Synthesis of bis(phenoxyimine) Ti alkyl complexes and observation of living species by 1H NMR spectroscopy

Article abstract of DOI:10.1246/cl.2005.1382

Synthesis of [C₆F₅N=CH(2-O-C₆H ₃-3-R)]₂TiMeX (R: ^tBu; X: Br (2), R: ^tBu; X: Me (3), R: H; X: Me (4)) was achieved by the reaction of methylated titanium halides with sodium salts of the phenoxyimine ligands. These complexes can be activated with equimolar amounts of B(C₆F ₅)₃ or Ph₃CB(C₆F₅) ₄ to form methyl cationic species, which, upon addition of a small amount of ethylene monomer allows room temperature observation of cationic polymeryl species. Copyright

Full text of DOI:10.1246/cl.2005.1382

1382
Chemistry Letters Vol.34, No.10 (2005)
Synthesis of Bis(phenoxyimine) Ti Alkyl Complexes and Observation
1
of Living Species by H NMR Spectroscopy
Haruyuki Makio,ꢀ Toshiyuki Oshiki,y Kazuhiko Takai,y and Terunori Fujitaꢀ
R & D Center, Mitsui Chemicals, Inc., 580-32 Nagaura, Sodegaura, Chiba 299-0265
yDivision of Chemistry and Biochemistry, Graduate School of Natural Science and Technology,
Okayama University, Tsushima, Okayama 700-8530
(Received June 10, 2005; CL-050753)
Synthesis of [C6F5N=CH(2-O-C6H3-3-R)]2TiMeX (R: tBu;
X: Br (2), R: tBu; X: Me (3), R: H; X: Me (4)) was achieved by
the reaction of methylated titanium halides with sodium salts of
the phenoxyimine ligands. These complexes can be activated
with equimolar amounts of B(C6F5)3 or Ph3CB(C6F5)4 to form
methyl cationic species, which, upon addition of a small amount
of ethylene monomer allows room temperature observation of
cationic polymeryl species.
F
F
F
C(2)
F
F
N(1)
O(2)
O(1)
C(1)
N
O
TiMe2
2
Ti(1)
tBu
N(2)
3
Synthesis of alkyl complexes is of great value since such
complexes can be activated in a well-defined manner using an
equimolar amount of perfluoroaryl borane or borate salts without
the need for alkylaluminums. Direct observations of these alkyl
cationic species have contributed to fundamental understanding
of olefin polymerization mechanisms with metallocenes and
related complexes.1
Figure 1. Molecular structure of a Ti-FI dimethyl complex,
[C6F5N=CH(2-O-C6H3-3-tBu)]2TiMe2 (3). Selected bond
ꢀ
lengths (A) and bond angles (deg): Ti–O(1) 1.869(2); Ti–O(2)
1.872(2); Ti–N(1) 2.345(3); Ti–N(2) 2.371(3); Ti–C(1)
2.111(4); Ti–C(1) 2.102(4); O(1)–Ti–O(2) 160.3(1); N(1)–Ti–
N(2) 93.9(1); C(1)–Ti–C(2) 93.9(2).
adopts a distorted octahedral structure having a trans-O, cis-N,
and cis-Me disposition as is commonly observed for FI dichlo-
ride complexes and is consistent with the observed NMR spec-
tra. To the best of our knowledge, this represents the first struc-
turally characterized alkyl complex bearing bis(phenoxyimine)
ligands. Complex 4 was prepared by a similar manner to com-
plex 3 and obtained as orange tan powder.10
Complex 2, upon activation with an equimolar amount of
NaB(C6F5)4, exhibited lower ethylene polymerization activity
than MAO activation under the same conditions (5700
g mmolꢁ1 hꢁ1 for borate; 35,800 g mmolꢁ1 hꢁ1 for MAO3a)
and was practically inactive for propylene polymerization. Com-
plex 3 upon activation with Ph3CB(C6F5)4 displayed compara-
We have reported the unique and versatile olefin polymer-
izations with Group 4 bis(phenoxyimine) complexes (FI Cata-
lysts) activated with MAO or iBu3Al/Ph3CB(C6F5)4,2 which in-
clude living (syndioselective) polymerizations of ethylene and
propylene by fluorinated Ti-FI complexes,3 e.g., [C6F5N=
CH(2-O-C6H3-3-tBu)]2TiCl2 (1). However, attempts to synthe-
size alkylated FI complexes have been hampered by various side
reactions, including nucleophilic attacks of alkylating agents on
imine functions, reduction of Ti(IV) to Ti species with lower
oxidation states, intramolecular migration of the alkyl group,4
and ligand transfer to alkylaluminum.5 Alkylation of 1 is even
more difficult because electron-withdrawing perfluorophenyl
groups make the imines more electrophilic. We now report on
the successful synthesis of [C6F5N=CH(2-O-C6H3-3-R)]2-
TiMeX (R: Bu; X: Br (2), R: Bu; X: Me (3), R: H; X: Me
(4)) by taking a new synthetic approach, i.e., synthesis of
Me2ꢁnTiX2þn (n ¼ 1, X ¼ Br (6) or n ¼ 0, X ¼ Cl (7)) fol-
lowed by complexation with sodium salt of phenoxyimine
ligands.
A titanium precursor 6 was prepared in 70% yield according
to modified procedure of the literature.6 The reaction of 6 and
2 equiv. of [C6F5N=CH(2-O-C6H3-3-tBu)]Na in toluene at
ꢁ78 ꢂC afforded complex 2 as reddish brown powder (71%).
1H NMR spectrum of 2 exhibits two sets of ligand signals, indi-
cating collapse of C2 symmetry seen for 1, as well as a H3C-Ti
signal at 1.97 ppm.7 Dimethyltitanium dichloride (7) was gener-
ated in situ by the reaction of 1,2-dimethoxyethane adduct of
ble activities with MAO for ethylene (23,400 g mmolꢁ1 hꢁ1
)
and propylene polymerizations (2.9 kg molꢁ1 hꢁ1 for MAO;3b
t
t
1
1.3 kg molꢁ1 hꢁ1 for borate). The H NMR spectra of complex
2 activated with NaB(C6F5)4 indicated a set of ligand signals
as a main component (2ꢀ), showing a symmetric structure on
an NMR time scale. However, the signals of 2ꢀ are significantly
up-field shifted (0.56 ppm for HC=N) relative to MAO activated
species.5 This can be attributed to less cationic nature of 2ꢀ pre-
sumably derived from formation of Br-bridged dinuclear mono-
cationic complex, [{C6F5N=CH(2-O-C6H3-3-tBu)}2TiMe]2-
Brþ.11 On the other hand, activation of 3 with the trityl salt
caused immediate precipitation of a dark red oily product (3ꢀ)
indicative of a species with strong ionic character. The reaction
in bromobenzene-d5 resulted in homogeneous solution of 3ꢀ,
whose 1H NMR exhibited a set of ligand signals and methyl
groups of Ph3CMe and Me–Ti at 1.88 and 1.78 ppm, respective-
ly, corroborating the formation of [C6F5N=CH(2-O-C6H3-3-
tBu)]TiMeþB(C6F5)4ꢁ. Activation of 3 with B(C6F5)3, which
usually affords more soluble and more stable inner sphere ion
.
TiCl4 (TiCl4 DME) and 2 equiv. of MeLi in diethyl ether at
ꢁ30 ꢂC8 and used without isolation in subsequent reactions with
the sodium salts. After removal of inorganic salts by filtration,
complex 3 was precipitated from pentane solution as dark red
crystallite.9 The molecular structure of 3 shown in Figure 1
Copyright Ó 2005 The Chemical Society of Japan

Chemistry Letters Vol.34, No.10 (2005)
1383
*
(CH2)n
5*
4*
CH3B
*
CH=N
4’
TiCH3(4*)
TiCH2-(5*)
TiCH2-(5*)
4’
*
*
*
*
8.0
7.0
6.0
3.0
2.0
1.0
PPM
Figure 2. 1H NMR spectra (270 MHz: BrC6D5) of [C6F5N=CH(2-O-C6H4)]TiMeþ[MeB(C6F5)3]ꢁ (4ꢀ, bottom) and [C6F5N=CH(2-
O-C6H4)]Ti(CH2CH2)nMeþ[MeB(C6F5)3]ꢁ (5ꢀ, top). The asterisks indicate protio impurities of NMR solvent and pentane.
pairs with metallocene dialkyls,1 also resulted in oily precipita-
References and Notes
1
2
E. Y.-X. Chen and T. J. Marks, Chem. Rev., 100, 1391 (2000).
H. Makio, N. Kashiwa, and T. Fujita, Adv. Synth. Catal., 344, 477
(2002).
tion in toluene at 8.1 mM [Ti], implying significant cation–anion
separation of this species.12 Sterically much less encumbered
complex 4 with B(C6F5)3 also gave an oily species (4ꢀ) in tol-
uene-d8 (13.3 mM [Ti]), thus the origin of the ion separation is
not primarily steric. 1H NMR spectra of 4ꢀ verify the formation
of a cationic methyl species with some minor species (40) that is
not observed for 3ꢀ. The identity of the minor species is un-
known at this moment but could be a Me-bridged dinuclear spe-
cies due to reduced steric bulk of 4ꢀ. The 1H NMR shifts of
3
4
a) M. Mitani, J. Mohri, Y. Yoshida, J. Saito, S. Ishii, K. Tsuru, S.
Matsui, R. Furuyama, T. Nakano, H. Tanaka, S. Kojoh, T.
Matsugi, N. Kashiwa, and T. Fujita, J. Am. Chem. Soc., 124,
3327 (2002). b) M. Mitani, R. Furuyama, J. Mohri, J. Saito, S.
Ishii, H. Terao, T. Nakano, H. Tanaka, and T. Fujita, J. Am. Chem.
Soc., 125, 4293 (2003).
P. D. Knight, A. J. Clarke, B. S. Kimberley, R. A. Jackson, and
P. Scott, J. Chem. Soc., Chem. Commun., 2002, 352.
H. Makio and T. Fujita, Macromol. Symp., 213, 221 (2004).
K.-H. Thiele, P. Zdunneck, and D. Baumgart, Z. Anorg. Allg.
Chem., 378, 62 (1970).
ꢁ
MeB(C6F5)3 for 4ꢀ as a measure of ionic character13 are ca.
5
6
0.8 ppm in both toluene and bromobenzene that is almost identi-
cal to LiMeB(C6F5)3. We attribute this enhanced ion separation
to the formation of sterically more relaxed five-coordinated
trigonal bipyramidal cationic species that can compensate for
the energy gain due to the charge separation.
7
1H NMR (270 MHz, benzene-d6) ꢂ 7.55 (s, 1H, HC=N), 7.47 (s,
1H, HC=N), 7.44 (dd, J ¼ 3:6, 1.7 Hz, 1H, ArH), 7.41 (dd,
J ¼ 3:6, 1.7 Hz, 1H, ArH), 6.91 (dd, J ¼ 7:9, 1.7 Hz, 1H, ArH),
6.81 (dd, J ¼ 7:9, 1.7 Hz, 1H, ArH), 6.69 (dt, J ¼ 7:6, 1.7 Hz,
1.51 (s, 9H, tBu); 13C NMR (67.5 MHz, benzene-d6) ꢂ 174.7,
174.0, 163.7, 162.9, 140.2, 139.2, 136.2, 135.1, 134.6, 134.1,
124.6, 123.7, 121.0, 86.3, 35.5, 29.8–29.3; Anal. Found: C,
51.16; H, 2.86; N, 3.44%. Calcd for C28H16F10N2O2Ti: C,
50.81; H, 3.53; N, 3.39%. The discrepancy in elemental analyses
is due to instability of the complex.
Upon addition of ethylene (ca. 5 equiv. to [Ti]) to 4ꢀ in bro-
mobenzene-d5, peak intensity of species 4ꢀ decreases along with
the appearance of new signals as shown in Figure 2. Particularly
notable are the signals at 2.88 and 1.45 ppm that are assignable
to the diastereotopic ꢀ-methylene protons of the propagating
‘‘living’’ species (5ꢀ). The minor species (40) is virtually inactive
to ethylene polymerization and there is no double bond forma-
tion indicative of chain transfers. It should be noted that NMR
experiments were conducted at room temperature and that after
9 h, species 5ꢀ remained almost unchanged though 4ꢀ slightly
decreased in intensity. The same experiment for 3ꢀ also exhibits
a living propagating species (4.59 and 3.15 ppm for –CH2Ti),
although it was rather difficult to control the reaction due to
the high activity of this species14 under the conditions currently
applied.
In conclusion, we have shown that monomethyl and dimeth-
yl Ti-FI catalysts with perfluorinated N-aryl groups are accessi-
ble. These complexes can be activated by common boron coca-
talysts without the use of alkylaluminums and thus without re-
duction of the imine and Ti. The species obtained are similar
to MAO derived species in character and prove the extraordinary
robust livingness. The living propagating species can be ob-
served even at room temperature. In order to investigate dynam-
ic solution structure and polymerization mechanism of FI Cata-
lysts, including the proposed ortho-F–ꢁ-H interaction,3 and con-
figurational fluxionality,3b,15 more extensive NMR and other
spectroscopic studies are currently underway and will be report-
ed in due course.
t
2H, ArH), 1.97 (d, J ¼ 2:0 Hz, 3H, CH3-Ti), 1.57 (s, 9H, Bu),
8
9
D. Duncan and T. Livinghouse, Organometallics, 18, 4421 (1999).
1H NMR (270 MHz, benzene-d6) ꢂ 7.52 (s, 2H, HC=N), 7.45 (dd,
J ¼ 7:6, 1.7 Hz, 2H, ArH), 6.88 (dd, J ¼ 7:6, 1.7 Hz, 2H, ArH),
6.72 (t, J ¼ 7:7 Hz, 2H, ArH), 1.65 (s, 6H, CH3-Ti), 1.58 (s,
t
18H, Bu); 19F NMR (471 MHz, benzene-d6) ꢂ ꢁ149:3 (d, o-F),
ꢁ160:3 (t, p-F), ꢁ164:2 (m, m-F); Anal. Found: C, 56.12; H,
3.07; N, 3.60%. Calcd for C36H32F10N2O2Ti: C, 56.71; H, 4.23;
N, 3.67%. The discrepancy in elemental analyses is due to instabil-
ity of the complex.
10 1H NMR (270 MHz, toluene-d8) ꢂ 7.38 (s, 2H, HC=N), 7.11–7.05
(m, 2H, ArH), 6.81–6.77 (m, 4H, ArH), 6.59 (dt, J ¼ 7:6, 1.0 Hz,
2H, ArH), 1.67 (s, 6H, CH3-Ti); 19F NMR (471 MHz, benzene-d6)
ꢂ ꢁ147:0, ꢁ149:5 (br, o-F), ꢁ160:9 (t, p-F), ꢁ163:3, ꢁ164:6 (br,
m-F); Anal. Found: C, 50.69; H, 1.33; N, 4.22%. Calcd for
C
28H16F10N2O2Ti: C, 51.72; H, 2.48; N, 4.31%. The discrepancy
in elemental analyses is due to instability of the complex.
11 In fact, a 1:1 mixture of 3ꢀ and 2 gave a similar spectrum to 2ꢀ.
12 C. Zuccaccia, N. G. Stahl, A. Macchioni, M.-C. Chen, J. A.
Roberts, and T. J. Marks, J. Am. Chem. Soc., 126, 1448 (2004).
13 S. Beck, S. Lieber, F. Schaper, A. Geyer, and H.-H. Brintzinger,
J. Am. Chem. Soc., 123, 1483 (2001).
14 R. Furuyama, M. Mitani, J. Mohri, R. Mori, H. Tanaka, and T.
Fujita, Macromolecules, 38, 1546 (2005).
15 Y. Tohi, H. Makio, S. Matsui, M. Onda, and T. Fujita, Macromo-
lecules, 36, 523 (2003).
Published on the web (Advance View) September 10, 2005; DOI 10.1246/cl.2005.1382