Cationic alkyl aluminium ethylene polymerization catalysts based on monoanionic N,N,N-pyridyliminoamide ligands

Article abstract of DOI:10.1039/a807950a

Treatment of the 2,6-bis(imino)pyridines {[2,6-(ArNCR)₂C₅H₃N]} [R = H, Ar = 2,6-i-Pr₂C₆H₃ or 2,4,6-Me₃C₆H₂; R = Me, Ar = 2,6-i-Pr₂C₆H₃] With AlMe₃ at elevated temperature gives, via migration of a methyl group to the ligand backbone, the pseudo-five coordinate dimethylaluminium species {2-[ArNCR((Me)],6-(ArNCR)C₅H₃N}-AlMe₂ (1a-c); upon treatment with B(C₆F₅)₃, 1a-c cleanly afford the cationic methyl complexes [{2-[ArNCR(Me)],6-(ArNCR)C₅H₃N}AlMe]⁺[MeB(C₆F₅)₃]^-(2a-c) which are active for ethylene polymerization.

Full text of DOI:10.1039/a807950a

Cationic alkyl aluminium ethylene polymerization catalysts based on
monoanionic N,N,N-pyridyliminoamide ligands
Michael Bruce, Vernon C. Gibson,* Carl Redshaw, Gregory A. Solan, Andrew J. P. White and David J.
Williams
Department of Chemistry, Imperial College, South Kensington, London, UK SW7 2AY. E-mail: V.Gibson@ic.ac.uk
Received (in Cambridge, UK) 13th October 1998
Treatment
of
the
2,6-bis(imino)pyridines
active as ethylene polymerization catalysts. The significance of
this observation is highlighted by a recent report that bidentate
N,N-chelate ligand systems can lead to undesirable exchange
reactions which thwart the generation of a polymerization-
active site.8
{[2,6-(ArNCR)2C5H3N]} [R = H, Ar = 2,6-i-Pr2C6H3 or
2,4,6-Me3C6H2; R = Me, Ar = 2,6-i-Pr2C6H3] with AlMe3 at
elevated temperature gives, via migration of a methyl group
to the ligand backbone, the pseudo-five coordinate dimethyl-
aluminium species {2-[ArNCR(Me)],6-(ArNCR)C5H3N}-
AlMe2 (1a–c); upon treatment with B(C6F5)3, 1a–c cleanly
afford the cationic methyl complexes [{2-[ArNCR(Me)],6-
(ArNCR)C5H3N}AlMe]+[MeB(C6F5)3]2 (2a–c) which are
active for ethylene polymerization.
Reaction
of
the
parent
2,6-bis(imino)pyridines
{[2,6-(ArNCR)2C5H3N]} [R = H, Ar = 2,6-i-Pr2C6H3 or
2,4,6-Me3C6H2; R = Me, 2,6-i-Pr2C6H3] with AlMe3 in
refluxing toluene (12 h) results in alkylation of the ligand
backbone to give the dimethylaluminium species
{2-[ArNCR(Me)],6-(ArNCR)C5H3N}AlMe2 [R = H; Ar =
2,6-i-Pr2C6H3 1a; R = H; Ar = 2,4,6-Me3C6H2 1b; R = Me,
Ar = 2,6-i-Pr2C6H3 1c] in high yield (Scheme 1).†
Crystals of 1b suitable for an X-ray structure determination
were grown from MeCN. The molecular structure‡ of 1b shows
the N(9)–C(9)–py–C(7)–N(7) portion of the ligand to be co-
planar to within 0.06 Å (Fig. 1), a geometry very similar to that
observed for the closely related bis(imino)pyridine ligand in its
iron complex.4 Here the aluminium atom lies 0.33 Å out of this
plane and adopts a severely distorted tetrahedral geometry with
Neutral aluminium alkyls are well known to act as ethylene
oligomerization1 and polymerization2 catalysts. However, the
potential of cationic aluminium alkyls as catalysts is just
emerging. Recent advances by Coles and Jordan3 have utilised
a number of chelating N,N-amidinate ligands, viz. {RC(NRA)2}
(R = Me, RA = i-Pr, Cy; R = t-Bu, RA = i-Pr, Cy, SiMe3)
which, upon reaction with Me3Al, afford complexes of the form
[{RC(NRA)2}AlMe2]. Cationic species are readily generated on
further reaction with B(C6F5)3 or [HNMe2Ph][B(C6F5)4], the
latter giving amine adducts. In the case of R = t-Bu, RA = i-Pr
the derived cation [from B(C6F5)3] polymerizes ethylene at
ambient temperature, albeit with low activity.
In a separate study, we4 and Brookhart and coworkers5 have
recently shown that iron and cobalt complexes bearing neutral,
6-electron donor 2,6-bis(imino)pyridine ligands afford excep-
tionally active polymerization catalysts when activated with
methylaluminoxane (MAO). We became interested in extend-
ing the range of N,N,N-chelate ligands to monoanionic
derivatives in which one of the imino groups is transformed into
an amido functionality, and nucleophilic attack on the imine
carbon using an alkylaluminium reagent offered a convenient
approach.6 The aluminium complexes so-derived can be used to
provide a source of free pyridyliminoamine ligands (via
hydrolysis); details of the synthetic utility of this reaction will
be reported elsewhere.7 Here, we show that the dimethylalumin-
ium complexes bearing such tridentate N,N,N-ligands also can
be converted cleanly to cationic alkyl derivatives which are
Fig. 1 The molecular structure of 1b. Selected bond lengths (Å) and angles
(°): Al–N(1) 2.029(4), Al–N(7) 1.876(4), Al–N(9) 2.575(4), Al–C(10)
1.990(6), Al–C(11) 1.951(5), C(7)–N(7) 1.444(6), C(9)–N(9) 1.273(5),
N(7)–Al–C(11) 103.6(2), N(7)–Al–C(10) 113.8(2), C(11)–Al–C(10)
114.0(3), N(7)–Al–N(1) 81.3(2), C(11)–Al–N(1) 136.0(2), C(10)–Al–N(1)
102.8(2).
Scheme 1 Reagents and conditions: (i) AlMe3, toluene, 110 °C, 12 h; (ii) B(C6F5)3, toluene, rt.
Chem. Commun., 1998, 2523–2524
2523

View Online
Table 1 Results of ethylene polymerization runs with cations 2a–c
Activity/
g mol21 h21 bar21
Runa
Cationb
Yield/g
Mn
Mw
Mwc/Mn
Mpk
c
c
c
c
1
2
3
2a
2b
2c
0.10
0.08
0.15
80
60
120
7800
5200
2400
23 000
33 000
13 000
2.9
6.3
5.5
19 000
13 000
9 800
a
b
All runs performed in toluene at 5 bar of ethylene, 40 °C, 60 min, using 0.25 mmol of cation. Generated in situ from the reaction of equimolar (0.25
mmol) amounts of 1a–c and B(C6F5)3. c Determined by GPC at 160 °C.
angles at aluminium ranging between 81.3(2) and 136.0(2)°.
There is a slight asymmetry in the Al–Me distances [1.990(6) Å
to C(10) and 1.951(5) Å to C(11)] but a more marked difference
between the two Al–N bonds, with that to the formally
negatively charged nitrogen N(7) being significantly shorter [at
1.876(4) Å] than that to the pyridyl nitrogen [2.029(4) Å].
Perhaps the most interesting feature of the structure is the
directing of the imino nitrogen N(9) into the flattened ‘basal’
face of the tetrahedron—the aluminium atom lies only 0.3 Å out
of the N(1)–C(10)–C(11) plane whereas it lies between 0.6 and
0.9 Å out of the other tetrahedral faces. The distance is long at
2.575(4) Å, but bearing in mind the potential for this nitrogen
and the pyridyl nitrogen to adopt an anti relationship in the
absence of a metal ion7 we believe that this interaction is real
and indeed a key feature in the subsequent cation formation in
2.
Notes and references
† Satisfactory microanalyses have been obtained. Selected spectroscopic
data: For 1a: 1H NMR (CD2Cl2, 293 K): d 8.61 (s, 1H, NNCH), 8.49 [d, 1H,
3J(HH) 7.6, Py-Hm], 8.15 [app. t, 1H, 3J(HH) 7.6 Py-Hp], 7.67 [d, 1H,
3J(HH) 7.6, Py-Hm], 20.67 (s, 3H, AlMe), 20.89 (s, 3H, AlMe). For 1b: 1H
NMR (CD2Cl2, 293 K): d 8.57 (s, 1H, NNCH), 8.30 [d, 1H, 3J(HH) 7.6, Py-
Hm], 8.11 [app. t, 1H, 3J(HH) 7.6, Py-Hp], 7.68 [d, 1H, 3J(HH) 7.6, Py-Hm],
20.73 (s, 3H, AlMe), 21.01 (s, 3H, AlMe). For 1c: 1H NMR (CD2Cl2, 293
K): d 8.31 [app. t, 1H, 3J(HH) 7.6 Py-Hp], 8.03 [d, 1H, 3J(HH) 7.6, Py-Hm],
7.78 [d, 1H, 3J(HH) 7.6, Py-Hm], 2.31 (s, 3H, NNCMe), 1.82 (s, 6H,
NCMe2), 20.72 (s, 3H, AlMe), 20.89 (s, 3H, AlMe). For 2a: 1H NMR
(CD2Cl2, 293 K): d 8.68 (s, 1H, NNCH), 8.15 [app. t, 1H, 3J(HH) 7.6, Py-
Hp], 8.17 [d, 1H, 3J(HH) 7.6, Py-Hm], 8.13 [d, 1H, 3J(HH) 7.6, Py-Hm], 0.43
(s, 3H, BMe), 20.70 (s, 3H, AlMe). For 2b: 1H NMR (CD2Cl2, 293 K): d
8.57 (s, 1H, NNCH), 8.33 [app. t, 1H, 3J(HH) 7.6, Py-Hp], 8.01 [d, 1H,
3J(HH) 7.6, Py-Hm], 7.98 [d, 1H, 3J(HH) 7.6, Py-Hm], 4.67 [q, 1H, 3J(HH)
6.7, CHMe], 1.26 (d, 3H, CHMe), 0.33 (s, 3H, BMe), 20.85 (s, 3H, AlMe).
For 2c: 1H NMR (CD2Cl2, 293 K): d 8.29 [app. t, 1H, 3J(HH) 7.6, 7.6, Py-
Hp], 8.07 [d, 1H, 3J(HH) 7.6 Py-Hm], 8.03 [d, 1H, 3J(HH) 7.6, Py-Hm], 2.30
(s, 3H, NNCMe), 1.79 (s, 6H, NCMe2), 0.40 (s, 3H, BMe), 20.77 (s, 3H,
AlMe).
‡ Crystal data for 1b: C28H36N3Al, M = 441.6, triclinic, space group P1
(no. 2), a = 7.992(2), b = 8.169(1), c = 20.979(3) Å, a = 82.28(1), b =
82.93(2), g = 71.92(1)°, V = 1285.4(4) Å3, Z = 2, Dc = 1.141 g cm23
m(Cu-Ka) = 8.21 cm21, F(000) = 476, T = 183 K; orange/red platy
needles, 0.23 3 0.17 3 0.03 mm, Siemens P4/RA diffractometer, w-scans,
3810 independent reflections. The structure was solved by direct methods
and the non-hydrogen atoms were refined anisotropically using full matrix
least-squares based on F2 to give R1 = 0.073, wR2 = 0.169 for 2432
independent observed reflections [|Fo| > 4s(|Fo|), 2q @ 120°] and 290
parameters. CCDC 182/1059.
The 1H NMR spectra are consistent with the solid-state
structures of 1 being maintained in solution. For 1a, the pyridyl
meta-protons resonate at d 8.49 and 7.67 while the coordinated
methyl groups appear as singlets at d 20.67 and 20.89
reflecting the C1 symmetry of the complex.
¯
The cationic complexes [{2-[ArNCR(Me)],6-(ArNCR)-
C5H3N}AlMe]+ (2a–c) are readily generated on treatment of
one equivalent of [B(C6F5)3] in toluene at ambient temperature
(Scheme 1).† For example, the 1H NMR spectrum arising from
2a reveals a sharp singlet at d 20.70 for the methyl group
coordinated to aluminium, while the methyl group coordinated
to boron of the [MeB(C6F5)3]2 counter-anion is clearly seen as
a broad singlet at d 0.43. The upfield shift of this resonance is
consistent with a free anion9 and contrasts with the more
downfield resonance (d 1.67) observed by Coles and Jordan in
which a B–Me···Al association is invoked.3
All the cationic complexes 2a–c are active for ethylene
polymerization (see Table 1) affording solid polyethylene with
activities between 80 and 120 g mol21 h21 bar21. The polymer
products in each case are low molecular weight, with Mws
ranging from 33 000 (run 2) to 13 000 (run 3). It is noteworthy
that by changing the ligand backbone (otherwise identical) in 2a
from a single methyl group to three methyl groups in 2c has the
effect of reducing the molecular weight by almost half (cf. runs
1 and 3).
In a series of experiments on the iron and cobalt catalyst
systems, we have shown that the bis(imino)pyridine ligands
bonded to iron and cobalt are not attacked by AlMe3 or MAO
under the conditions of the polymerization experiment: free
bis(imino)pyridine can be isolated in quantitative yield follow-
ing hydrolytic work-up after the polymerization, i.e. no
alkylation of the ligand backbone occurs of the type described
here.
,
1 K. Ziegler, H.-G. Tosel, E. Holzkamp, J. Schneider, M. Söll and W. R.
Kroll, Justus Leibigs Ann. Chem., 1960, 629, 121.
2 H. Martin and H. Bretinger, Makromol. Chem., 1992, 193, 1283.
3 M. P. Coles and R. F. Jordan, J. Am. Chem. Soc., 1997, 119, 8125.
4 G. J. P. Britovsek, V. C. Gibson, B. S. Kimberley, P. J. Maddox, S. J.
McTavish, G. A. Solan, A. J. P. White and D. J. Williams, Chem.
Commun., 1998, 848.
5 B. L. Small, M. Brookhart and M. A. Bennett, J. Am. Chem. Soc., 1998,
120, 4049.
6 (a) J. M. Klerks, D. J. Stufkens, G. van Koten and K. Vrieze,
J. Organomet. Chem., 1979, 181, 271; (b) V. C. Gibson, C. Redshaw,
A. J. P. White and D. J. Williams, J. Organomet. Chem., 1998, 550,
453.
7 V. C. Gibson, S. Mastroianni, C. Redshaw, G. A. Solan, A. J. P. White
and D. J. Williams, to be submitted.
8 B. Qiani, D. L. Ward and M. R. Smith III, Organometallics, 1998, 17,
3070.
9 X. Yang, C. L. Stern and T. J. Marks, J. Am. Chem. Soc., 1994, 116,
10015.
BP Chemicals Ltd is thanked for financial support. Dr J.
Boyle and G. Audley are thanked for NMR and GPC
measurements, respectively.
Communication 8/07950A
2524
Chem Commun., 1998, 2523–2524