Propylene polymerization with cyclopentadienyltitanium(IV) hydroxylaminato complexes

Article abstract of DOI:10.1021/om800571j

Complexes of the type Cp*TiX₂(ONR'R") (Cp* = n⁵-C₅Me₅; X = Me, Cl; R = R' = Et, TEMPO (2,2,6,6-tetramethylpiperidine-n-oxyl); X = R' = Me, R" = Bu) were synthesized by several routes. Upon activation with [Ph

Full text of DOI:10.1021/om800571j

Organometallics 2009, 28, 405–412
405
Propylene Polymerization with Cyclopentadienyltitanium(IV)
Hydroxylaminato Complexes
Andrew P. Dove,† Elizabeth T. Kiesewetter, Xavier Ottenwaelder,‡ and
Robert M. Waymouth*
Department of Chemistry, Stanford UniVersity, Stanford, California 94305
ReceiVed June 19, 2008
Complexes of the type Cp*TiX2(ONR′R′′) (Cp* ) η5-C5Me5; X ) Me, Cl; R ) R′ ) Et, TEMPO
t
(2,2,6,6-tetramethylpiperidine-N-oxyl); X ) R′ ) Me, R′′ ) Bu) were synthesized by several routes.
Upon activation with [Ph3C]+[B(C6F5)4]- and AliBu3, these complexes generate highly active catalysts
for propylene polymerization, although significant catalyst deactivation was observed. Activation with
B(C6F5)3/AliBu3 or methylaluminoxane (MAO) resulted in reduced polymerization activity, the latter
leading to increased catalyst lifetime. Model studies showed that the interaction of AlMe3 with
Cp*Ti(Me)2(ONEt2) led to the formation of Cp*Ti(Me)2(η2-O(AlMe3)NEt2). The X-ray crystal structure
confirmed that the hydroxylaminato ligand remained η2 bound to the titanium center with the AlMe3
bound to the complex through the oxygen atom of the hydroxylaminato ligand. Exchange reactions with
organic ethers revealed the metalloether to be a comparable donor to PhOMe. Cp*Ti(Me)2(ONtBu(Me))
was revealed to be a weaker donor than Cp*Ti(Me)2(ONEt2); Cp*Ti(Me)2(TEMPO) did not bind to AlMe3.
AliBu3 bound more weakly to Cp*Ti(Me)2(ONEt2) than AlMe3. Reaction of Cp*Ti(Me)2(ONEt2) and
Cp*Ti(Me)2(ONtBu(Me)) with B(C6F5)3 resulted in clean formation of the zwitterionic contact ion pairs
[Cp*Ti(Me)(η2-ONEt2)]+[MeB(C6F5)3]- and [Cp*Ti(Me)(η2-ONtBu(Me))]+[MeB(C6F5)3]-, respectively,
whereas reaction of Cp*Ti(Me)2(η2-ONtBu(Me)) with [Ph3C]+[B(C6F5)4]- resulted in the clean formation
of the solvent-separated ion pair [Cp*Ti(Me)(η2-ONtBu(Me))]+[B(C6F5)4]-. Reaction of Cp*Ti(Me)2(η1-
TEMPO) with B(C6F5)3 results in the elimination of methane to result in the formation of the contact ion
paired [Cp*Ti(η2-TEMPO)]+[MeB(C6F5)3]-, in which one of the TEMPO methyl groups has undergone
C-H activation, resulting in a η2-bound TEMPO ligand, confirmed by 1H, gROESY, and 1H-15N HMBC
NMR. Addition of AliBu3 or AlMe3 to the cationic [Cp*Ti(Me)(ONtBu(Me))]+[B(C6F5)4]- resulted in
decomposition of the cations.
have focused on bis-cyclopentadienyl metallocenes,1,4,11-14
investigations of coordination compounds lacking cyclopenta-
dienyl ligands15-18 or half-metallocenes containing only one
Introduction
New classes of coordination complexes have spawned a
renaissance in olefin polymerization catalysis. The application
and study of these catalysts has led to both improved catalysts
and polymers on an industrial scale and new insights into the
mechanisms of enchainment.1-10 While considerable studies
(11) Kaminsky, W. Macromol. Chem. Phys. 1996, 197, 3907–3945.
(12) Dankova, M.; Waymouth, R. M. Macromolecules 2003, 36, 3815–
3820.
(13) Jensen, T. R.; Yoon, S. C.; Dash, A. K.; Luo, L. B.; Marks, T. J.
J. Am. Chem. Soc. 2003, 125, 14482–14494.
(14) Lofgren, B.; Kokko, E.; Seppala, J. AdV. Polym. Sci. 2004, 169,
1–12.
(15) Gibson, V. C.; Spitzmesser, S. K. Chem. ReV. 2003, 103, 283–
315.
(16) Zuo, W. W.; Sun, W. H.; Zhang, S.; Hao, P.; Shiga, A. J. Polym.
Sci. Part A 2007, 45, 3415–3430.
* Corresponding author. E-mail: waymouth@stanford.edu.
† Department of Chemistry, University of Warwick, Coventry, U.K. CV4
7A.
‡ Current address: Department of Chemistry, Concordia University,
Montreal, Quebec, Canada.
(1) Brintzinger, H. H.; Fischer, D.; Mu¨lhaupt, R.; Rieger, B.; Waymouth,
R. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1143–1170.
(2) Coates, G. W. Chem. ReV. 2000, 100, 1223–1252.
(3) Kaminsky, W. J. Polym. Sci. Part A 2004, 42, 3911–3921.
(4) Resconi, L.; Cavallo, L.; Fait, A.; Piemontesi, F. Chem. ReV. 2000,
100, 1253–1345.
(17) Gendler, S.; Groysman, S.; Goldschmidt, Z.; Shuster, M.; Kol, M.
J. Polym. Sci. Part A 2006, 44, 1136–1146.
(18) Capacchione, C.; Proto, A.; Ebeling, H.; Mu¨lhaupt, R.; Okuda, J.
J. Polym. Sci. Part A 2006, 44, 1908–1913.
(19) Nomura, K.; Liu, J. Y.; Padmanabhan, S.; Kitiyanan, B. J. Mol.
Catal. A Chem. 2007, 267, 1–29.
(20) Dove, A. P.; Xie, X.; Waymouth, R. M. Chem. Commun. 2005,
2152–2154.
(5) Lai, S. Y.; Wilson, J. R.; Knight, G. W.; Stevens, J. C.; Chum, P. S.
U.S. Patent 5,272,236, 1993.
(6) Scheirs, J.; Kaminsky, W. Metallocene-based Polyolefins: Prepara-
tion, Properties, and Technology.: John Wiley & Sons Ltd.: Chichester,
2000; Vol. 1, p 2.
(7) Stevens, J. C. Stud. Surf. Sci. Catal. 1994, 89, 277–284.
(8) Landis, C. R.; Rosaaen, K. A.; Sillars, D. R. J. Am. Chem. Soc.
2003, 125, 1710–1711.
(9) Ferreira, M. J.; Martins, A. M. Coord. Chem. ReV. 2006, 250, 118–
132.
(10) Gladysz, J. A. E Chem. ReV. 2000, 100, Special Issue on Frontiers
in Metal-Catalyzed Polymerization.
(21) Mahanthappa, M. K.; Cole, A. P.; Waymouth, R. M. Organome-
tallics 2004, 23, 836–845.
(22) Stephan, D. W.; Stewart, J. C.; Guerin, F.; Courtenay, S.; Kickham,
J.; Hollink, E.; Beddie, C.; Hoskin, A.; Graham, T.; Wei, P. R.; Spence,
R. E. V.; Xu, W.; Koch, L.; Gao, X. L.; Harrison, D. G. Organometallics
2003, 22, 1937–1947.
(23) Stephan, D. W.; Stewart, J. C.; Guerin, F.; Spence, R. E. V. H.;
Xu, W.; Harrison, D. G. Organometallics 1999, 18, 1116–1118.
(24) Guerin, F.; Beddie, C. L.; Stephan, D. W.; Spence, R. E.; Wurz,
R. Organometallics 2001, 20, 3466–3471.
10.1021/om800571j CCC: $40.75
2009 American Chemical Society
Publication on Web 12/11/2008

406 Organometallics, Vol. 28, No. 2, 2009
Scheme 1. Synthetic Routes to Cp*M(hydroxylaminato) Complexes
DoVe et al.
cyclopentadienyl ligand19-36 have expanded the scope of
coordination catalysts for olefin polymerization. The remarkable
1-octene,19,25 but the activities and catalyst lifetimes depend
sensitively on the nature of the catalyst precursor as well as the
activation strategy. Studies to examine the interaction of these
complexes with the cocatalyst system have revealed several
deactivation pathways.32,42-45
Hydroxylamines (or nitroxides) are an intriguing class of
anionic ligands with tunable M-L bond strengths46-49 that can
bind as either monodentate (η1) or bidentate (η2) ligands,
depending on the hydroxylamine substituents and the ancillary
ligation at the metal.21,49-58 We recently reported that mono-
cyclopentadienyl titanium hydroxylaminato complexes exhibit
high activities for propylene polymerization.20 In this paper, we
describe the synthesis and propylene polymerization behavior
of a series of monocyclopentadienyl hydroxylaminato com-
plexes, their propylene polymerization behavior, and the reaction
chemistry of these complexes with alkylaluminum reagents and
a series of activators (MAO, [Ph3C]+[B(C6F5)4]-, and B(C6F5)3).
success of the titanium “constrained geometry” catalysts such
7,37-41
as [(Me2Si)(η5-C5Me4)(N(tBu)]TiCl2
has stimulated re-
newed interest in the behavior of monocyclopentadienyl com-
plexes with different monoanionic coligands.19-36 In conjunction
with cocatalysts such as methylaluminoxane (MAO), triphenyl-
carbenium
tetrakis(pentafluorophenyl)borate
([Ph3C]+-
[B(C6F5)4]-), or tris(pentafluorophenyl)borane (B(C6F5)3), many
of these complexes have been shown to be effective catalysts
for the homopolymerization of ethylene and the copolymeri-
zation of ethylene with higher olefins such as 1-hexene or
(25) Mcmeeking, J.; Gao, X.; Von Haken Spence, R. E.; Brown, S. J.;
Jeremic, D. US Patent 6,114,481, 2000.
(26) Zhang, W.; Sita, L. R. J. Am. Chem. Soc. 2008, 130, 442.
(27) Zhang, W.; Sita, L. R. AdV. Synth. Catal. 2008, 350, 439–447.
(28) Harney, M. B.; Zhang, Y. H.; Sita, L. R. Angew. Chem., Int. Ed.
2006, 45, 2400–2404.
(29) Nomura, K.; Fujii, K. Macromolecules 2003, 36, 2633–2641.
(30) Nomura, K.; Fujita, K.; Fujiki, M. J. Mol. Catal., A Chem. 2004,
220, 133–144.
(31) Manz, T. A.; Phomphrai, K.; Medvedev, G.; Krishnamurthy, B. B.;
Sharma, S.; Haq, J.; Novstrup, K. A.; Thomson, K. T.; Delgass, W. N.;
Caruthers, J. M.; Abu-Omar, M. M. J. Am. Chem. Soc. 2007, 129, 3776–
3777.
(32) Zhang, S.; Piers, W. E.; Gao, X. L.; Parvez, M. J. Am. Chem. Soc.
2000, 122, 5499–5509.
(33) Dias, A. R.; Duarte, M. T.; Fernandes, A. C.; Fernandes, S.;
Marques, M. M.; Martins, A. M.; Da Silva, J. F.; Rodrigues, S. S. J.
Organomet. Chem. 2004, 689, 203–213.
(34) Fenwick, A. E.; Phomphrai, K.; Thorn, M. G.; Vilardo, J. S.; Trefun,
C. A.; Hanna, B.; Fanwick, P. E.; Rothwell, I. P. Organometallics 2004,
23, 2146–2156.
(35) Tamm, M.; Randoll, S.; Herdtweck, E.; Kleigrewe, N.; Kehr, G.;
Erker, G.; Rieger, B. Dalton Trans. 2006, 459–467.
(36) Sinnema, P. J.; Spaniol, T. P.; Okuda, J. J. Organomet. Chem. 2000,
598, 179–181.
Results and Discussion
Synthesis of (Cyclopentadienyl)titanium(hydroxylaminato)
Complexes. We have developed several synthetic routes to
monocyclopentadienyl titanium hydroxylaminato complexes
(Scheme 1).20,21,49,53 For hydroxylamines that form stable
nitroxyl radicals such as TEMPO (2,2,6,6-tetramethylpiperidine-
N-oxyl), coupling of the nitroxyl radical with a reduced Ti(III)
precursor generates stable TEMPO complexes (Scheme 1A).21,51,53
Nitroxides generated by deprotonation of hydroxylamines or
by alkylation of nitrones react with titanium halides to generate
the hydroxylaminato titanium complexes (Scheme 1B,C).21,53
(42) Nomura, K.; Fudo, A. Inorg. Chim. Acta 2003, 345, 37–43.
(43) Byun, D. J.; Fudo, A.; Tanaka, A.; Fujiki, M.; Nomura, K.
Macromolecules 2004, 37, 5520–5530.
(44) Zhang, S. B.; Piers, W. E. Organometallics 2001, 20, 2088–2092.
(45) Kickham, J. E.; Guerin, F.; Stewart, J. C.; Urbanska, E.; Stephan,
D. W. Organometallics 2001, 20, 1175–1182.
(46) Huang, K. W.; Han, J. H.; Cole, A. P.; Musgrave, C. B.; Waymouth,
R. M. J. Am. Chem. Soc. 2005, 127, 3807–3816.
(47) Huang, K. W.; Han, J. H.; Musgrave, C. B.; Waymouth, R. M.
Organometallics 2006, 25, 3317–3323.
(37) Mcknight, A. L.; Waymouth, R. M. Chem. ReV. 1998, 98, 2587–
2598.
(38) Okuda, J. Chem. Ber. 1990, 123, 1649–1651.
(39) Stevens, J. C. In Studies in Surface Science and Catalysis;
Hightower, J. W., Delglass,W. N., Iglesia, E., Bell, A. T., Eds.; Elsevier:
Amsterdam,1996; Vol. 101, pp 11-20.
(40) Canich, J. M.; Hlatky, G. G.; Turner, H. W. WO 92-00333, 1992.
(41) Chen, Y. X.; Marks, T. J. Organometallics 1997, 16, 3649–3657.

Cp*Ti(hydroxylaminato) Catalysts for Propylene Polymerization
Organometallics, Vol. 28, No. 2, 2009 407
Table 1. Propylene Polymerization Using Cp* Titanium Hydroxylaminato Complexes
entry
cat.
activator
scav.
pP (psig)
[Ti] (mM)
yield (g)
prodg
Mwh (kDa)
PDIh
m4i
r4i
d
d
d
d
1
2
3
4
5
6
7
8
8a
9a
5a
8a
8a
8b
8b
8b
4b
8b
8b
[Ph3C]+
[Ph3C]+
[Ph3C]+
B(C6F5)3
MAOc
AliBu3
AliBu3
AliBu3
AliBu3
liq
liq
liq
liq
liq
60
60
60
60
60
60
0.25
1.25
1.25
25
25
0.3
30
0.3
0.3
30
1.3
1.6
0.5
0.2
1.3
1.5
trace
1.1
0.4
153.2
37.4
11.4
0.3
1.5
50.0
2505
2490
2500
1380
1675
1340
1.90
1.90
2.00
2.45
2.03
2.02
2.6
0.9
0.9
3.6
1.1
1.3
17.3
16.6
16.3
17.6
21.1
19.8
d
d
[Ph3C]+
B(C6F5)3
MAOc
AliBu3
AliBu3
37.3
13.3
1870
1610
2.07
1.87
2.1
2.1
18.6
17.3
9
10
11
MAOc
e
[Ph3C]+
[Ph3C]+
AlMe3
MADf
30
a Polymerizations were carried out in 90 mL of liquid propylene, 10 mL of toluene containing 60 mg of AltBu3 at 20 ( 1 °C for 20 min, [Ph3C]+
)
b
[Ph3C]+[B(C6F5)4]-. Polymerizations were carried out in 100 mL of toluene containing 60 mg of AltBu3 at 20 ( 1 °C for 60 min. c Dry MAO.
d AliBu3 ) triisobutylaluminum. e AlMe3 ) trimethylaluminum. f MAD ) bis(2,6-di-tert-butyl-4-methylphenoxy)methylaluminum. g Productivity kg
i
poly(propylene) (mmol Ti-1) (h)-1
.
h Determined by gel permeation chromatography. Determined by 13C NMR.
Conversion of the dichloro complexes to the dimethyl analogues
can be achieved by treatment with 2 equiv of methyl lithium or
methyl Grignard reagent (Scheme 1D).
site behavior. The productivity of Cp*Ti(Me)2(ONEt2), 8, at
room temperature approaches 153 (kg PP) (mmol Ti)-1 (h)-1
,
comparable to some of the most active constrained geometry
type catalysts.62 Catalysts derived from the η2-bound hydroxy-
laminato ligands 8 and 9 are more active than those derived
from the η1-bound TEMPO ligand 5. In contrast, the hafnium
complex 10 shows no activity for the polymerization of
propylene upon activation with [Ph3C]+[B(C6F5)4]- at room
temperature in the presence of AliBu3.
In liquid propylene, the highest activities were observed upon
activation with [Ph3C]+[B(C6F5)4]-; activation of 8 with
B(C6F5)3 in the presence of AliBu3 or with dry MAO resulted
in a significantly reduced polymerization activity (Table 1,
entries 1, 4 and 5).
Solution Propylene Polymerization. The time course of
propylene conversion was monitored in toluene with a gaseous
monomer feed at 4 atm to study the catalyst lifetime as a
function of activation conditions. Propylene was introduced at
constant pressure from a ballast vessel of known volume,
temperature, and propylene pressure, and gas uptake was
recorded from the pressure drop in the gas ballast.63 Activation
of 8 with B(C6F5)3/AliBu3 in solution at 4 atm of propylene
Hydroxylaminato titanium alkyl complexes can also be
prepared from the alkyl precursors. Addition of HONEt2 to
Cp*TiMe3, 7, in pentane generates Cp*Ti(Me)2(ONEt2), 8
(Scheme 1E). Alternatively nitrosoalkyl compounds can be
cleanly and selectively inserted into a single titanium or hafnium
methyl bond to generate a hydroxylaminato ligand in situ.20,59-61
Thus, Cp*Ti(Me)2(ONtBu(Me)), 9,20 or Cp*Hf(Me)2-
(ONtBu(Me)), 10, was generated in high yield by treatment of
t
Cp*TiMe3 or Cp*HfMe3 with 1 equiv of BuNO in pentane
(Scheme 1F).
The hapticity of the hydroxylaminato ligands depends sen-
sitively on the nature of the hydroxylamine and the ancillary
ligation at the metal.20,21,49,51,56,57 X-ray crystallographic analysis
of the TEMPO complex 2 reveals a η1-bound coordination
geometry, whereas the hydroxylamine in 820 and in
Cp*Ti(Cl)2(ONMe2)56 is bound η2 through both oxygen and
nitrogen. The sterically demanding TEMPO ligand also adopts
an η2 coordination geometry in the absence of cyclopentadienyl
ligands as in (TEMPO)TiCl3.53,57
In solution, the 15N NMR chemical shift provides a useful
indicator of the hapticity of the hydroxylamine ligand. The 15
N
(48) George, S. D.; Huang, K. W.; Waymouth, R. M.; Solomon, E. I.
Inorg. Chem. 2006, 45, 4468–4477.
(49) Huang, K. W.; Waymouth, R. M. J. Am. Chem. Soc. 2002, 124,
8200–8201.
(50) Nagy, S.; Etherton, B. P.; Krishnamurti, R.; Tryrell, J. A. US Patent
6,204,216 B1, March 20, 2001.
(51) Mahanthappa, M. K.; Cole, A. P.; Waymouth, R. M. Organome-
tallics 2004, 23, 1405–1410.
(52) Huang, K. W.; Waymouth, R. M. J. Chem. Soc., Dalton Trans.
2004, 354–356.
(53) Mahanthappa, M. K.; Huang, K. W.; Cole, A. P.; Waymouth, R. M.
Chem. Commun. 2002, 502–503.
(54) Mitzel, N. W.; Parsons, S.; Blake, A. J.; Rankin, D. W. H. J. Chem.
Soc., Dalton Trans. 1996, 2089–2093.
(55) Wieghardt, K.; Tolksdorf, I.; Weiss, J.; Swiridoff, W. Z. Anorg.
Allg. Chem. 1982, 490, 182–190.
NMR chemical shifts can be measured by 1H-15N heteronuclear
multiple-bond correlation spectroscopy (HMBC). These studies
revealed that Cp*Ti(Me)2(TEMPO), 5, displays two resonances
at δ ) 217 and 228 ppm, which we attribute to two ring-flip
isomers of an η1-bound TEMPO ligand. The 15N NMR
resonance of 9 occurs farther upfield at δ ) 134 ppm, indicative
of an η2 coordination geometry in solution, consistent with that
observed in the solid state. Similar analysis of the diethyl
hydroxylaminato complexes 4 and 8 revealed 15N NMR chemi-
cal shifts of δ ) 129 and 143 ppm, respectively, indicative of
a bidentate coordination of the less sterically hindered hydroxy-
1
lamine ligands. Analysis of high-field (500 MHz) H NMR
(56) Hughes, D. L.; Jimeneztenorio, M.; Leigh, G. J.; Walker, D. G.
J. Chem. Soc., Dalton Trans. 1989, 2389–2395.
(57) Golubev, V. A.; Voronina, G. N.; Chernaya, L. I.; Dyachkovskii,
F. S.; Matkovskii, P. E. Zh. Obs. Khim. 1977, 47, 1825–1832.
(58) Brindley, P. B.; Scotton, M. J. J. Organomet. Chem. 1981, 222,
89–96.
spectra of 4 reveals a doublet of doublets for the methylene
resonances of the diethyl hydroxylamine, indicative of restricted
rotation of the N-ethyl groups due to the η2 binding of the
hydroxylaminato ligand.
Propylene Polymerization Using Cp*TiX2(hydroxyl-
aminato) Complexes. In liquid propylene, activation of the Cp*
titanium dialkyl complexes 5, 8, and 9, bearing an η2-bound
hydroxylaminato ligand, with [Ph3C]+[B(C6F5)4]- at room
temperature in the presence of AliBu3 leads to highly active
catalysts for propylene polymerization and yields high molecular
weight atactic polypropylene.20 Under these conditions, narrow
molecular weight distributions are observed, indicative of single-
(59) Doxsee, K. M.; Juliette, J. J. J.; Weakley, T. J. R.; Zientara, K.
Inorg. Chim. Acta 1994, 222, 305–315.
(60) Nakamoto, M.; Tilley, T. D. Organometallics 2001, 20, 5515–5517.
(61) Cummings, S. A.; Radford, R.; Erker, G.; Kehr, G.; Frohlich, R.
Organometallics 2006, 25, 839–842.
(62) Grandini, C.; Camurati, I.; Guidotti, S.; Mascellani, N.; Resconi,
L.; Nifant’ev, I. E.; Kashulin, I. A.; Ivehenko, P. V.; Mercandelli, P.; Sironi,
A. Organometallics 2004, 23, 344–360.
(63) Lin, S.; Tagge, C. D.; Waymouth, R. M.; Nele, M.; Collins, S.;
Pinto, J. C. J. Am. Chem. Soc. 2000, 122, 11275–11285.

408 Organometallics, Vol. 28, No. 2, 2009
DoVe et al.
Figure 2. ORTEP diagram of Cp*Ti(Me)2(O(AlMe3)NEt2) (11).
Representative bond lengths (Å) and angles (deg): Ti-C(1)
2.120(2); Ti-C(2) 2.107(2); Ti-N(1) 2.1696(16); Ti-O(1)
2.0027(12);O(1)-N(1)1.4496(18);O(1)-Al1.9311(13);N(1)-Ti-O(1)
41.41(5); C(1)-Ti-C(2) 91.68(9); Ti-O(1)-Al 147.07(7).
Figure 1. Representative activity versus time profiles for
propylene polymerization by Cp*Ti(Me)2(ONEt2), 8, activated
by (a) [Ph3C]+[B(C6F5)4]-/AliBu3, (, and (b) MAO, 0 (Table 1,
entries 6 and 8).
and scavengers. Studies by Stephan,45 Piers,32,44 and Park64 have
revealed that monocyclopentadienyl titanium complexes with
monoanionic coligands undergo a variety of reactions with
alkylaluminum reagents that have helped to identify some of
the catalyst decomposition pathways.
resulted in only a trace amount of polymer. Activation of 8 with
[Ph3C]+[B(C6F5)4]-/AliBu3 yielded an active catalyst, but activa-
tion with [Ph3C]+[B(C6F5)4]- in the presence of AlMe3 as a
scavenger led to an inactive catalyst for propylene polymeri-
zation. Analysis of propylene conversion upon activation of 8
with [Ph3C]+[B(C6F5)4]-/AliBu3 reveals that the polymerization
activity approached a maximum of 10 000 (mol propylene) (mol
Ti)-1 (h)-1 after approximately 6 min, followed by a decrease
in monomer uptake; after 15 min, the activity is half of its
maximum (Figure 1a). In contrast, activation of 8 with MAO
revealed a lower rate of monomer consumption and a gradual
increase in activity over 12 min followed by a constant rate of
monomer conversion (Figure 1b). One interpretation of these
data is that activation of 8 with [Ph3C]+[B(C6F5)4]-/AliBu3
generates a high concentration of active sites that decay over
time, whereas MAO is less efficient in activating the metallocene
initially but results in a constant generation of active sites at a
rate comparable to catalyst deactivation. Activation of the
dichloride precursor 4 with MAO generated a less active catalyst
than the dimethyl 8, but yielded a polymer with similar
molecular weight, molecular weight distribution, and tacticity.
Previous studies on the ethylene/hexene copolymerization
with the TEMPO derivatives 5 and 2 revealed that activation
ofthedialkylderivative5with[(2,6-iPrC6H3)NMe2H]+[B(C6F5)4]-/
AliBu3 yielded a more active catalyst than that obtained upon
activation of the dichloride derivative 2 with MAO. Moreover,
bimodal molecular weight distributions were observed for the
To assess the stability of the hydroxylamines in the presence
of alkylaluminum scavengers, we first investigated reactions of
the neutral hydroxylaminato complexes 5, 8, and 9 with
alkylaluminum reagents. Addition of AlMe3 to Cp*Ti(Me)2-
(ONEt2), 8, led to the formation of a stable trimethylaluminum
adduct, Cp*Ti(Me)2(O(AlMe3)NEt2), 11. X-ray quality crystals
of 11 were grown from a saturated pentane solution at -30 °C;
an ORTEP diagram is shown in Figure 2. The X-ray diffraction
structure of 11 reveals AlMe3 bound to the oxygen of the η2-
bound diethylhydroxylamine ligand. In contrast to complex 920
and the previously reported CpTi(Cl)2(ONMe2),56 the geometry
of the ONEt2 ligand is twisted with respect to the Ti-Cp
centroid vector such that O(1) and N(1) are bound almost
paralleltotheplanedescribedbyTi,C(1),andC(2)[C(2)-Ti-O(1)
o
) 91.63(7) , C(1)-Ti-N(1) ) 89.65(7)o]. The binding of
AlMe3 to the hydroxylamine ligand results in an increase in
the Ti-O distances in 11 (O(1)-Ti ) 2.0027(12) Å) relative
to that of 9 (O-Ti ) 1.895(3) Å)20 and CpTi(Cl)2(ONMe2)
(O-Ti ) 1.866(1)),56 whereas the Ti-N distances are similar
(N-Ti ) 2.1696(16), 2.178(4), and 2.128 Å, respectively).
1
The H NMR spectrum of complex 11 at 25 °C in solution
(Figure S1, Supporting Information) reveals only one set of
resonances for the AlMe3 and N-Et groups, indicating that either
the molecule is symmetric in solution or it is fluxional as a result
of either the fast rotation of the η2-bound hydroxylaminato
ligand or rapid dissociation/reassociation of the AlMe3. Exami-
ethylene/hexene polymers derived from 2/MAO (Mw/Mn
)
18.3), whereas those from 5/[(2,6-iPrC6H3)NMe2H]+[B(C6F5)4]-/
AliBu3 were narrower (Mw/Mn ) 3.1). For the propylene
polymerizations reported here, monomodal molecular weight
distributions were observed irrespective of the activator ([Ph3C]+-
[B(C6F5)4]-/AliBu3 vs B(C6F5)3/AliBu3 vs MAO); nevertheless
the bimodal molecular weight distributions observed in EH
copolymerizations imply that under some conditions activation
by MAO can generate multiple active sites, presumably by
transmetalation of the hydroxylamine ligand from Ti to Al.
Model Studies. In an effort to identify the nature of the
organometallic species generated upon reaction of the mono-
cyclopentadienyl precursors with either alkylaluminum scav-
engers or activators, we carried out model studies on the
interaction of the catalyst precursors with selected activators
1
nation of the H NMR spectrum at -80 °C reveals only slight
broadening of the resonances, suggesting that even under these
conditions the dynamic process is fast. Nevertheless, the
multiplet assigned to the NCH2 protons at δ ) 3.21 ppm and
triplet at δ ) 0.85 ppm for 11 are indicative of an η2-bound
hydroxylaminato ligand.
To assess the stability of the AlMe3 adducts, we examined
the interaction of 11 with a range of organic ethers (Scheme
2). Addition of 1 equiv of Et2O to 11 led to the complete
(64) Park, J. T.; Yoon, S. C.; Bae, B. J.; Seo, W. S.; Suh, I. H.; Han,
T. K.; Park, J. R. Organometallics 2000, 19, 1269–1276.

Cp*Ti(hydroxylaminato) Catalysts for Propylene Polymerization
Organometallics, Vol. 28, No. 2, 2009 409
Scheme 2. Reaction of 11 with Ethers
Scheme 3. Activation of 8 and 9 with B(C6F5)3 or [Ph3C]-
[B(C6F5)4]
any 1H NMR resonances of 5, suggesting that the more sterically
hindered TEMPO ligand does not form an adduct with AlMe3.
The reversible formation of AlR3 adducts of 5, 8, and 9 is
analogous to the behavior observed by Park for dimethylsilyl-
bridged tetramethylcyclopentadienylhydroazidotitanium com-
plexes (eq 1).64 The η2 hydrazido complex 13 reacts with AlMe3
to form the stable AlMe3 adduct 14, and treatment of 14 with
Et3N regenerates 13. However, though AlMe3 binds to the
oxygen of the η2 hydroxylaminato complexes 5, 8, and 9, it
binds to the terminal Me2N of the hydrazido ligand, inducing a
change in the hapticity of the hydrazido ligand from an η2 to
an η1 coordination geometry.
dissociation of the AlMe3 adduct, whereas the addition of 1
equiv of anisole, PhOMe, led to an equilibrated mixture. At 20
°C the H NMR spectrum of the latter reaction displays only
1
one Cp* resonance at δ ) 1.79 ppm, which is distinct from
Cp* resonance of 8 (δ ) 1.88 ppm) and 11 (δ ) 1.74 ppm),
indicative of a rapid equilibrium. At -80 °C, separate reso-
nances for 8 and 11 could be resolved, enabling a determination
of the equilibrium constant for the reaction of 11 with PhOMe
of Keq ) 0.066 ( 0.003 (-80 °C). In contrast, the addition of
1 equiv of hexamethyldisiloxane led to no significant change
1
in the resonances attributed to 11 in the H NMR spectrum.
This indicates that the bound hydroxylamine of 8 is a signifi-
cantly stronger donor than (Me3Si)2O. This study demonstrates
the order of ether donor strength is Et2O > PhOMe ∼ 8 >
(Me3Si)2O. In addition, this suggests that the AlMe3 can readily
dissociate from 11 in solution.
Addition of 1 equiv of AliBu3 to 8 led to significant shifts in
1
the HNMR spectrum, indicative of the formation of an adduct
analogous to 11. Addition of 1 equiv of anisole to this mixture
regenerated 8 and the anisole adduct of AliBu3, Ph(Me)O · AliBu3.
These data suggest that AlMe3 binds more strongly than AliBu3 to
8.
Reaction of Cp*Ti(Me)2(hydroxylaminato) Complexes
with B(C6F5)3 and [Ph3C]+[B(C6F5)4]-. Treatment of
Cp*Ti(Me)2(ONEt2), 8, with B(C6F5)3 in d8-toluene yielded the
zwitterionic complex [Cp*Ti(Me)(ONEt2)]+[MeB(C6F5)3]-, 15
(Scheme 2). The complex displays two broad resonances
attributable to the ethyl CH2 at δ ) 2.89 and 2.56 ppm, a single
broad resonance due to the ethyl CH3 at δ ) 0.62 ppm, and
Cp*, TiMe, and [MeB(C6F5)3]- resonances occurring at δ )
1.46, 0.45, and 0.73 ppm, respectively. The broadening of the
1H NMR spectra is likely a consequence of a dynamic process
involving migration of the [MeB(C6F5)3]- counterion between
binding sites at the titanium.32 In contrast, treatment of
Cp*Ti(Me)2(ONtBu(Me)), 9, with B(C6F5)3 in d8-toluene resulted
Addition of AlMe3 to Cp*Ti(Me)2(ONtBu(Me)), 9, affords a
related adduct, Cp*Ti(Me)2(O(AlMe3)NtBu(Me)), 12, as evi-
1
denced by significant shifts in the H NMR chemical shifts of
the Cp* and AlMe3 resonances at δ ) 1.82 and -0.31 ppm,
respectively (9: Cp* δ ) 1.87 ppm; free AlMe3 δ ) -0.37
ppm, Figure S1b). Attempts to isolate and crystallize this
complex on a preparatory scale were unsuccessful. The equi-
librium constant for the reaction of anisole with 12 at -80 °C
was measured to be Keq ) 1.74 ( 0.23, a much larger value
that observed for the reaction of 11 with anisole. This result
indicated that the ONEt2 ligand binds more strongly to AlMe3
than the ONtBu(Me) ligand.
Investigations of 11 and 12 in C6D6 over a longer period show
that some degradation of the titanium species occurs. In the
case of 11, degradation is slow. After 3 days approximately
30% Cp*TiMe3 is observed. However, the degradation of 12 is
much more rapid. The 1H NMR spectrum becomes broad after
only a few hours and the reaction becomes increasingly green
in color due to the probable formation of a Ti(III). These
observations are consistent with the transmetalation of the
hydroxylaminato ligand, a possible degradation route for the
catalytic species. Addition of AlMe3 to the η1 hydroxylaminato
complex, Cp*Ti(Me)2(TEMPO), 5, does not cause a shift in
1
in sharp H NMR resonances consistent with formation of the
contact ion pair [Cp*Ti(Me)(ONtBu(Me))]+[MeB(C6F5)3]-, 16
(Scheme 3). This complex was isolated as a pentane-insoluble
orange powder on a 500 mg scale. Investigations by 1H
1
gROESY NMR and H-15N HMBC NMR spectroscopy (15N
δ ) 158 ppm) were consistent with a bidentate η2 coordination
geometry for the hydroxylamine ligand of 16. NOEs were
observed between N-CH3 (δ ) 2.49 ppm) and Ti-CH3 (δ )
0.46 ppm), between N-C(CH3)3 (δ ) 0.46 ppm), and between
Ti-CH3 and N-C(CH3)3 and (C6F5)3B-CH3 (δ ) 0.89 ppm).
The absence of an NOE between N-CH3 and (C6F5)3B-CH3
suggests that the [MeB(C6F5)3]- counterion binds strongly and
t
selectively syn to the Bu group. Similarly, the reaction of 9
with [Ph3C]+[B(C6F5)4]- in d5-chlorobenzene results in the clean

410 Organometallics, Vol. 28, No. 2, 2009
DoVe et al.
Scheme 4. Activation of 5 with B(C6F5)3
this reaction. Of the cation that reacted, no signals attributable
to the hydroxylamine could be observed, and for each hydroxy-
lamine consumed, approximately half an equivalent of methane
is generated. After 2 h, complete decomposition occurs to
unidentified products. Addition of 2 equiv of AlMe3 results in
complete disappearance of resonances due to the hydroxylamine
ligand along with the liberation of approximately 1 equiv of
methane. These results suggest that the hydroxylamine ligands
of the cationic species are much more reactive toward trialky-
laluminums and likely undergo rapid transfer to the aluminum
species along with activation of either the Al-Me or Ti-
Me bonds to liberate methane. The rate of decomposition of
these complexes appears to be sensitive to both the nature of
the activator and the alkylaluminum. We were unable to
characterize the organometallic products of these reactions, but
the liberation of methane is analogous to the reactivity observed
by Piers32 and Stephan,45 where CpTi(Me)2(NPCy3) reacts with
3 equiv of AlMe3 to generate the carbide complexes CpTi(µ2-
Me)(µ2-NPCy3)(µ4-C)(AlMe2)3, 19, along with the liberation of
methane (eq 2).
abstraction of one methyl group from the titanium center with
the elimination of Ph3CMe to yield [Cp*Ti(Me)(ONtBu-
(Me))]+[B(C6F5)4]-, 17. 1H NMR resonances attributable to Cp*,
ONMe, ONtBu, and TiMe are observed at δ ) 1.73, 2.57, 0.55,
1
1
and 0.50 ppm, respectively. H gROESY NMR and H-15N
HMBC NMR spectroscopy confirm that the hydroxylaminato
ligand remains η2 bound to the titanium in 17 (15N δ ) 153
ppm). Examination of the 19F NMR spectrum of 17 indicates
that there is no interaction between the titanium cation and aryl
groups of the borate counterion.65,66
Activation of Cp*Ti(Me)2(TEMPO), 5, with B(C6F5)3 also
results in the clean formation of the bridged zwitterionic
titanium-µ-methylborate species 18, which has undergone C-H
activation of one of the methyl groups of the TEMPO ligand
1
(Scheme 4). The H NMR spectrum reveals the evolution of
methane, and three singlets at δ ) 0.86, 0.65, and 0.50 ppm
attributable to three TEMPO methyl signals and a doublet of
doublet resonance are observed at δ ) 0.80 and -0.59 ppm
Summary
2
with JH-H splitting constant ) 11.4 Hz for the diastereotopic
methylene resonances. The 1H NMR shifts of these resonances
are comparable to those observed for the recently reported C-H
activation of one methyl group on TEMPO upon the reaction
of (TEMPO)Ti(CH2Ph)3 with B(C6F5)3.51 Thus, upon activation
of 5, the zwitterionic product undergoes rapid C-H activation
from one TEMPO-methyl group to result in the formation of
[Cp*Ti(κ2-ON(C(Me)2(CH2)3C(Me)CH2)]+[MeB(C6F5)3]-, 18
We have synthesized a variety of Cp*Ti(hydroxylaminato)
catalysts via several different synthetic routes. These complexes
show a high activity toward the polymerization of propylene in
the presence of [Ph3C]+[B(C6F5)4]- and AliBu3 and show a lower
activity in both the B(C6F5)3/AliBu and MAO catalyst systems.
In contrast, these complexes are inactive toward the polymer-
ization of propylene in the presence of [Ph3C]+[B(C6F5)4]- and
AlMe3. The interaction of AlMe3 with Cp*Ti(Me)2(ONEt2) led
to the formation of Cp*Ti(Me)2(η2-O(AlMe3)NEt2), while AliBu3
was found to bind more weakly to Cp*Ti(Me)2(ONEt2) than
AlMe3. Reaction of Cp*Ti(Me)2(ONEt2) and Cp*Ti(Me)2-
(ONtBu(Me)) with B(C6F5)3 resulted in clean formation of the
1
(Scheme 4). H gROESY NMR data support this conclusion.
Analysis of 18 using 1H-15N HMBC NMR reveals a 15N
resonance at δ ) 163 ppm. Though this is slightly downfield
of other 15N resonances for η2-bound hydroxylaminato ligands,
this is indicative of an η2 binding of the hydroxylaminato ligand
in this complex. Our polymerization studies imply that despite
the C-H activation mechanism, 18 is still able to initiate the
polymerization of propylene.
zwitterionic
contact
ion
pairs
[Cp*Ti(Me)(η2-
ONEt2)]+[MeB(C6F5)3]- and [Cp*Ti(Me)(η2-ONtBu(Me))]+-
[MeB(C6F5)3]-, respectively, whereas reaction of Cp*Ti(Me)2(η2-
ONtBu(Me)) with [Ph3C]+[B(C6F5)4]- resulted in the clean
formation of the solvent-separated ion pair [Cp*Ti(Me)(η2-
ONtBu(Me))]+[B(C6F5)4]-. However, addition of AliBu3 or
AlMe3 to the cationic [Cp*Ti(Me)(ONtBu(Me))]+[B(C6F5)4]-
resulted in decomposition of the cations, likely through rapid
transfer of the aluminum species and activation of either the
Al-Me or Ti-Me bond to eliminate methane. These results
suggest that the hydroxylamine ligands of the cationic species
are much more reactive toward trialkylaluminums than those
in the neutral complex.
Reaction of [Cp*Ti(Me)(ONtBu(Me)]+[B(C6F5)4]- with
Alkylaluminum Complexes. Our investigations on the reactiv-
ity of the neutral complexes 5, 8, and 9 with trialkylaluminum
reagents reveal that the hydroxylaminato ligands do not readily
transmetalate, but rather form stable adducts. However the
cations derived from these complexes are much more reactive
toward alkylaluminum reagents. The cation [Cp*Ti(Me)(ONtBu-
(Me))]+[B(C6F5)]4]-, 17, began to degrade immediately upon
addition of the AliBu3 to give a mixture of isobutylene and other
unidentified products. Resonances due to the hydroxylamine
decrease rapidly, suggesting the decomposition pathway may
involve loss of the nitroxide ligand via transmetalation. After
2 h, complete decomposition occurs. Addition of AlMe3 to the
cation 17 also results in decomposition, but the reaction occurs
more slowly and depends on the Al/Ti stoichiometry. With 1
equiv of AlMe3 only two-thirds of the cation was consumed,
suggesting that more than 1 equiv of AlMe3 is participating in
Experimental Section
General Considerations. All reactions were performed under a
nitrogen atmosphere, using either standard Schlenk techniques or
a dinitrogen-filled glovebox. All solvents and reagents were
purchased from Aldrich unless otherwise stated. Pentane and toluene
were passed through purification columns packed with activated
alumina and copper catalyst, and diethyl ether was passed through
a purification column packed with activated alumina. 2-Methyl-2-
nitrosopropane was sublimed and then recrystallized from dry
diethyl ether prior to use. Diethylhydroxylamine was distilled from
(65) Krossing, I.; Raabe, I. Angew. Chem., Int. Ed. 2004, 43, 2066–
2090.
(66) Jia, L.; Yang, X. M.; Stern, C. L.; Marks, T. J. Organometallics
1997, 16, 842–857.

Cp*Ti(hydroxylaminato) Catalysts for Propylene Polymerization
Organometallics, Vol. 28, No. 2, 2009 411
potassium hydride. Propylene was purchased from Matheson TriGas
(Research Purity) and passed through a purification column packed
with activated alumina and copper catalyst. Cp*TiCl3, 1, was
purchased from Strem, and Cp*TiMe3, 7, was prepared by a
literature procedure.67 Cp*HfCl3 was purchased from Strem, and
Cp*HfMe3 was prepared by a literature procedure.68 Triphenyl-
carbenium tetrakis(pentafluorophenyl)borate and tris(pentafluo-
rophenyl)borane were kindly donated by Albermarle Corporation.
Triphenylcarbenium tetrakis(pentafluorophenyl)borate was repre-
cipitated from a dichloromethane/pentane mixture, and tris(pen-
tafluorophenyl)borane was recrystallized from a saturated pentane
solution at -50 °C. Modified methylaluminoxane (type IV) was
purchased from Akzo Nobel as a solution in toluene; before use,
the volatile materials were removed in Vacuo to yield a powdery
solid.
Physical Methods. NMR spectra were recorded using a Varian
UI-600, UI-500, UI-400, or UI-300 spectrometer and referenced
to the residual proton peaks (C6D6 ) δ 7.15 ppm). 13C NMR spectra
for polymer analysis were obtained at 75.4 MHz using a 10 mm
broadband probe operating at 100 °C. Samples were prepared as
solutions of ca. 80 mg polymer in 2.5 mL of 90:10 (v/v)
1,2-dichlorobenzene/benzene-d6 containing ca. 2 mg of chro-
mium(III) acetylacetonate as a spin relaxation agent. An inverse-
gated decoupled pulse sequence was used. GPC measurements were
taken using a Waters GPC using THF as an eluent. The samples
were analyzed through a Waters Styragel HR5E, HR4 and two HR2
columns and analyzed using a Waters 410 differential refractometer.
The molecular weight of the samples was determined by comparison
to PS standards between 1.25 × 103 and 2.95 × 106 g/mol using
Millenium v3.2. Elemental analyses were performed at Atlantic
Microlabs Ltd. Kinetic measurements were made by drawing gas
from a ballast vessel of known pressure, volume, and temperature,
and results were extrapolated using a previously determined
method.63
General Polymerization Procedures. Liquid Propylene Poly-
merizations. In a 300 mL stainless steel Parr reactor a solution of
60 mg of AliBu3 (0.30 mmol) in 8 mL of toluene was equilibrated
for 30 min at 20 °C with 90 mL of liquid propylene. A solution of
catalyst in toluene (1 mL) was injected into the reactor followed
by a solution of trityl borate (1 mL) via argon pressure. The
polymerization was run for 20 min and quenched by addition of
methanol (10 mL). The polymer was precipitated from acidified
methanol, filtered, washed with methanol, and dried under vacuum
at 60 °C. Solution Propylene Polymerizations, Kinetic Studies. In
a 300 mL stainless steel Parr reactor a solution of 60 mg of AliBu3
(0.30 mmol) in 98 mL of toluene was equilibrated for 30 min at
20 °C. A solution of catalyst in toluene (1 mL) was injected into
the reactor followed by a solution of trityl borate (1 mL) via a
small vent (1-2 psig) and injection with propylene pressure. The
polymerization was run for 1 h and quenched by addition of
methanol (10 mL). The polymer was precipitated from acidified
methanol, filtered, washed with methanol, and dried under vacuum
at 60 °C.
Synthesis of Cp*Ti(Cl)2(ONEt2) (4). n-Butyl lithium (2.5 M
in hexanes, 1.36 mL, 3.40 mmol) was added dropwise to a stirred
solution of HONEt2 (0.35 mL, 3.40 mmol) in toluene (50 mL) at
0 °C. The solution was stirred for ca. 30 min before addition to a
rapidly stirred suspension of Cp*TiCl3 (0.99 g, 3.40 mmol) in
toluene (70 mL) at -78 °C. The solution was stirred for 4 h at
room temperature before being filtered and reduced under vacuum
to yield an orange solid (0.94 g, 3.22 mmol, 94.7%). Crystals were
grown from a saturated pentane solution at -30 °C.
1H NMR (C6D6, 500 MHz): δ 3.10, 3.06 (ddq, ONCH2CH3, 3JH-H
2
) 6.3, 6.5, 7.0, 7.3 Hz, JH-H ) 13.9, 14.0 Hz, 4H), 1.97 (s,
C5(CH3)5, 15H), 0.94 (t, ONCH2CH3, 3JH-H ) 7.3 Hz, 6H). 13C{1H}
NMR (C6D6, 500 MHz): δ 129.4 (C5(CH3)5), 51.0 (ON CH2CH3),
12.2 (C5(CH3)5), 10.2 (ONCH2CH3). 15N NMR (C6D6, 600 MHz):
δ 143. Anal. Calcd for C14H25ONTiCl2 (found): C, 49.13 (49.80);
H, 7.37 (7.36); N, 4.09 (4.34).
Synthesis of Cp*Ti(Me)2(ONEt2) (8). HONEt2 (0.35 mL, 3.4
mmol) was added dropwise to a stirred solution of Cp*TiMe3 (0.78
g, 3.4 mmol) in pentane (70 mL) at 0 °C. The solution was stirred
for 4 h before being reduced under vacuum to yield a yellow oil.
The oil was washed in 10 mL of pentane at -78 °C and yielded a
yellow solid (0.78 g, 2.6 mmol, 76%). Crystals were grown from
a saturated pentane solution at -30 °C.
1H NMR (C6D5CD3, 300 MHz): δ 3.16, 3.00 (ddq, ONCH2CH3,
2
3JH-H ) 6.8, 7.3 Hz, JH-H ) 13.5 Hz, 4H), 1.86 (s, C5(CH3)5,
3
15H), 1.07 (t, ONCH2CH3, JH-H ) 7.2 Hz, 6H), 0.05 (s, TiCH3,
6H). 13C{1H} NMR (C6D6, 500 MHz): δ 120.2 (C5(CH3)5), 51.4
(ON CH2CH3), 46.6 (TiCH3), 11.3 (C5(CH3)5), 9.8 (ONCH2CH3).
15N NMR (C6D6, 600 MHz): δ 129. Anal. Calcd for C16H31ONTi
(found): C, 63.78 (63.72); H, 10.37 (10.34); N, 4.65 (4.61).
Synthesis of Cp*Ti(Me)2(ONtBu(Me)) (9). A solution of
2-nitroso-2-methylpropane (0.39 g, 4.5 mmol) in pentane (20 mL)
was added slowly to a stirred solution of Cp*TiMe3 (1.00 g, 4.4
mmol) in pentane (70 mL) at 0 °C. The solution was stirred for
4 h before being reduced under vacuum to yield a pale yellow solid
(0.97 g, 3.1 mmol, 70%). X-ray quality crystals were grown from
a saturated pentane solution at -30 °C.
1H NMR (C6D6, 500 MHz): δ 2.88 (s, ONCH3, 3H), 1.87 (s,
C5(CH3)5, 15H), 1.06 (s, ONC(CH3)3, 9H), 0.47, -0.03 (s, TiCH3,
3H). 13C{1H} NMR (C6D6, 500 MHz): δ 120.5 (C5(CH3)5), 62.9
(ONC(CH3)3), 52.4, 44.3 (Ti CH3), 43.7 (ON CH3), 26.4 (ON-
C(CH3)3), 11.5 (C5(CH3)5). 15N NMR (C6D6, 600 MHz): δ 134.
Anal. Calcd for C17H33ONTi (found): C, 64.75 (64.56); H, 10.55
(10.58); N, 4.44 (4.49).
Synthesis of Cp*Hf(Me)2(ONtBu(Me)) (10). A solution of
2-nitroso-2-methylpropane (0.081 g, 0.93 mmol) in pentane (10 mL)
was added slowly to a stirred solution of Cp*HfMe3 (0.33 g, 0.93
mmol) in pentane (20 mL) at 0 °C. The solution was stirred for
4 h before being reduced under vacuum to yield a white solid (0.27
g, 0.61 mmol, 66%). Crystals were grown from a saturated pentane
solution at -30 °C.
1H NMR (C6D6, 500 MHz): δ 2.64 (s, ONCH3, 3H), 1.97 (s,
C5(CH3)5, 15H), 0.98 (s, ONC(CH3)3, 9H), -0.035, -0.28 (s,
HfCH3, 3H). 13C{1H} NMR (C6D6, 500 MHz): δ 117.1 (C5(CH3)5),
61.9 (ONC(CH3)3), 44.4, 42.8 (HfCH3), 40.9 (ONCH3), 25.7
(ONC(CH3)3), 10.9 (C5(CH3)5). Anal. Calcd for C17H33ONHf
(found): C, 45.79 (45.50); H, 7.46 (7.43); N, 3.14 (2.99).
Synthesis of Cp*Ti(Me)2(O(AlMe3)NEt2) (11). A solution of
AlMe3 (0.1 g, 1.43 mmol) in pentane (15 mL) was added dropwise
to a stirred solution of 8 (0.43 g, 1.43 mmol) in pentane (30 mL)
in a drybox. The solution was removed from the drybox and stirred
for 20 min before being reduced under vacuum to yield a yellow
solid (0.44 g, 1.18 mmol, 82.5%). Crystals were grown from a
saturated pentane solution at -30 °C.
1H NMR (C6D6, 500 MHz): δ 3.51, 2.90 (ddq, ONCH2CH3, 3JH-H
) 13.9, 14.1 Hz, 2JH-H ) 6.8, 7.0 Hz, 4H), 1.74 (s, C5(CH3)5, 15H),
3
0.85 (t, ONCH2CH3, JH-H ) 7.3 Hz, 6H), 0.48 (s, Ti CH3, 6H),
-0.21 (s, Al(CH3)3, 9H). 13C{1H} NMR (C6D6, 500 MHz): δ 123.9
(C5(CH3)5), 62.06 (TiCH3), 53.6 (ONCH2CH3), 12.1 (C5(CH3)5),
11.2 (ONCH2CH3), -4.0 (Al(CH3)3).
NMR Scale Synthesis of Cp*Ti(Me)2(O(AliBu3)NEt2). 1H
(67) Mena, M.; Royo, P.; Serrano, R.; Pellinghelli, M. A.; Tiripicchio,
A. Organometallics 1989, 8, 476–482.
(68) Schock, L. E.; Marks, T. J. J. Am. Chem. Soc. 1988, 110, 7701–
7715.
3
NMR (C6D6, 500 MHz): δ 3.30, 3.01 (ddq, ONCH2CH3, JH-H
)
7.5 Hz, 2JH-H ) 13.5 Hz, 4H), 2.11 (sept, AlCH2CH(CH3)2, 3JH-H
) 13.5 Hz, 3H), 1.81 (s, C5(CH3)5, 15H), 1.18 (d, AlCH2CH(CH3)2,

412 Organometallics, Vol. 28, No. 2, 2009
DoVe et al.
3
3JH-H ) 6.6 Hz, 6H), 0.93 (t, ONCH2CH3, JH-H ) 7.2 Hz, 6H),
13C{1H} NMR (C6D5Cl, 600 MHz): δ 149.3-125.8, (Ar), 125.6
(C5(CH3)5), 74.2 (ON C(CH3)3), 65.0 (Ti CH3), 43.3 (ON CH3),
30.5 (ONC(CH3)3), 11.5 (C5(CH3)5). 15N NMR (C6D5Cl, 600 MHz):
δ 153. 19F NMR (C6D5Cl, 400 MHz): δ -132.2, -162.7, -166.6.
3
0.31 (s, Ti CH3, 6H), 0.30 (d, AlCH2CH(CH3)2, JH-H ) 9.6 Hz,
6H).
Synthesis of [Cp*Ti(Me)(ONtBu(Me))]+[MeB(C6F5)3]- (16).
A solution of B(C6F5)3 (0.54 g, 1.05 mmol) in pentane (10 mL)
was added slowly to a stirred solution of 9 (0.34 g, 1.07 mmol) in
pentane (50 mL) at 0 °C. The solution was stirred for 1 h before
the colorless solution was filtered from the orange precipitate. The
solid was dried in Vacuo (0.785 g, 0.95 mmol, 90.4%).
1H NMR (C6D5CD3, 500 MHz): δ 2.49 (s, ONCH3, 3H), 1.50
(s, C5(CH3)5, 15H), 0.88 (br s, CH3BAr3, 3H), 0.46 (s, ONC(CH3)3,
9H), 0.46 (s, TiCH3, 3H). 13C{1H} NMR (C6D5CD3, 500 MHz): δ
150.2, 147.0, 140.8, 139.2, 137.7, 135.7 (CAr) 137.3 (C5(CH3)5),
71.2 (ONC(CH3)3), 63.9 (TiCH3), 42.8 (ONCH3), 24.4 (ON-
C(CH3)3), 15.1 (br CH3BAr3), 11.3 (C5(CH3)5). 15N NMR (C6D5CD3,
600 MHz): δ 153. 19F NMR (C6D6, 400 MHz): δ -132.9, -160.6,
-165.3 (CH3B(C6 F5)3). Anal. Calcd for C35H33ONTiF15B (found):
C, 50.78 (48.94); H, 4.02 (3.93); N, 1.69 (1.86).
1
[Cp*Ti(Me)(TEMPO)]+[MeB(C6F5)3]- (18). H NMR (C6D6,
600 MHz): 1.51 (s, C5(CH3)5, 15H), 1.39-1.10 (m, CH2, 6H), 1.09
(br s, CH3BAr3, 3H), 0.86, 0.65, 0.50 (s, TEMPO-CH3, 3H), 0.80,
-0.59 (dd, TEMPO-CH2Ti, 2H). 15N NMR (C6D6, 600 MHz): δ
162.
Crystal Data for 17. See Supporting Information for details:
C19H40AlNOTi, M ) 373.4, rhombic, a ) 9.143(1), b ) 18.171(1),
c ) 13.708(1) Å, U ) 2253.8(2) Å3, T ) 166 K, space group P21/
n, Z ) 4, µ(0.71073 Å radiation) 0.42 cm-1, 13 110 reflections
measured, 4601 unique (Rint ) 0.0261), which were used in all
calculations, residuals: R1; wR2 0.0368; 0.1011.
Acknowledgment. We acknowledge the NSF (CHE-
0611563) for financial support. Albermarle Inc. is thanked
for their donation of borate activators. Dr. Stephen Lynch is
gratefully acknowledged for assistance with NMR measure-
ments. We thank Dr. Frederick J. Hollander and Allen G.
Oliver of the UC Berkeley Diffraction Facility (CHEXRAY)
for collection of X-ray diffraction data.
NMR Scale Activations. The borane and titanium complexes
were combined in ca. 0.6 mL of C6D6, C6D5CD3, or C6D5Cl in a
scintillation vial before being transferred to a J-Young NMR tube.
Spectra were collected immediately.
[Cp*Ti(Me)(ONEt2)]+[MeB(C6F5)3]- (15). 1H NMR (C6D5CD3,
600 MHz): δ 2.89, 2.56 (br s, ONCH2CH3, 4H), 1.46 (s, C5(CH3)5,
15H), 0.73 (br s, CH3BAr3, 3H), 0.62 (br s, ONCH2CH3, 6H), 0.45
(s, TiCH3, 3H).
Supporting Information Available: A CIF file and crystal-
lographic details for 17 are available free of charge via the Internet
at http://pubs.acs.org.
[Cp*Ti(Me)(ONtBu(Me))]+[B(C6F5)4]- (17). 1H NMR (C6D5Cl,
600 MHz): δ 6.97-7.15 (m, Ar, 15H), 2.57 (s, ONCH3, 3H), 1.73
(s, C5(CH3)5, 15H), 0.55 (s, ONC(CH3)3, 9H), 0.50 (s, TiCH3, 3H).
OM800571J