Rapid access to dimethylcyclopentadienyltitanium(IV) amidinate, (C5R5)TiMe2[NR1C(R 2)NR3] (R = H and Me; R2 = Me), libraries

Article abstract of DOI:10.1021/om980706q

Facile insertion of carbodiimides, R¹N=C=NR³, into a Ti-C_Me bond of (C₅R₅)TiMe₃ (R = H and Me) (2a and 2b, respectively) in pentane solutions at 25 °C provides a wide range of derivatives of (C₅R₅)TiMe₂-[NR¹C(Me)NR³] (1) in high yield. Low barriers to racemization (≤15 kcal mal^-1) have been determined for derivatives of 1 where R¹ ≠ R³, and further, a preliminary screen has found some of these to be Ziegler-Natta catalyst precursors for the polymerization of ethylene upon activation with methylaluminoxane.

Full text of DOI:10.1021/om980706q

5228
Organometallics 1998, 17, 5228-5230
Ra p id Access to Dim eth ylcyclop en ta d ien yltita n iu m (IV)
Am id in a te, (C5R5)TiMe2[NR1C(R2)NR3] (R ) H a n d Me; R2
) Me), Libr a r ies
Lawrence R. Sita* and J ason R. Babcock
Searle Chemistry Laboratory, Department of Chemistry, The University of Chicago,
5735 South Ellis Avenue, Chicago, Illinois 60637
Received August 18, 1998
Summary: Facile insertion of carbodiimides, R1NdCd
NR3, into a Ti-CMe bond of (C5R5)TiMe3 (R ) H and
Me) (2a and 2b, respectively) in pentane solutions at 25
°C provides a wide range of derivatives of (C5R5)TiMe2-
[NR1C(Me)NR3] (1) in high yield. Low barriers to
racemization (e15 kcal mol-1) have been determined for
derivatives of 1 where R1 * R3, and further, a prelimi-
nary screen has found some of these to be Ziegler-Natta
catalyst precursors for the polymerization of ethylene
upon activation with methylaluminoxane.
a large number of derivatives is to identify synthetic
methods that are amenable to providing a diverse range
of supported amidinate metal complexes in high yield,
and preferably, in a few number of steps, for the rapid
production and screening of catalytic properties by
combinatorial techniques.5 Herein, we identify one such
method that involves facile insertion of carbodiimides,
R1NdCdNR3, into a Ti-CMe bond of the readily avail-
able trimethylcyclopentadienyl titanium complexes
(C5R5)TiMe3 (R ) H and Me) (2a and 2b, respectively),6
to directly provide a range of new derivatives of 1 for
M ) Ti, R ) H or Me, and X ) R2 ) Me according to
Scheme 1 and Tables 1 and 2. As the series of racemic
asymmetric complexes 1b and 1d -j represent the first
examples of amidinates that possess a chiral titanium
metal center and as the configurational stability of Cp-
based Ziegler-Natta catalysts are known to directly
influence the degree of stereochemical control (i.e.,
tacticity) that can be achieved in the polymerization of
R-olefins,1 barriers to racemization for these complexes
were determined by variable-temperature NMR, which
serves to provide insights regarding the influence of
steric and electronic effects of R1 and R3 on this
parameter. Finally, by using a minimal set of conditions
under which catalytic activity might be expressed, a
preliminary screen found that two members of the
library of compounds listed in Table 1, 1a and 1b, are
indeed active for the polymerization of ethylene in the
presence of MAO, with the level of this activity being
closely tied to both the steric and electronic nature of
the organic substituents on the amidinate ligand. With
the availability of polymer-supported carbodiimides,7
the results reported here now provide a strong founda-
tion and impetus for the directed combinatorial search
for new R-olefin polymerization catalyst precursors
based on the structure of 1.
The high activities and stereochemical control exhib-
ited by many d0 group 4 mono- and biscyclopentadienyl
(Cp) metal-based catalysts in the Ziegler-Natta polym-
erization of R-olefins have generated considerable aca-
demic and commercial interest in the search for addi-
tional Cp and non-Cp metal complexes that can serve
in a similar capacity.1-3 In this regard, symmetric
mono-Cp, mono-N,N′-bis(trimethylsilyl)benzamidinate
metal complexes of the general structure (C5R5)MX2-
[NR1C(R2)NR3] (1), where M ) Ti or Zr; R ) H or Me;
X ) Cl or Me; R1 ) R3 ) SiMe3, and R2 ) Ph, are known
to function as Ziegler-Natta catalyst precursors upon
activation by methylaluminoxane (MAO).4 Given this,
the next logical step of inquiry would be to systemati-
cally vary both the steric and electronic nature of the
amidinate organic substituents, R1, R2 and R3, in 1 over
a wide range in order to realize the full potential of this
class of compound for the polymerization of R-olefins.
Nowadays, one attractive approach to achieve this for
(1) For
a recent review, see: Brintzinger, H. H.; Fischer, D.;
Mu¨lhaupt, R.; Rieger, B.; Waymouth, R. M. Angew. Chem., Int. Ed.,
Engl. 1995, 34, 1143-1170.
(2) For recent examples of Cp-based catalyst precursors, see: (a)
Amor, F.; du Plooy, K. E.; Spaniol, T. P.; Okuda, J . J . Organomet.
Chem. 1998, 558, 139-146. (b) Sinnema, P.-J .; Liekelema, K.; Staal,
O. K. B.; Hessen, B.; Teuben, J . H. J . Mol. Catal. A: Chem. 1998, 128,
143-153. (c) Nomura, K.; Naga, N.; Miki, M.; Yanagi, K.; Imai, A.
Organometallics 1998, 17, 2152-2154. (d) Doherty, S.; Errington, R.
J .; J arvis, A. P.; Collins, S.; Clegg, W.; Elsegood, M. R. J . Organome-
tallics 1998, 17, 3408-3410, and references therein.
Carbodiimides have previously been shown to insert
into a metal-carbon bond of early transition and group
13 organometallic compounds to form amidinate com-
(3) For recent examples of non-Cp-based catalyst precursors, see:
(a) Scollard, J . D.; McConville, D. H. J . Am. Chem. Soc. 1996, 118,
10008-10009. (b) Shah, S. A. A.; Dorn, H.; Voigt, A.; Roesky, H. W.;
Parisini, E.; Schmidt, H. G.; Noltemeyer, M. Organometallics 1996,
15, 3176-3181. (c) Baumann, R.; Davis, W. M.; Schrock, R. R. J . Am.
Chem. Soc. 1997, 119, 3830-3831, and references therein. (d) Kim, I.;
Nishihara, Y.; J ordan, R. F.; Rogers, R. D.; Rheingold, A. L.; Yap, G.
P. A. Organometallics 1997, 16, 3314-3323. (e) Gibson, V. C.; Kim-
berely, B. S.; White, A. J . P.; Williams, D. J .; Howard, P. Chem.
Commun. 1998, 313-314.
(4) (a) Chernega, A. N.; Go´mez, R.; Green, M. L. H. J . Chem. Soc.,
Chem. Commun. 1993, 1415-1417. (b) Buijink, J . K.; Noltemeyer, M.;
Edelmann, F. T. J . Naturforsch 1991, 46B, 1328-1332. (c) Go´mez, R.;
Green, M. L. H.; Haggitt, J . L. J . Chem. Soc., Chem. Commun. 1994,
2607-2608. (d) Go´mez, R.; Duchateau, R.; Chernega, A. N.; Teuben,
J . H.; Edelmann, F. T.; Green, M. L. H. J . Organomet. Chem. 1995,
491, 153-158.
(5) For some recent examples, see: (a) Sigman, M. S.; J acobsen, E.
N. J . Am. Chem. Soc. 1998, 120, 4901-4902. (b) Shimizu, K. D.; Cole,
B. M.; Krueger, C. A.; Kuntz, K. W.; Snapper, M. L.; Hoveyda, A. H.
Angew. Chem., Int. Ed., Engl. 1997, 36, 1704-1707. (c) Gilbertson, S.
R.; Wang, X. Tetrahedron Lett. 1996, 37, 6475-6478.
(6) For 2a , see: Giannini, U.; Cesca, S. Tetrahedron Lett. 1960, 14,
19-20. Compound 2b is commercially available from Strem Chemicals,
Inc.
(7) (a) Weinshenker, N. M.; Shen, C.-M. Tetrahedron Lett. 1972, 32,
3281-3284. (b) Drewry, D. H.; Gerritz, S. W.; Linn, J . A. Tetrahedron
Lett. 1997, 38, 3377-3380. (b) Gelbard, G.; Vielfaure-J oly, F. Tetra-
hedron Lett. 1998, 39, 2743-2746.
10.1021/om980706q CCC: $15.00 © 1998 American Chemical Society
Publication on Web 11/06/1998

Communications
Organometallics, Vol. 17, No. 24, 1998 5229
Sch em e 1
having occurred. On a preparative scale, simple re-
moval of the volatiles provided the desired compounds
listed in Table 1 in essentially pure form as either dark-
red or orange crystalline solids that provided analyti-
cally pure samples upon recrystallization from pentane
at -35 °C.10,11 The derivatives listed in Table 2,
however, possess physical properties, such as noncrys-
tallinity and/or low thermal stability, which have so far
thwarted attempts to secure satisfactory chemical analy-
ses. Nonetheless, for comparison purposes, we feel
justified in including this latter group of compounds for
discussion, as their 1H NMR spectra are exceedingly
clean and strongly supportive of their structural as-
signments being that of 1.11 With regard to the antici-
pated generality of the reaction shown in Scheme 1, it
can be noted that the only limitation that we have
encountered so far is with attempts to prepare deriva-
tives of 1 that bear exceedingly bulky organic substit-
Ta ble 1. An a lytica lly P u r e Der iva tives of 1
P r ep a r ed Accor d in g to Sch em e 1 a n d Ba r r ier s to
Ra cem iza tion a t th e Coa lescen ce Tem p er a tu r e
R1
R3
∆Gcq (kcal mol-1
)
Tc (°C)
R ) H
1a
1b
R ) Me
1c
1d
1e
1f
Cy
Cy
tBu
2,6-Me2C6H3
12.6
233
Cy
Cy
Et
tBu
tBu
tBu
14.9
14.1
15.2
308
295
293
iPr
Cy
t
uents, such as R1 ) R3 ) Bu or SiMe3. In these
instances, the process fails at 25 °C, as insertion is
either extremely slow at this temperature or it does not
occur at all. Unfortunately, at more elevated temper-
atures (e.g., g50 °C), thermal decomposition of the
titanium starting materials now becomes competitive
with productive insertion of these bulky carbodiimides.
A common feature of the variable-temperature 1H
NMR spectra for all the asymmetric derivatives of 1
documented here is the occurrence of two separate
resonances for the diastereotopic methyl groups at-
tached to the chiral titanium center that are observed
at the slow exchange limit for racemization.11 As
temperature is increased, these two resonances are
observed to first coalesce and then to sharpen to a single
strong resonance at the fast exchange limit. Using a
standard proceedure,12 analyses of these spectra yield
barriers to racemization at the coalescence temperature,
Ta ble 2. Ad d ition a l Der iva tives of 1 (R ) H)
P r ep a r ed Accor d in g to Sch em e 1 for Wh ich
Ba r r ier s to Ra cem iza tion a t th e Coa lescen ce
Tem p er a tu r e Ha ve Been Mea su r ed
R1
R3
∆Gcq (kcal mol-1
)
Tc (K)
R ) H
1g
1h
1i
1j
tBu
tBu
tBu
tBu
Et
iPr
Cy
Ph
12.1
13.7
14.7
11.0
243
283
283
223
plexes,8 and as Tables 1 and 2 show, this was also found
to be true for 2a and 2b, where similar insertion
processes occurred rapidly and exothermically with a
range of carbodiimides9 at 25 °C in pentane solution.
Importantly, in each case, 1H NMR (400 MHz, benzene-
d6) spectroscopy revealed that the desired amidinate 1
was produced in a near quantitative yield with no
evidence of multiple insertions of the carbodiimide
q
∆Gc , that are recorded in Tables 1 and 2 for each of
the asymmetric complexes. As can be noted upon
inspection, all the values so far obtained are associated
with barriers that are too low in energy to allow for
configurational stability of these complexes at room
temperature.13 Such low observed values for asym-
metric derivatives of 1 are in keeping with theoretical
calculations of the potential energy surface for [M(bi-
dentate)(unidentate)3] complexes which predict very
shallow barriers to racemization for bidentate ligands
that have a small bite angle, θb, such as that of the
amidinate group, where θb is known to be in the range
60-62°.14,15 Upon further analysis of the experimental
data, however, two interesting trends can be discerned
that bear on the steric and electronic influences of the
R1 and R3 substituents on the racemization barrier of
these compounds. To begin, it can be seen by Table 2
that as the steric bulk of R3 increases, so does the
racemization barrier. An exception to this trend is
noted for R3 ) Ph, which has a smaller barrier than
the case where R3 ) Et; however, increasing the steric
(8) (a) Drew, M. G. B.; Wilkins, J . D. J . Chem. Soc., Dalton Trans.
1974, 1579-1582. (b) Drew, M. G. B.; Wilkins, J . D. J . Chem. Soc.,
Dalton Trans. 1975, 2611-2617. (c) Gambarotta, S.; Strologo, S.;
Floriani, C.; Chiesi-villa, A.; Guastini, C. Inorg. Chem. 1985, 24, 654-
660. (d) Lechler, R.; Hausen, H.-D.; Weidlein, J . J . Organomet. Chem.
1989, 359, 1-12. (e) Coles, M. P.; Swenson, D. C.; J ordan, R. F.
Organometallics 1997, 16, 5183-5194.
t
(9) Except for R3 ) Et, which is commercially available, the BuNd
CdNR3 derivatives used in this study were conveniently obtained in
high yield by tin(II)-mediated heterocumulene metathesis of the
corresponding isocyanates R3NCO using commercially available HN-
(SiMe3)tBu; see: Babcock, J . R.; Sita, L. R. J . Am. Chem. Soc. 1998,
120, 5585-5586.
(10) Representative procedure (conducted in drybox). To a solution
of 140 mg (0.89 mmol) of CpTiMe3 (2a ) in 1 mL of pentane was added
a solution of 183 mg (0.89 mmol) of 1,3-biscyclohexylcarbodiimide in 1
mL of pentane by pipet at 25 °C. After ∼5 min, the reaction mixture
turned red in color, and after 15 min, a brick-red crystalline material
began to precipitate from solution. The reaction mixture was left to
stir for 18 h, and then the solvent removed in vacuo to provide a brick-
red solid, which was recrystallized from pentane at -35 °C to provide
313 mg (97% yield) of 1a as dark-red crystals: 1H NMR (400 MHz,
benzene-d6) δ 1.05 (s, 6H), 0.9-1.2 (m, 6H), 1.35 (dq, 4H, J ) 2.7 Hz,
J ) 11.5 Hz), 1.5-1.7 (m, 10H), 1.66 (s, 3H), 2.95 (tt, 2H, J ) 3.9 Hz,
J ) 11.4 Hz), 6.30 (s, 5H). Anal. Calcd for C21H36N2Ti: C, 69.22; H,
9.96; N, 7.67. Found: C, 69.10; H, 10.07; N, 7.73. In a similar fashion,
1b (from 2a ) and 1c-f (from 2b) were prepared from the corresponding
carbodiimides. For 1H NMR spectra of these compounds, see the
Supporting Information. Anal. Calcd for C21H32N2Ti (1b): C, 70.00;
H, 8.95; N, 7.77. Found: C, 69.52; H, 8.85; N, 8.07. Anal. Calcd for
(11) Detailed information is provided in the Supporting Information.
(12) Sandstro¨m, J . Dynamic NMR Spectroscopy; Academic Press:
New York, 1982.
C
26H46N2Ti (1c): C, 71.87; H, 10.67; N, 6.45. Found: C, 71.62; H, 10.80;
(13) (a) Veciana, J .; Crespo, M. I. Angew. Chem., Int. Ed., Engl. 1991,
30, 74-76. (b) Pirkle, W. H.; Welch, C. J .; Zych, A. J . J . Chromatogr.
1993, 648, 101-109.
N, 6.47. Anal. Calcd. for C24H44N2Ti (1d ): C, 70.57; H, 10.86; N, 6.86.
Found: C, 70.51; H, 11.01; N, 6.80. Anal. Calcd for C21H40N2Ti (1e):
C, 68.46; H, 10.94; N, 7.60. Found: C, 67.83; H, 10.28; N, 7.65. Anal.
Calcd for C20H38N2Ti (1f): C, 67.78; H, 10.81; N, 7.90. Found: C, 67.62;
H, 10.76; N, 7.89.
(14) Kepert, D. L. Inorganic Stereochemistry; Springer-Verlag: New
York, 1982; pp 52-56.
(15) Barker, J .; Kilner, M. Coord. Chem. Rev. 1994, 133, 219-300.

5230 Organometallics, Vol. 17, No. 24, 1998
Communications
size of this aromatic group once more increases barrier
height (cf. 1b vs 1j). In comparing the Cp and Cp* (C5-
Me5) derivatives, in each case, there is an increase in
barrier height on going from the former to the latter
with the same R3 group, but note that now between 1d
and 1f there is not much change on going from R3 ) Et
to R3 ) Cy, which suggests that the steric effects of the
amidinate substituents quickly become “saturated” in
the Cp* series. These steric effects further suggest that
the mechanism by which racemization occurs is through
a “flipping” of the amidinate group that most likely
proceeds via a distorted trigonal bipyramidal state in
which steric interactions between R3 and the cyclopen-
tadienyl group are augmented relative to a preferred
square-pyramidal configuration of these complexes.16
Further, aromatic substituents on the amidinate might
lower this barrier by encouraging an unsymmetric mode
of amidinate binding, and structural studies are now
being pursued in order to verify this.
duce a new method for the direct production of a
subclass of Ziegler-Natta catalyst precursors based on
1 that is characterized by extremely high yields and the
absence of problematic side products that might also
express catalytic activity. Importantly, this methodol-
ogy should be quite compatible with libraries of sup-
ported carbodiimides to provide a diverse range of
supported derivatives of 1 for rapid screening. Although
the activity for PE production reported here is only
modest for 1a relative to what can be currently achieved
with known d0 group 4 metal complexes, it is important
to point out that our polymerization studies have so far
been conducted with only a very limited library of
derivatives obtained by classical preparative methods,
and we have employed only a simple screen for activity.
It can be anticipated, therefore, that much higher
activities might be achievable with an expanded library
of 1, higher MAO/1 ratios, the isolation of discrete
“single-site” cationic complexes from 1 possessing non-
coordinating anions, and the formation of zirconium-
based derivatives of 1 using a reaction analogous to
Scheme 1 with known trimethylcyclopentadienylzir-
conium complexes, (C5R5)ZrMe3 (R ) H and Me).1,18
Studies along these lines are now in progress.
In a preliminary screen, studies were undertaken to
assess the ability of the analytically pure complexes
1a -f to function as Ziegler-Natta catalyst precursors.17
Gratifyingly, it was found that upon activation by only
80-90 equiv of MAO in toluene, which yielded golden-
yellow homogeneous solutions in each case, 1a was
capable of polymerizing ethylene at 25 °C with a modest
Note Ad d ed in P r oof: The crystal structures of 1a -d
and 1q have now been determined, and full structural
analyses will be detailed elsewhere.
level of activity (22 kg PE molcat h-1 atm-1), whereas
-1
1b/MAO was substantially less active at 0.1 kg PE
-1
molcat h-1 atm-1
. Unfortunately, none of the Cp*
Ack n ow led gm en t. This work was supported in part
by the Petroleum Research Fund and by the MRSEC
program under the National Science Foundation (DMR-
9400379), for which we are grateful.
derivatives, 1c-f, were found to be active in this
screening.
In conclusion, the present work has served to intro-
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra of
compounds 1b-j in both the fast and slow exchange regions
(16 pages). Ordering information is given on any current
masthead page.
(16) A similar amidinate flipping has been observed for the inter-
conversion of diasteromers of the chiral molybdenum complex
(C5H5)Mo(CO)2[N(CH2Ph)C(Me)N{(S)-CH(Et)Ph}]; see: Brunner, H.;
Agrifoglio, G. J . Organomet. Chem. 1980, 202, C43-C48.
(17) In a typical procedure, a solution of 20 mg (55 µmol) of 1a in 1
mL of toluene was rapidly mixed with 50 mL of a 10 wt % solution of
MAO (∼80 equiv) in toluene in a pressure-reaction vessel that was
charged with ethylene (2 atm). After 10 min, the vessel was quickly
vented and the reaction quenched by pouring into a large volume (600
mL) of aqueous 10% HCl. After stirring 1 h, the PE was collected,
sequentially washed with water, methanol, acetone, and pentane, and
dried at 70 °C (10-3 Torr) for 48 h to provide 400 mg of polymer.
OM980706Q
(18) For CpZrMe3, see: (a) Aitken, C.; Barry, J .-P.; Gauvin, F.;
Harrod, J . F.; Malek, A.; Rousseau, D. Organometallics 1989, 8, 1732-
1736. For Cp*ZrMe3, see: (b) Wolczanski, P. T.; Bercaw, J . E.
Organometallics 1982, 1, 793-799