁰Thus while κ³-bound

HC(Me₂pz)₃ligands have clefts between the Me₂pz rings, which

are able to favorably direct the ortho substituent of an aryl imido

ligand to protect the potentially reactive^31-33TidNR bond

and/or highly electrophilic metal center, Me₃[9]aneN₃has no

such feature. Nevertheless, based on the successes of Ic and

Id, we prepared the new compounds Ti(N-2-C₆H₄R)(Me₃[9]-

t

(N-2-C₆H₄Bu){HC(Me₂pz)₃}Cl₂(Id)/MAO (ambient temper-

ature) gave a ca. 55% decrease in productivity and distinctly

bimodal PE (the two fractions arising from chain transfer to

AlMe₃(lower M_wpart) and â-H transfer to monomer).¹⁰⁰

aneN₃)Cl₂(R ) CF₃(57), ^tBu (58)) and Ti(N-2-C₆H₄Bu)(Me₃-

t

[6]aneN₃)Cl₂(59) to test the effect of these bulkier aryl imido

groups when used in conjunction with triazacyclic ligands.

Compounds 57-59 were obtained in ca. 65-89% yield using

the “one-pot” method starting from Ti(NMe₂)₂Cl₂and the

appropriate aniline and triazacycle.

As shown in Table 2 (entries 14, 15), the catalyst systems

57/MAO and 58/MAO have ambient temperature productivities

of 220 and 580 kg mol^-1h^-1bar^-1, which compare favorably

with those for 1 (630 kg mol^-1h^-1bar^-1) and 3 (120 kg mol^-1

h^-1bar^-1) under identical conditions. The PDIs for the PEs

formed with 57 (2.4) and 58 (3.3) are indicative of single-site-

type behavior, and the PE for 57 had a particularly high M_w.

Disappointingly, the productivity for the smaller triazacyclic

system 59/MAO (Table 2, entry 16) was still very low (4.1 kg

mol^-1h^-1bar^-1) and gave PE with a relatively broad PDI. No

further studies of this system were undertaken.

Polymerization Experiments Using MAO Modified by

BHT-H. It has been known for some time that bulky aryloxide

derivatives of AlMe₃, for example MeAl(BHT)₂(readily

formed^101,102from AlMe₃and BHT-H, where BHT-H ) 2,6-

di-tert-butyl-4-methylphenol), can act as “noninteracting”

scavengers.^99,103-105More recently, Busico et al. showed that

addition of 2,6-di-tert-butylphenol to commercial MAO in

appropriate quantities scavenged the “free” AlMe₃as the

expected bis(phenoxide).¹⁰⁶Use of such a 2,6-di-tert-butylphe-

nol/MAO cocatalyst with certain key metallocene and post-

metallocene precatalysts afforded higher catalyst productivities

The catalyst systems 57/MAO and 58/MAO were evaluated

at 100 °C in a conventional semibatch reactor (Table 3, entries

(93) Bochmann, M.; Lancaster, S. J. Angew. Chem., Int. Ed. Engl. 1994,

33, 1634.

(94) Bochmann, M.; Lancaster, S. J. J. Organomet. Chem. 1995, 497,

55.

(95) Bryliakov, K. P.; Babushkin, D. E.; Talsi, E. P.; Voskobynikov, A.

Z.; Gritzo, H.; Schro¨der, L.; Damrau, H.-R. H.; Wieser, U.; Schaper, F.;

Brintzinger, H. H. Organometallics 2005, 24, 894.

(96) Bryliakov, K. P.; Talsi, E. P.; Bochmann, M. Organometallics 2004,

23, 149.

(97) Petros, R. A.; Norton, J. R. Organometallics 2005, 23, 5105.

(98) Britovsek, G. J. P.; Cohen, S. A.; Gibson, V. C.; van Meurs, M. J.

Am. Chem. Soc. 2004, 126, 10701.

(99) Vollmerhaus, R.; Rahim, M.; Tomaszewski, R.; Xin, S.; Taylor, N.

J.; Collins, S. Organometallics 2000, 19, 2161.

(100) Sealey, A. J. D. D.Phil. Thesis, University of Oxford, 2004.

(101) Shreve, A. P.; Mulhaupt, R.; Fultz, W.; Calabrese, J.; Robbins,

W.; Ittel, S. D. Organometallics 1988, 7, 409.

(102) Healy, M. D.; Wierda, D. A.; Barron, A. R. Organometallics 1988,

7, 2543.

13, 14). The productivities (1230 and 690 kg mol^-1h^-1bar^-1

,

respectively) were only ca. 1% of that for Ti(N^tBu)(Me₃[9]-

aneN₃)Cl₂(1) and were also remarkably poor in comparison to

those of Ic and Id (133 700 and 152 500 kg mol^-1h^-1bar^-1

,

respectively) under analogous conditions. The PE formed with

57 had a higher M_wthan for 1 but featured a slightly broadened

PDI (4.4); that for 58 was clearly bimodal (overall PDI ) 24).

The polymerization results for 57 and 58 confirm the general

trend found for the mono- or di-ortho-substituted phenyl imides

2, 3, and 40-56 (Figure 3). In particular, the low productivity

of 57 parallels that of Ti(N-2,6-C₆H₃(F)CF₃)(Me₃[9]aneN₃)Cl₂

(51), which yielded no PE when assessed using the parallel

microreactors.

(103) Williams, V. C.; Dai, C.; Li, Z.; Collins, S.; Piers, W. E.; Clegg,

W.; Elsegood, M. R. J.; Marder, T. B. Angew. Chem., Int. Ed. 1999, 38,

3695.

(104) Williams, V. C.; Irvine, G. J.; Piers, W. E.; Li, Z.; Collins, S.;

Clegg, W.; Elsegood, M. R.; Marder, T. B. Organometallics 2000, 19, 1619.

(105) Metcalfe, R. A.; Kreller, D. I.; Tian, J.; Kim, H.; Taylor, N. J.;

Corrigan, J. F.; Collins, S. Organometallics 2002, 21, 1719.

(106) Busico, V.; Cipullo, R.; Cutillo, F.; Friederichs, N.; Ronca, S.;

Wang, B. J. Am. Chem. Soc. 2003, 125, 12402.

Effect of Added 2,6-Di-tert-butyl-4-methyl Phenol (BHT-

H) and Its Reactions with Imido Titanium Alkyl Complexes.

Commercial MAO typically contains significant amounts of

“free” AlMe₃(actually Al₂Me₆), which may act as both a chain

transfer agent (“chain transfer to aluminium”) and inhibitor of

catalyst activity through formation of adducts of the type [(L)M-

(µ-Me)(µ-R)AlMe₂]⁺(R ) Me or polymeryl chain) with the

3898 Organometallics, Vol. 25, No. 16, 2006

Adams et al.

Table 4. Ethylene Polymerization Data for the Ti(N^tBu)(Me₃[9]aneN₃)Cl₂(1)/MAO/BHT-H Catalyst System (100 °C)^a

estimated BHT-H:total Al

ratio (BHT-H:AlMe₃

ratio)

vinyl end

groups per

chain^b

internal

unsaturations

per chain^b

total

unsaturations

per chain

yield of

PE (g)

productivity (kg

M_w

M_n

entry

mol^-1h^-1bar^-1

)

(g mol^-1

)

(g mol^-1

)

M_w/M_n

1

2

3

4

0.0 (0)

0.4 (2)

0.8 (4)

1.2 (6)

5.63

5.46

8.48

4.41

96 600

93 600

145 400

75 400

100 000

120 000

235 000

280 000

37 000

40 000

67 000

125 000

2.7

3.0

3.5

2.2

0.37

0.40

0.53

0.91

0

0.37

0.46

0.77

1.09

0.06

0.24

0.16

^aConditions: (A) 7 bar C₂H₄on demand, 1000 mL of heptanes, Al:Ti ratio 1600, 0.05 µmol of 1, polymerization run time 10 min, BHT-H and MAO

premixed in reactor for 10 min prior to addition of 1 at 100 °C. ^bDetermined by NMR and/or FTIR spectroscopy.

and/or average polymer molecular weight compared with

nonmodified MAO alone.

unchanged compared to that without BHT-H added (690 kg

mol^-1h^-1bar^-1).

Reactions of Imido Titanium Compounds with BHT-H.

Previous work has shown that five-coordinate tert-butyl imido

titanium compounds (L)TiN^tBu can react readily with phenols

Table 4 summarizes analogous experiments carried out on

the Ti(N^tBu)(Me₃[9]aneN₃)Cl₂(1)/MAO catalyst system. Entry

1 (which is the same as entry 5 in Table 3) uses “as supplied”

MAO (Crompton, in which ca. 20% of the Al present is in the

form of AlMe₃). Entries 2-4 report the effect of premixing the

MAO with the stated amount of BHT-H for 10 min prior to

injection of 1. We address different aspects of the data in turn.

With regard to productivity, addition of 0.4 equiv (per total

Al present) of BHT-H gave negligible change. However,

addition of 0.8 and 1.2 equiv gave first an increase and then a

decrease in productivity. The PDI values followed a similar

trend, first increasing and then decreasing. In contrast, the total

average number of unsaturations per chain (and also the number

of vinyl end groups per chain) increases steadily with added

BHT-H, as do the molecular weights. These general features

may be readily interpreted. The increase in M_wand average

number of unsaturations per chain is consistent with decreasing

chain transfer to aluminum as “free” AlMe₃is scavenged. The

variation in PDI is also explained in this way, with lowest values

for 0 equiv and 1.2 equiv added BHT-H, reflecting first one

dominant chain transfer mechanism (i.e., chain transfer to

aluminum) and then another (chain transfer to monomer). The

decrease in productivity in entry 4 may indicate that addition

of an excess of BHT-H has a detrimental effect on either the

precatalyst and/or the activated species, or BHT-H may react

with Al-Me groups of MAO itself, therefore diminishing its

ability to activate the dichloride precatalyst.¹⁰⁶It may also/

alternatively indicate less efficient scavenging of impurities in

this system since AlMe₃is more effective in this regard than

MeAl(BHT)₂.¹⁰³Experiments to probe the underlying reaction

chemistry of BHT-H with relevant neutral and cationic tert-

butyl imido titanium compounds are described below. Interest-

ingly, the systems presumably containing the lowest amounts

of “free” AlMe₃(entries 3 and 4) appear to contain more internal

unsaturations per chain than in the other cases. The NMR data

were unclear regarding the exact nature of these unsaturations,

which may arise from “chain walking” for example. It is also

interesting that while the experiment described in entry 2 (ratio

of estimated “free” BHT-H to AlMe₃) 2) has, in principle,

sufficient BHT-H to scavenge all of the “free” AlMe₃, the

beneficial effect on productivity continues in entry 3, and the

increases in M_wand chain unsaturations continue through to

entry 4. This may be associated with the ability of BHT-H to

react with Al-Me bonds of MAO itself (or AlMe₃gradually

released from it).¹⁰⁶

t

to give the corresponding bis(aryloxides) (L)Ti(OAr)₂and -

BuNH₂.^33,107,108Furthermore, Collins has shown that BHT-H

reacts slowly with [Cp₂ZrMe]⁺at room temperature (t_1/2≈ 2

h) in C₆D₅Br to form [Cp₂Zr(BHT)]⁺.¹⁰⁹Both types of proto-

nolysis reaction would help account for the diminished pro-

ductivity in entry 4 of Table 4. We therefore examined the

reactions of the following species^41-43with BHT-H at room

temperature and/or 70 °C: Ti(N^tBu)(Me₃[9]aneN₃)Cl₂(1, the

primary catalyst precursor), Ti(N^tBu)(Me₃[9]aneN₃)Me₂(a likely

dialkyl product of the reaction of 1 with MAO), [Ti(N^tBu)-

(Me₃[9]aneN₃)Me]⁺(a likely first-formed catalytically active

species), and [Ti(N^tBu)(Me₃[9]aneN₃)(µ-Me)₂AlMe₂]⁺(II⁺, a

model for the AlMe₃-trapped active species resting state).

No reaction was observed between BHT-H and Ti(N^tBu)-

(Me₃[9]aneN₃)Cl₂(1) either at room temperature or at 70

°C. In contrast, the dimethyl derivative⁴³Ti(N^tBu)(Me₃[9]-

aneN₃)Me₂reacted completely within 10 min at room temper-

ature. The only identifiable product of this reaction was free

Me₃[9]aneN₃. Notably, ^tBuNH₂was not one of the products. It

appears that the tert-butyl group provides significant steric

protection of the TidN^tBu bond and that it is the Ti-Me bonds

that undergo attack by BHT-H.

The reaction between BHT-H and the in situ-generated

cation^41,42[Ti(N^tBu)(Me₃[9]aneN₃)Me]⁺(in C₆D₅Br) was slow

at room temperature, and after 1 h the spectrum still showed

resonances mainly for [Ti(N^tBu)(Me₃[9]aneN₃)Me]⁺and BHT-

H. However, heating at 70 °C for 1 h afforded a single new C_s

symmetric species, and subsequent scale-up gave [Ti(N^tBu)-

(Me₃[9]aneN₃)(BHT)][BAr^F] (60-BAr^F) as a fully character-

4

ized yellow powder in 68% isolated yield (eq 1).

1

The H NMR spectrum of 60⁺featured resonances for a

triazacyclononane ligand in a C_ssymmetric complex along with

those for a single BHT and N^tBu ligand. Although 60⁺is a

five-coordinate cation, at room temperature the ¹H NMR

spectrum shows no evidence for any fluxional behavior. This

Encouraged by the improved productivity of the Ti(N^tBu)-

(Me₃[9]aneN₃)Cl₂(1)/MAO catalyst system in the presence of

t

BHT-H, we repeated the screening of Ti(N-2-C₆H₄Bu)(Me₃-

(107) Swallow, D.; McInnes, J. M.; Mountford, P. J. Chem. Soc., Dalton

Trans. 1998, 2253.

(108) Dubberley, S. R.; Friedrich, A.; Willman, D. A.; Mountford, P.;

Radius, U. Chem. Eur. J. 2003, 9, 3634.

(109) Stapleton, R. A.; Galan, B. R.; Collins, S.; Simons, R. S.; Garrison,

J. C.; Youngs, W. J. J. Am. Chem. Soc. 2003, 125, 9246.

[9]aneN₃)Cl₂(58) under the same conditions as entry 14, Table

3, but this time pretreating the MAO with BHT-H (total Al:

BHT-H ) 1.0). Disappointingly, the activity was still very low

(710 kg mol^-1h^-1bar^-1) compared to that of 1 and essentially

Imido Titanium Ethylene Polymerization Catalysts

Organometallics, Vol. 25, No. 16, 2006 3899

is in contrast to the related five-coordinate cations [Ti(N^tBu)-

(Me₃[9]aneN₃)R]⁺(R ) Me, CH₂SiMe₃, or Cl)^41,42and is

probably due to the steric bulk of the tert-butyl groups of the

aryloxide ligand. As mentioned, Collins and co-workers have

reported the zirconocenium analogue of 60⁺, namely, [Cp₂Zr-

(BHT)]⁺. The X-ray crystal structure of [Cp₂Zr(BHT)]⁺featured

a Zr‚‚‚H-C agostic interaction between one of the ortho tert-

butyl substituents and the metal center. Unfortunately, we were

Al-Me signals as the rate of methyl group exchange increased,

but the proposed (entropically disfavored) equilibrium shifted

more to the left-hand side. The lack of a significant interac-

tion between [Ti(N^tBu)(Me₃[9]aneN₃)Me]⁺and MeAl(BHT)₂

contrasts strongly with that for AlMe₃, in which the adduct

[Ti(N^tBu)(Me₃[9]aneN₃)(µ-Me)₂AlMe₂]⁺(II⁺) is instanta-

neously formed in quantitative yield.⁴¹The NMR experiments

therefore support the view that there is limited interaction

between [Ti(N^tBu)(Me₃[9]aneN₃)Me]⁺and MeAl(BHT)₂in the

catalyst systems described in Table 4.

unable to obtain diffraction-quality crystals of 60-BAr^F. Low-

4

1

temperature H NMR spectroscopy was used to seek evidence

for an agostic interaction in 60⁺. No decoalescence of any of

the resonances was observed down to a temperature of -80

°C. Furthermore, there was no significant broadening of the

singlet at 1.58 ppm for the tert-butyl protons of the aryloxide

ligand relative to the other resonances. No evidence for an

agostic interaction¹¹⁰was detected in the expected region

Conclusions

An extensive range of triazacycle-supported imido titanium

compounds have been prepared and structurally characterized,

and their ethylene polymerization capability has been assessed.

Compounds with the smaller triazacyclohexane ligand, as well

as polystyrene-supported systems, showed negligible activity.

Compounds with the larger Me₃[9]aneN₃ligand showed the best

productivities at room temperature for both bulky alkyl- and

aryl-imido ligands. Assessment of a library of Me₃[9]aneN₃-

supported compounds at 100 °C found that only compounds

with bulky alkyl imido ligands gave the highest productivities

at this temperature. Under the particular MAO-activation

conditions assessed, the catalysts derived from Ti(NR)(Me₃[9]-

aneN₃)Cl₂exhibit two types of chain transfer mechanism. Use

of BHT-H scavenger in appropriate quantities increased the

productivity and PE molecular weights associated with the

catalyst system Ti(N^tBu)(Me₃[9]aneN₃)Cl₂/MAO, and mainly

PE terminated by â-hydrogen transfer to monomer was produced

with the highest amounts of added BHT-H. The underlying

organometallic chemistry in these systems was probed, and no

reaction between BHT-H and the imido ligand was observed.

The present data suggest that a primary role of the imido

N-substituent in these catalyst systems (and also the comple-

mentary Ti(NR){HC(Me₂pz)₃}Cl₂(I)/MAO⁴⁰) is to protect the

TidNR linkage and/or titanium center from attack by AlMe₃.

Bulkier N-R groups might also prevent homo-dimerization by

low-coordinate monoalkyl cations and/or inhibit attack at the

TidNR bond by the ethylene substrate itself, a process re-

cently proposed on the basis of DFT calculations to occur in

the Cr(NR)₂Cl₂/MAO catalyst system.^111,112DFT and experi-

mental studies of the reactions of well-defined imido titanium

alkyl cations with AlMe₃, olefins, and other unsaturated

substrates are underway and will be reported in due course.

(2800-2000 cm^-1) of the IR spectrum of 60-BAr^F.

4

In a final experiment BHT-H (2 equiv) was added to the

AlMe₃adduct⁴¹[Ti(N^tBu)(Me₃[9]aneN₃)(µ-Me)₂AlMe₂]⁺in

C₆D₅Br. The ¹H NMR spectrum was monitored at room

temperature over the course of 2 h, after which time the

following species were present in the relative approximate

proportions indicated: [Ti(N^tBu)(Me₃[9]aneN₃)(µ-Me)₂AlMe₂]⁺

(II⁺, 1.0), [Ti(N^tBu)(Me₃[9]aneN₃)Me]⁺(2.5), [Ti(N^tBu)(Me₃-

[9]aneN₃)(BHT)]⁺(60⁺, 1.0), BHT-H (1.8), MeAl(BHT)₂(3.7).

It therefore appears that BHT-H can trap the (previously)

coordinated AlMe₃of [Ti(N^tBu)(Me₃[9]aneN₃)(µ-Me)₂AlMe₂]⁺

as MeAl(BHT)₂to liberate [Ti(N^tBu)(Me₃[9]aneN₃)Me]⁺. How-

ever, this species in turn can react slowly with BHT-H to give

60⁺.

The reactions of these various neutral and cationic imido

titanium species with BHT-H indicate quite clearly that control

of the amount of BHT-H present in the catalyst system, and

the order in which the BHT-H and titanium precatalyst are

added, is critical with regard to obtaining optimum levels of

“free” monoalkyl cation active species.

Since the data in Table 4 imply that the BHT-modified

AlMe₃is less able to stabilize the cationic monoalkyl active

species in polymerization, the interaction of in situ-generated

Experimental Section

General Methods and Instrumentation. All operations de-

scribed herein are air- and moisture-sensitive and were carried out

using standard Schlenk-line (Ar) and drybox (N₂) techniques. Protio-

and deutero-solvents were purified, dried, and distilled using

conventional techniques. NMR samples were prepared under

dinitrogen in Wilmad 505-PS tubes fitted with J. Young NMR/5

valves. Samples for magic angle spinning (MAS) NMR spectros-

[Ti(N^tBu)(Me₃[9]aneN₃)Me][BAr^F] with “noninteracting” MeAl-

4

(BHT)₂^101,102was probed by NMR spectroscopy (C₆D₅Br). The

1

resultant H NMR spectrum showed resonances for the indi-

vidual species [Ti(N^tBu)(Me₃[9]aneN₃)Me]⁺and MeAl(BHT)₂,

except that those for the Ti-Me (δ 0.66 ppm) and Al-Me (δ

-0.43 ppm) groups were significantly broadened. These ob-

servations are consistent with the dynamic process shown in

eq 2, in which the equilibrium lies well to the left-hand side.

Cooling the NMR sample in an attempt to shift the equilibrium

in favor of the adduct 61⁺and slow the dynamic process caused

1

copy were prepared in 2 mm Zirconia rotors. Solution phase H

and ¹³C{¹H} NMR spectra were recorded on Bruker DPX 300,

Varian Unity Plus 500, and Varian Mercury-VX 300 spectrometers

at 298 K unless otherwise stated. Spectra were referenced internally

to residual protio-solvent (¹H) or solvent (¹³C) resonances and are

reported relative to tetramethylsilane (δ ) 0 ppm). ¹⁹F spectra were

externally referenced to CFCl₃(δ ) 0 ppm). Chemical shifts are

[Ti(N^tBu)(Me₃[9]aneN₃)Me][BAr^F] to precipitate, and at 0 °C

4

only resonances for MeAl(BHT)₂were visible. Warming the

sample from 20 to 100 °C led first to increased broadening,

and then (after ca. 60 °C) sharpening, of the Ti-Me and

quoted in δ (ppm) and coupling constants in Hz. H and ¹³C{¹H}

1

(110) Brookhart, M.; Green, M. L. H.; Wong, L.-L. Prog. Inorg. Chem.

1988, 36, 1.

(111) Jensen, V. R.; Børve, K. J. Organometallics 2001, 20, 616.

(112) Jensen, V. R.; Børve, K. J. Chem. Commun. 2002, 542.

3900 Organometallics, Vol. 25, No. 16, 2006

Adams et al.

MAS NMR spectra were recorded on a Chemagnetics 400 Infinity

spectrometer. Spectra were referenced internally to tetramethylsi-

lane. IR spectra were recorded on Perkin-Elmer 1600 and Perkin-

Elmer 1710 spectrometers as Nujol mulls between CsBr, KBr, or

NaCl windows unless otherwise stated, and data are quoted in

wavenumbers (cm^-1). Mass spectra were recorded on a Micromass

GCT TOF instrument using a solid probe inlet temperature program,

and data are quoted in mass/charge ratio (m/z). Elemental analyses

were carried out by the analytical laboratory of the School of

Chemistry, University of Nottingham, Inorganic Chemistry Labora-

tory, University of Oxford, or by the Elemental Analysis Service

at the London Metropolitan University. GPC analyses were carried

out by Rapra Technology Ltd. in 1,2,4-trichlorobenzene solvent at

160 °C relative to polystyrene calibrants and using the appropriate

Mark Houwink parameters for polyethylene. IR and NMR analyses

of polymer samples were performed by the analytical services

department of DSM Research or SABIC Europe, Research &

Development.

to a stirred solution of Ti(N-2,6-C₆H₃Me₂)cl₂(py)₃(2.376 g, 5.00

mmol) in CH₂Cl₂(25 mL). The resulting dark brown solution was

stirred for 20 h, and the volatiles were removed under reduced

pressure. The product was extracted into CH₂Cl₂(15 mL) and then

cautiously precipitated by the slow addition of hexane (30 mL)

with vigorous stirring. The resulting brown precipitate was filtered

off, washed with hexanes (5 mL), and dried in vacuo to give a

brown powder. Yield: 1.76 g (86%). ¹H NMR (300.1 MHz,

3

CDCl₃): 6.76 (2 H, d, J ) 7.4 Hz, meta-C₆H₃Me₂), 6.52 (1 H, t,

³J ) 7.4 Hz, para-C₆H₃Me₂), 3.91 (2 H, m, CH₂ring), 3.34 (2 H,

m, CH₂ring), 3.33 (6 H, s, NMe cis to NAr), 3.12 (2 H, m, CH₂

ring), 3.10 (2 H, m, CH₂ring), 2.78 (6 H, s, C₆H₃Me₂), 2.76 (2 H,

m, CH₂ring), 2.70 (3 H, s, NMe trans to NAr), 2.55 (2 H, m, CH₂

ring). ¹³C{¹H} NMR (75.5 MHz, CDCl₃): 148.2 (ipso-C₆H₃Me₂),

133.6 (ortho-C₆H₃Me₂), 127.9 (meta-C₆H₃Me₂), 121.0 (para-C₆H₃-

Me₂), 57.8 (CH₂ring), 57.2 (CH₂ring), 54.7 (CH₂ring), 54.3 (NMe

cis to NAr), 49.2 (NMe trans to NAr), 22.1 (C₆H₃Me₂). Anal. Found

(calcd for C₁₇H₃₀Cl₂N₄Ti): C 49.7, (49.9); H, 7.5 (7.4); N, 13.6

(13.7).

Starting Materials. The compounds Me₃[9]aneN₃,¹¹³Ti(NR)-

i

Cl₂(py)₃(R ) ^tBu, 2,6-C₆H₃Me₂, 2,6-C₆H₃Pr₂),⁴⁴Ti(NR)Cl₂-

i

Ti(N-2,6-C₆H₃Pr₂)(Me₃[9]aneN₃)Cl₂(3).

A solution of

t

(NHMe₂)₂(R ) Bu, Ph, C₆F₅),⁴⁷Ti(NMe₂)₂Cl₂,¹¹⁴Ti(NR)(Me₃-

Me₃[9]aneN₃(0.856 g, 5.00 mmol) in CH₂Cl₂(5 mL) was added

i

[6]aneN₃)Cl₂(R ) ^tBu, 2,6-C₆H₃Pr₂),³⁴and MeAl(BHT)₂¹⁰²were

i

to a stirred solution of Ti(N-2,6-C₆H₃Pr₂)cl₂(py)₃(2.657 g, 5.00

prepared according to the literature methods. Liquid primary amines

for preparative scale experiments were predried over freshly ground

CaH₂and distilled prior to use. BHT-H was recrystallized from

hexanes. MAO was obtained from Albermarle and Crompton. All

other compounds and reagents were purchased and used without

further purification.

Ti(N^tBu)(Me₃[9]aneN₃)Cl₂(1). From Ti(N^tBu)Cl₂(py)₃. A

solution of Me₃[9]aneN₃(0.110 g, 0.642 mmol) in CH₂Cl₂(5 mL)

was added to a stirred solution of Ti(N^tBu)cl₂(py)₃(0.273 g, 0.639

mmol) in CH₂Cl₂(15 mL). The initially deep orange solution

became paler over 2 h, after which the volatiles were removed under

reduced pressure to give an orange solid, which was washed with

pentane (2 × 5 mL) and dried in vacuo. The crude material was

dissolved in CH₂Cl₂(8 mL) and the solution cooled to -25 °C for

3 days, which gave 1 as orange crystals, which were filtered off

and dried in vacuo. Yield: 0.196 g (85%).

mmol) in CH₂Cl₂(20 mL). The resulting dark brown solution

was stirred for 14 h, and the volatiles were removed under re-

duced pressure. The product was extracted into CH₂Cl₂(15 mL)

and then cautiously precipitated by the slow addition of hexane

(35 mL) with vigorous stirring. The resulting microcrystalline green-

brown precipitate was filtered off, washed with hexanes (5 mL),

and dried in vacuo to give a green-brown powder. Yield: 2.02 g

(87%). ¹H NMR (300.1 MHz, CDCl₃): 6.90 (2 H, d, ³J ) 7.6 Hz,

i

meta-C₆H₃Pr₂), 6.72 (1 H, t, ³J ) 7.6 Hz, para-C₆H₃Pr₂), 4.51 (2

H, sept, ³J ) 6.9 Hz, CHMe₂), 3.74 (2 H, m, CH₂ring), 3.33 (2 H,

m, CH₂ring), 3.27 (6 H, s, NMe cis to NAr), 3.16 (2 H, m, CH₂

ring), 3.11 (2 H, m, CH₂ring), 2.77 (2 H, m, CH₂ring), 2.69 (3 H,

s, NMe trans to NAr),2.56 (2 H, m, CH₂ring), 1.25 (12 H, d, ³J )

6.9 Hz, CHMe₂). ¹³C{¹H} NMR (75.5, CDCl₃, MHz): 155.7 (ipso-

i

C₆H₃Pr₂), 145.0 (ortho-C₆H₃Pr₂), 122.8 (meta-C₆H₃Pr₂), 121.3

i

(para-C₆H₃Pr₂), 58.3 (CH₂ring), 57.3 (CH₂ring), 54.4 (CH₂ring),

(b) From Ti(NMe₂)₂Cl₂. To a deep orange solution of

Ti(NMe₂)₂Cl₂(0.50 g, 2.4 mmol) in C₆H₆(10 mL) was added

^tBuNH₂(0.25 mL, 2.4 mmol). Immediately after addition, the

orange solution darkened and subsequently lightened over the space

of several minutes to give a lemon yellow solution. After 16 h the

solution was filtered and Me₃[9]aneN₃(0.41 g, 2.4 mmol) was

added, and the mixture was left to stir for another 16 h. Soon after

addition of the ligand, a yellow solid began to precipitate. The

solvent volume was reduced by two-thirds, and the solids were

filtered off and washed with hexanes (3 × 10 mL). The resulting

yellow powder was dried in vacuo. Yield: 0.73 g (85%). ¹H NMR

(300.1 MHz, CDCl₃): 3.61 (2 H, m, CH₂ring), 3.30 (6 H, s, NMe

trans to N^tBu), 3.18 (2 H, m, CH₂ring), 3.10 (2 H, m, CH₂ring),

2.94 (2 H, m, CH₂ring), 2.80 (2 H, m, CH₂ring), 2.57 (2 H, m,

CH₂ring), 2.51 (6 H, s, NMe cis to N^tBu), 1.13 (9 H, s, ^tBu). ¹³C-

{¹H} NMR (75.5 MHz, CDCl₃): 69.2 (NCMe₃), 56.7 (CH₂ring),

56.4 (CH₂ring), 54.2 (CH₂ring). 53.8 (NMe cis to N^tBu), 48.1

(NMe trans to N^tBu), 30.9 (NCMe₃). IR (CsBr plates, Nujol, cm^-1):

1498(w), 1462 (s), 1377 (m), 1351 (w), 1298 (w), 1244 (m), 1206

(w), 1155 (w), 1120 (w), 1074 (w), 1065 (w), 1002 (m), 892 (w),

784 (m), 747 (w), 730 (w), 598 (w), 584 (w), 536 (w), 438 (w),

421 (w). EI-MS: m/z ) 345 (75%) [M - Me]⁺, 289 (100%) [M

- N^tBu]⁺. Anal. Found (calcd for C₁₃H₃₀Cl₂N₄Ti): C, 43.1 (43.2);

H, 8.4 (8.4); N, 15.4 (15.5).

54.0 (NMe cis to NAr), 49.1 (NMe trans to NAr), 27.4 (CHMe₂),

25.3 (CHMe₂). Anal. Found (calcd for C₂₁H₃₈Cl₂N₄Ti): C, 53.9

(54.2); H, 8.4 (8.2); N, 12.0 (12.0).

Ti(NPh)(Me₃[9]aneN₃)Cl₂(4). To a solution of Ti(NPh)Cl₂-

(NHMe₂)₂(0.4 g, 1.33 mmol) in C₆H₆(5 mL) was added Me₃[9]-

aneN₃(0.26 mL, 1.33 mmol). Soon after addition, an orange solid

began to precipitate. The mixture was stirred for 16 h, after which

time the solids were filtered off, washed with hexanes (3 × 10

1

mL), and dried in vacuo. Yield: 0.50 g (99%). H NMR (500.0

3

MHz, CD₂Cl₂): 7.06 (2 H, app t, app J ) 8.2 Hz, meta-C₆H₅),

6.92 (2 H, d, ³J ) 8.8 Hz, ortho-C₆H₅), 6.75 (1 H, t, ³J ) 7.7 Hz,

para-C₆H₅), 3.48 (2 H, m, CH₂ring), 3.23 (2 H, m, CH₂ring),

3.21 (6 H, s, NMe cis to NPh), 3.02 (4 H, overlapping 2 × m, CH₂

ring), 2.78 (2 H, m, CH₂ring), 2.64 (3 H, s, NMe trans to NPh),

2.61 (2 H, m, CH₂ring). ¹³C{¹H} NMR (125.7 MHz, CD₂Cl₂):

128.6 (ipso-C₆H₅), 128.5 (meta-C₆H₅), 123.8 (para-C₆H₅), 121.4

(ortho-C₆H₅), 57.3 (CH₂ring), 56.9 (CH₂ring), 55.2 (CH₂ring),

53.6 (NMe cis to NPh), 49.2 (NMe trans to NPh). IR (KBr pellet,

cm^-1): 3064 (w), 3004 (w), 2928 (s), 2912 (s), 2858 (s), 2844 (s),

1576 (m), 1496 (m), 1294 (w), 1282 (w), 1280 (m), 1162 (w), 1072

(m), 1064 (m), 1008 (m), 962 (w), 896 (w), 788 (m), 772 (s), 750

(w), 670 (w), 584 (w), 520 (m), 442 (w), 426 (w). EI-MS: m/z )

380 (50%) [M]⁺, 288 (14%) [M - NPh - H]⁺. Anal. Found (calcd

for C₁₅H₂₆Cl₂N₄Ti): C, 47.6 (47.3); H, 7.3 (6.9); N, 14.4 (14.7).

Ti(N-2,6-C₆H₃Me₂)(Me₃[9]aneN₃)Cl₂(2).

A solution of

Ti(NC₆F₅)(Me₃[9]aneN₃)Cl₂(5). To a solution of Ti(NC₆F₅)-

Cl₂(NHMe₂)₂(0.62 g, 1.59 mmol) in C₆H₆(10 mL) was added

Me₃[9]aneN₃(0.31 mL, 1.59 mmol). A yellow solid precipitated

from solution soon after addition. The mixture was stirred for 16

h, after which time the solvent volume was reduced by two-thirds,

Me₃[9]aneN₃(0.856 g, 5.00 mmol) in CH₂Cl₂(5 mL) was added

(113) Madison, S. A.; Batal, D. J. (Unilever PLC, UK.; Unilever N. V.)

United States Patent 5,284,944, 1994.

(114) Froneman, M.; Cheney, D. L.; Modro, T. L. Phosphorus, Sulfur,

Silicon Relat. Elem. 1990, 47.

Imido Titanium Ethylene Polymerization Catalysts

Organometallics, Vol. 25, No. 16, 2006 3901

and the solids were filtered off, washed with hexanes (3 × 10 mL),

reduced pressure was applied to the oil, an orange powder was

1

and dried in vacuo. Yield: 0.59 g (79%). H NMR (300.1 MHz

obtained. Yield: 127 mg (68%). H NMR (499.9 MHz, CD₂Cl₂):

t

CD₂Cl₂): 3.56 (2 H, m, CH₂ring), 3.21 (6 H, overlapping 3 × m,

CH₂ring), 3.15 (6 H, s, NMe cis to NC₆F₅), 3.01 (2 H, m, CH₂

ring), 2.75 (2 H, m, CH₂ring), 2.74 (3 H, s, NMe trans to NC₆F₅).

¹⁹F NMR (282.4 MHz, CD₂Cl₂): -152.1 (2 F, unresolved m, ortho-

C₆F₅), -167.6 (2 F, unresolved m, meta-C₆F₅), -169.6 (1 F,

unresolved m, para-C₆F₅). ¹³C{¹H} NMR (75.5 MHz, CD₂Cl₂):

57.4 (NMe cis to NC₆F₅), 55.4 (overlapping, 2 x CH₂ring), 53.5

(NMe trans to NC₆F₅), 49.7 (CH₂ring), the C₆F₅ring carbons were

not observed due to ¹⁹F coupling. IR (KBr plates, Nujol, cm^-1):

1320 (w), 1292 (w), 1262 (w), 1206 (w), 1156 (w), 1053 (m), 1038

(m), 1000 (m), 978 (m), 892 (w), 790 (m), 750 (w), 722 (m), 580

(br, w), 500 (w), 462 (w), 448 (w), 426 (m), 416 (w). EI-MS: m/z

) 470 (45%) [M]⁺, 435 (7%) [M - Cl]⁺. Anal. Found (calcd for

C₁₅H₂₁Cl₂F₅N₄Ti): C, 38.3 (38.3); H, 4.7 (4.5); N, 11.7 (11.9).

7.10 (2 H, C₆H₂Bu₂Me), 4.09 (2 H, m, CH₂ring), 3.38 (2 H, m,

CH₂ring), 3.16 (6 H, s, NMe cis to N^tBu), 3.01 (2 H, m, CH₂

ring), 2.85 (2 H, m, CH₂ring), 2.74 (2 H, m, CH₂ring), 2.63 (2 H,

m, CH₂ring), 2.41 (3 H, s, NMe trans to N^tBu), 2.26 (3 H, s,

t

C₆H₂Bu₂Me), 1.58 (18 H, s, C₆H₂Bu₂Me), 0.79 (9 H, s, Bu).

t

¹³C{¹H} NMR (75.4 MHz, CD₂Cl₂): 162.7 (ipso-C₆H₂Bu₂Me),

1

148.5 (br d, J_C-F) 238 Hz, ortho-C₆F₅), 138.6 (br d, J_C-F

)

247 Hz, para-C₆F₅), 136.7 (br d, ¹J_C-F) 250 Hz, meta-C₆F₅), 135.1

t

(ortho-C₆H₂Bu₂Me), 127.8 (para-C₆H₂Bu₂Me), 126.7 (meta-

t

C₆H₂Bu₂Me), 72.7 (NCMe₃), 56.8 (CH₂ring), 56.6 (CH₂ring),

t

54.0 (CH₂ring), 53.7 (NMe cis to Bu), 49.0 (NMe trans to Bu),

35.4 (CCMe₃), 31.6 (NCMe₃), 31.4 (CCMe₃). ¹⁹F NMR (282.1

3

MHz, CD₂Cl₂): -133.5 (d, J ) 10.6 Hz, ortho-C₆F₅), -164.0 (t,

³J ) 20.4 Hz, para-C₆F₅), -167.9 (app t, app ³J ) 18.1 Hz, meta-

C₆F₅). IR (NaCl plates, Nujol, cm^-1): 1643 (m), 1514 (s), 1420

(m), 1265 (m), 1239 (m), 1216 (m), 1087 (s), 1001 (m), 980 (s),

844 (w), 775 (m), 756 (m), 684 (m), 662 (m). Anal. Found (calcd

for C₅₂H₅₃BF₂₀N₄OTi): C 52.1 (52.5), H 4.4 (4.5), N 4.7 (4.7).

Ti(NCH₂Ph)(Me₃[9]aneN₃)Cl₂(6). To a solution of Ti(NMe₂)₂-

Cl₂(0.5 g, 2.4 mmol) in C₆H₆(5 mL) was added PhCH₂NH₂(0.25

mL, 2.4 mmol). Immediately after addition the orange color of the

solution darkened noticeably. The solution was stirred for 16 h,

and then Me₃[9]aneN₃(0.47 mL, 2.4 mmol) was added. Almost

immediately after addition, a pale yellow solid began to precipitate

from the solution. The mixture was stirred for another 16 h, the

solvent volume was reduced by two-thirds, and the solids were

filtered off, washed with hexanes (3 × 10 mL), and dried in vacuo

to yield pale cream colored 6. Yield: 0.86 g (96%). ¹H NMR (500.0

MHz, CD₂Cl₂): 7.53 (2 H, d, ³J ) 7.3 Hz, ortho-C₆H₅), 7.30 (2 H,

NMR Tube Scale Reaction of [Ti(N^tBu)(Me₃[9]aneN₃)Me]⁺

with MeAl(BHT)₂. To a solution of [Ti(N^tBu)(Me₃[9]aneN₃)Me]⁺

(generated in situ from Ti(N^tBu)(Me₃[9]aneN₃)Me₂(7.0 mg, 0.022

mmol) and [Ph₃C][BAr^F] (20.2 mg, 0.022 mmol)) in C₆D₅Br (0.75

4

mL) was added MeAl(BHT)₂(10.5 mg, 0.022 mmol). After 10 min

1

the H NMR spectrum showed broadened resonances for [Ti(N^t-

Bu)(Me₃[9]aneN₃)Me]⁺and MeAl(BHT)₂.

3

app t, app J ) 7.8 Hz, meta-C₆H₅), 7.17 (1 H, t, J ) 7.3 Hz,

para-C₆H₅), 4.53 (2 H, s, CH₂Ph), 3.31 (2 H, m, CH₂ring), 3.13

(2 H, m, CH₂ring), 3.06 (6 H, s, NMe cis to NCH₂Ph), 2.94 (4 H,

overlapping 2 × m, CH₂ring), 2.69 (2 H, m, CH₂ring), 2.50 (2 H,

m, CH₂ring), 2.49 (3 H, s, NMe trans to NCH₂Ph). ¹³C{¹H} NMR

(125.7 MHz, CD₂Cl₂): 142.0 (ipso-C₆H₅), 128.3 (meta-C₆H₅), 127.6

(ortho-C₆H₅), 126.4 (para-C₆H₅), 68.6 (CH₂Ph), 57.1 (CH₂ring),

56.8 (CH₂ring), 54.9 (CH₂ring), 53.5 (NMe cis to NCH₂Ph), 48.9

(NMe trans to NCH₂Ph). IR (KBr plates, Nujol, cm^-1): 3064 (w),

3020 (m), 3002 (m), 2958 (m), 2898 (s), 2866 (s), 2861 (s), 2768

(s), 1496 (vs), 1452 (vs), 1388 (w), 1336 (m), 1326 (m), 1298 (vs),

1244 (vs), 1186 (m), 1158 (m), 1124 (m), 1074 (vs), 1066 (vs),

1030 (m), 1004 (vs), 896 (s), 786 (vs), 744 (vs), 708 (vs), 690 (m),

618 (w), 586 (s), 458 (m), 426 (s), 442 (s), 412 (s). EI-MS: m/z )

395 (6%) [M⁺], 289 (100%) [M - CH₂Ph]⁺, 218 (78%) [M -

NCH₂Ph - 2 Cl]⁺. Anal. Found (calcd for C₁₆H₂₈Cl₂N₄Ti): C, 48.5

(48.7); H, 7.2 (7.2); N, 13.5 (14.2).

NMR Tube Scale Reaction of [Ti(N^tBu)(Me₃[9]aneN₃)(µ-

Me)₂AlMe₂]⁺(II⁺) with BHT-H (2 equiv). To a solution of

[Ti(N^tBu)(Me₃[9]aneN₃)(µ-Me)₂AlMe₂]⁺(II⁺) (generated in situ

from Ti(N^tBu)(Me₃[9]aneN₃)Me₂(7.0 mg, 0.022 mmol), [Ph₃C]-

[BAr^F] (20.2 mg, 0.022 mmol), and Al₂Me₆(2.1 µl, 0.022 mmol))

4

in C₆D₅Br (7.5 mL) was added BHT-H (9.6 mg, 0.044 mmol). The

¹H NMR spectrum was monitored over the course of 2 h, after

which time the following species were present in the relative

approximate proportions indicated: [Ti(N^tBu)(Me₃[9]aneN₃)-

(µ-Me)₂AlMe₂]⁺(II⁺, 1.0), [Ti(N^tBu)(Me₃[9]aneN₃)Me]⁺(2.5),

[Ti(N^tBu)(Me₃[9]aneN₃)(BHT)]⁺(60⁺, 1.0), BHT-H (1.8), MeAl-

(BHT)₂(3.7).

Procedure for Ti(NR)(Me₃[9]aneN₃)Cl₂Library Synthesis. In

preparation of the parallel synthesis procedure, stock solutions of

Ti(NMe₂)₂Cl₂(0.32 M), Me₃[9]aneN₃(1.06 M), and the amine

library (0.97 M) in benzene were prepared. Amines (1.94 mmol of

amine per vial) were purchased as “CombiKits” from Aldrich and

used as received. An M. Braun Labmaster glovebox was equipped

with a Tecan Miniprep liquid handling robot, two low-vortex flatbed

shakers, and a vacuum/N₂connection. The first step was carried

out in sealable wells having a maximum volume capacity of 10

mL. In the first step, 3 mL of the titanium amide stock solution

(corresponding to 200 mg of Ti(NMe₂)₂Cl₂, 0.97 mmol/well) was

dispensed into each well, followed by 1 mL of the appropriate amine

stock solution (corresponding to 0.97 mmol amine/well). The wells

were sealed with a Teflon cover to prevent evaporation of the

solvent and shaken for 16 h. After this time, the resulting solutions

of Ti(NR)Cl₂(NHMe₂)₂were transferred to preweighed solid phase

synthesis cartridges equipped with Teflon filters and housed in a

specially constructed holder. The holder allowed the application

of either an overpressure or a vacuum from underneath the

cartridges. After the solution transfer had been completed, the plates

were manually checked for signs of crystallization. If crystallization

had occurred, the crystals were gathered and recombined with the

corresponding solution. During the solution transfer, an overpressure

was applied from underneath the cartridges to keep the solutions

above the frit. Subsequently, 1 mL of Me₃[9]aneN₃stock solution

(containing 1.06 mmol of ligand) was dispensed into each cartridge

under agitation. Addition of the macrocycle typically resulted in

the formation of a thick precipitate of the corresponding Ti(NR)-

NMR Tube Scale Reaction of Ti(N^tBu)(Me₃[9]aneN₃)Cl₂with

BHT-H. Ti(N^tBu)(Me₃[9]aneN₃)Cl₂(8.0 mg, 0.022 mmol) and

BHT-H (4.9 mg, 0.022 mmol) were dissolved in CD₂Cl₂(0.75 mL).

¹H NMR spectra after 10 min and 1 h showed resonances corre-

sponding to unreacted Ti(N^tBu)(Me₃[9]aneN₃)Cl₂and BHT-H. After

1

the sealed NMR tube was heated to 70 °C for 1 h the H NMR

spectrum remained unchanged.

NMR Tube Scale Reaction of Ti(N^tBu)(Me₃[9]aneN₃)Me₂with

BHT-H. Ti(N^tBu)(Me₃[9]aneN₃)Me₂(2.5 mg, 0.008 mmol) and

BHT-H (1.7 mg, 0.008 mmol) were dissolved in C₆D₆(0.75 mL).

1

The H NMR spectrum after 10 min showed resonances corre-

sponding to free Me₃[9]aneN₃and a mixture of other unidentified

products. The ¹H NMR spectrum was unchanged after 1 h at room

temperature and after heating to 70 °C for 1 h.

[Ti(N^tBu)(Me₃[9]aneN₃)(BHT)][BAr^F] (60-BAr^F). To a solu-

4

tion of Ti(N^tBu)(Me₃[9]aneN₃)Me₂(50 mg, 0.156 mmol) in

C₆H₅Cl (2 mL) was added [Ph₃C][BAr^F] (144 mg, 0.156 mmol)

4

in C₆H₅Cl (2 mL) to give a bright orange solution. After 5 min

BHT-H (34 mg, 0.156 mmol) was added. The reaction mixture was

stirred for 1 h at 70 °C and then allowed to cool. The yellow solution

was concentrated to approximately 1 mL, then pentane (4 mL) was

added with stirring to give an orange oil. The supernatant was

decanted and the oil washed with pentane (3 × 2 mL). When a

3902 Organometallics, Vol. 25, No. 16, 2006

Adams et al.

Table 5. X-ray Data Collection and Processing Parameters for Ti(N^tBu)(Me₃[9]aneN₃)Cl₂‚CH₂Cl₂(1‚CH₂Cl₂) and

i

Ti(NR)(Me₃[9]aneN₃)Cl₂(R ) 2,6- C₆H₃Me₂(2), 2,6-C₆H₃Pr₂(3), Ph (4), C₆F₅(5))

parameter

1‚CH₂Cl₂

2

3

4

5

empirical formula

fw

temp/K

C₁₃H₃₀N₄Cl₂Ti‚CH₂Cl₂

446.15

150

C₁₇H₃₀Cl₂N₄Ti

409.26

150

C₂₁H₃₈N₄Cl₂Ti

465.35

150

C₁₅H₂₆Cl₂N₄Ti

381.21

150

C₁₅H₂₁Cl₂F₅N₄Ti

471.16

150

wavelength/Å

space group

a/Å

b/Å

c/Å

0.71073

Pbca

15.104(6)

14.515(6)

19.933(7)

90

4370(3)

8

1.36

0.71073

Pna2₁

17.327(4)

8.214(2)

13.868(4)

90

1974.0(6)

4

1.38

0.71073

Pbca

16.130(9)

16.049(6)

18.348(8)

90

4750(4)

8

1.302

0.71073

P2₁/c

7.3615(2)

15.1070(4)

16.0577(6)

90

93.061(1)

90

1783.2(1)

4

1.42

0.78

R₁) 0.0283

R_w) 0.0213

[I > 3σ(I)]

0.71073

P2₁/c

17.3911(2)

8.1056(1)

13.3658(2)

90

96.218(1)

90

1873.03(4)

4

1.67

0.80

R₁) 0.0262

R_w) 0.0262

[I > 3σ(I)]

R/deg

â/deg

γ/deg

V/Å³

Z

d(calcd)/Mg‚m^-1

abs coeff/mm^-1

R indices^a

0.88

0.71

0.60

R

_{1 )}0.0470

R

_{1 )}0.0319

R

_{1 )}0.0425

wR₂) 0.0672

[I > 2σ(I)]

R_w) 0.0310

[I > 2σ(I)]

wR₂) 0.106

[I > 2σ(I)]

2

^aR₁) ∑||F_o| - |F_c||/∑|F_o|; R_w) {∑w(|F_o| - |F_c|)²/∑w|F_o| }^1/2. wR₂) {∑w(F_o- F_c)²/∑w(F_o)²}^1/2

.

Cl₂(Me₃[9]aneN₃) complex. The cartridges were shaken for 2 h,

after which time the pressure underneath the cartridge was reversed

to remove the reaction solvent, which was caught in the plate

underneath. The resulting solids were washed with hexanes (3 × 1

mL), the cartridges stoppered, and the solids dried in vacuo. The

compounds were characterized by ¹H NMR spectroscopy and

EI-mass spectrometry (high resolution) where possible. Character-

izing data and yields are given in the Supporting Information. The

data for 1-4 and 6 were identical to those for samples prepared

conventionally using Schlenk techniques.

Typical Procedure for Ambient-Temperature Ethylene Po-

lymerization Studies. To a sealable metal reactor containing a glass

insert was added MAO (Albermarle, 20 mL, 10% in toluene, w/w,

30 mmol) made up to 200 mL with toluene. The solution was stirred

at 250 rpm for 5 min, and then the precatalyst (20 µmol) in toluene

(50 mL) was added to the reactor and the mixture stirred for a

further 30 min. The reaction vessel was placed under dynamic

vacuum for 10 s, the stirring was increased to 750 rpm, and a

dynamic pressure of ethylene (6 bar) was admitted. After 60 min,

the reactor was isolated and the pressure was released. Methanol

(5 mL) was added to the mixture, followed by water (50 mL) with

stirring. The mixture was acidified to pH ) 1 using a solution of

10% HCl (aq) in MeOH and stirred overnight. The precipitated

polymers were filtered, washed with water (total volume ca. 1000

mL), and dried to constant weight at room temperature. Analogous

procedures were used for the polystyrene-supported catalysts and

solution phase catalysts at 2.5 µmol precatalyst loadings.

Conditions for Parallel Screening of the Catalyst Library

Ti(NR)(Me₃[9]aneN₃)Cl₂. Proprietary equipment was used with

the following conditions: 0.16 µmol of precatalyst with 192 µmol

of MAO (Ti:Al ) 1:1200) in 62 mL of PMH (2,2,4,4,6-

pentamethylheptane); precatalyst injected at 100 °C and stirred

under 7 bar dynamic ethylene pressure; polymerization time 10 min.

Conditions for Ethylene Polymerization Studies Using Semi-

batch Reactors at 100 °C. MAO (Eurecen Al 5100/10T, 10% in

toluene) was obtained from Crompton and used as received. BHT-H

was obtained from Aldrich. The BHT-H was dissolved in toluene

and dried in vacuo and subsequently stored under nitrogen. Ethylene

and nitrogen are purified over deoxocatalyst (BTS catalyst BASF

AG) and 4 Å molecular sieves. Heptane was purified by degassing

with nitrogen and subsequently purified over 13× molecular sieves.

Toluene was purified by distilling from sodium/benzophenone.

equipped with baffles and two interMIG stirrers, operated at 1400

rpm. A 500 mL amount of heptanes was dosed into the autoclave.

Ethylene was dosed via a Brooks mass flow controller into the

headspace, and the reactor pressure was set at 8 bar. The temperature

was set at 100 °C. Subsequently the MAO was dosed together with

an additional 350 mL of heptanes. In case BHT was used, this was

premixed with the MAO in the reactor. After stirring the contents

of the reactor for 10 min at 100 °C, a solution of the catalyst in

approximately 1 mL of toluene was mixed with 50 mL of heptanes

and subsequently pumped into the reactor. The catalyst feeding

section was flushed with additional heptanes. After dosing the

catalyst components, the total volume of added heptanes was 1 L.

The reactor temperature was kept at 100 ( 2 °C by cooling with

an oil system. The reactor pressure was kept constant via supplying

ethylene. After 10 min, the mixture was drawn off via a bottom

valve. A mixture of 2-propanol and Irganox 1076 was added. The

reactor was rinsed with PMH at 150 °C until the washings were

visibly clear from polymer. The complete polymer slurry, including

the washings with PMH, was put in a vacuum oven and dried to

constant weight at 50 °C. To minimize mass transfer limitations,

the amount of catalyst was adjusted to obtain between 5 and 10 g

of polymer.

Crystal Structure Determinations of Ti(N^tBu)(Me₃[9]aneN₃)-

Cl₂‚CH₂Cl₂(1‚CH₂Cl₂), Ti(NR)(Me₃[9]aneN₃)Cl₂(R ) 2,6-

i

C₆H₃Me₂(2), 2,6-C₆H₃Pr₂(3), Ph (4), C₆F₅(5)), Ti(N-2,6-

i

C₆H₃Pr₂)(Me₃[6]aneN₃)Cl₂(11), Ti(NPh)(Me₃[6]aneN₃)Cl₂‚

0.5CH₂Cl₂(12‚0.5CH₂Cl₂)), and Ti(NC₆F₅)(Me₃[6]aneN₃)Cl₂‚

CH₂Cl₂(13‚CH₂Cl₂). Crystal data collection and processing

parameters are given in Table 5 (1-5) or Table S2 of the Supporting

Information. Crystals were mounted on glass fibers using perfluo-

ropolyether oil and cooled rapidly in a stream of cold N₂using an

Oxford Cryosystems cryostream unit. Diffraction data were mea-

sured using either a Stoe¨ Stadi-4 four-circle or a Enraf-Nonius

KappaCCD diffractometer. As appropriate, absorption and decay

corrections were applied to the data and equivalent reflections

merged.¹¹⁵The structures were solved by direct methods (SIR92¹¹⁶

or SHELXS-97¹¹⁷), and further refinements and all other crystal-

lographic calculations were performed using SHELXL-97¹¹⁸or the

CRYSTALS program suite.¹¹⁹Details of the structure solution and

refinements are given in the Supporting Information (CIF data). A

(115) Otwinowski, Z.; Minor, W. Processing of X-ray Diffraction Data

Collected in Oscillation Mode; Academic Press: New York, 1997.

(116) Altomare, A.; Cascarano, G.; Giacovazzo, G.; Guagliardi, A.; Burla,

M. C.; Polidori, G.; Camalli, M. J. Appl. Crystallogr. 1994, 27, 435.

(117) Sheldrick, G. M. Acta Crystallogr., Sect. A 1990, 46, 467.

(118) Sheldrick, G. M.; Schneider, T. R. Methods Enzymol. 1997, 277,

319.

Typical Example of an Ethylene Polymerization. Polymeriza-

tions were carried out in a stainless steel autoclave, equipped with

a magnetically coupled mechanical stirrer. The autoclave has an

internal volume of 2 L, and for these experiments the reactor was

Imido Titanium Ethylene Polymerization Catalysts

Organometallics, Vol. 25, No. 16, 2006 3903

full listing of atomic coordinates, bond lengths and angles, and

displacement parameters for all the structures has been deposited

at the Cambridge Crystallographic Data Centre. See Notice to

Authors, Issue No. 1.

Supporting Information Available: X-ray crystallographic data

in CIF format for the structure determinations of 1-5 and 11-13;

displacement ellipsoid plot and two further views of the supramo-

lecular arrangement for Ti(NC₆F₅)(Me₃[9]aneN₃)Cl₂(5); further

details of the polystyrene-supported and R₃[6]aneN₃-supported

catalyst systems 7-9, 12-15, and 59 (discussion, syntheses, and

X-ray structures); characterizing data and yields for compounds

16-58. This information is available free of charge via the Internet

at http://pubs.acs.org.

Acknowledgment. We thank the EPSRC, DSM Research,

and Millenium Pharmaceuticals Ltd. for support and Albermarle

for samples of MAO.

(119) Betteridge, P. W.; Cooper, J. R.; Cooper, R. I.; Prout, K.; Watkin,

D. J. J. Appl. Crystallogr. 2003, 36, 1487.

OM0603316

Article Doi

Imido titanium ethylene polymerization catalysts containing triazacyclic ligands

DOI: 10.1021/om0603316

Source and publish data:

Authors:

Article abstract of DOI:10.1021/om0603316

Full text of DOI:10.1021/om0603316

Products guided by the article

R&D Labs maybe for 906666-50-0

Relevant to this article

Hot Product