Titanium and zirconium complexes that contain the tridentate diamido ligands [(i-PrN-o-C6H4)2O]2- ([i-PrNON]2-) and [(C6H11N-o- C6H4)2O]2- ([CyNON]2-)

Article abstract of DOI:10.1021/ja983549v

A variety of five- and six-coordinate titanium and zirconium dialkyl complexes that contain the [(i-PrN-o-C₆H₄)₂O]^2- ([i-PrNON]^2-) ligand have been prepared, among them [i-PrNON]Ti(CH₂CHMe₂)₂, [i- PrNON]Ti(CH₂CMe₃)₂, [i-PrNON]Zr(CH₂CH₃)₂, and [i- PrNON]Zr(CH₂CHMe₂)₂. These species serve as sources of complexes such as {[i-PrNON]Ti(PMe₃)₂}₂(μ-N₂), [i-PrNON]Ti(CHCMe₃)(PMe₃)₂, [i- PrNON]Zr(CH₂CHMe₂)₂(PMe₃), and [i-PrNON]Zr(H₂C=CMe₂)(PMe₃)₂. The reaction between [i-PrNON]Ti(CH₂CMe₃)₂ and Me₂PCH₂CH₂PMe₂ in the absence of dinitrogen yields (i-PrNC₆H₄)(i-PrNC₆H₄O)Ti(dmpe), a pseudooctahedral species in which one aryl-oxygen bond has been cleaved. In all structurally characterized complexes in which the [i-PrNON]^2- ligand is intact, it adopts a mer configuration in which the donor oxygen atom is planar. Analogous dialkyl complexes [CyNON]MR₂ (Cy = cyclohexyl; M = Zr, R = Me, Et, i-Bu, CH₂CMe₃, allyl; M = Ti, R = Me, CH₂CMe₃, i-Bu) have also been prepared. Decomposition of [i-PrNON]Zr(CH₂CH₃)₂ in the absence of phosphine has been found to proceed in a first-order manner to yield {[iPrNON]Zr(CH₂CH₃)}₂(μ-C₂H₄) via rate-limiting β-hydrogen abstraction to give transient [i-PrNON]Zr(CH₂CH₂) followed by either intermolecular selective β-hydrogen transfer or ethyl group transfer from [i-PrNON]Zr(CH₂CH₃)₂ to [i-PrNON]Zr(CH₂CH₂), as suggested by ²H and ¹³C labeling studies. Decompositions of [i-PrNON]ZrR₂ complexes are dramatically accelerated in the presence of PMe₃. One equivalent of PMe₃ is proposed to bind to give a pseudooctahedral adduct, [i-PrNON]ZrR₂(PMe₃), in which the two alkyl (R) groups are pushed close to one another and β- hydrogen abstraction is thereby accelerated. Consistent with the relatively uncrowded coordination sphere in [i-PrNON]^2- or [CyNON]^2- complexes, activated dialkyls will only oligomerize 1-hexene. Compounds whose structures have been determined include [i-PrNON]Ti(CH₂CHMe₂)₂, [i-PrNON]ZrMe₂. [i- PrNON]Zr(H₂C=CHMe₂)₂(PMe₃), [i-PrNON]Zr(CH₂CMe₂)(PMe₃)₂, [i- PrNON]Ti(PMe₃)₂]₂(μ-N₂), [i-PrNON]Ti(CHCMe₃)(PMe₃)₂, and (i- PrNC₆H₄)(i-PrNC₆H₄O)Ti(dmpe).

Full text of DOI:10.1021/ja983549v

7822
J. Am. Chem. Soc. 1999, 121, 7822-7836
Titanium and Zirconium Complexes That Contain the Tridentate
Diamido Ligands [(i-PrN-o-C6H4)2O]2- ([i-PrNON]2-) and
[(C6H11N-o-C6H4)2O]2- ([CyNON]2-)
Robert Baumann, Ru1diger Stumpf, William M. Davis, Lan-Chang Liang, and
Richard R. Schrock*
Contribution from the Department of Chemistry, 6-331, Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139
ReceiVed October 8, 1998
Abstract: A variety of five- and six-coordinate titanium and zirconium dialkyl complexes that contain the
[(i-PrN-o-C6H4)2O]2- ([i-PrNON]2-) ligand have been prepared, among them [i-PrNON]Ti(CH2CHMe2)2,
[i-PrNON]Ti(CH2CMe3)2, [i-PrNON]Zr(CH2CH3)2, and [i-PrNON]Zr(CH2CHMe2)2. These species serve as
sources of complexes such as {[i-PrNON]Ti(PMe3)2}2(µ-N2), [i-PrNON]Ti(CHCMe3)(PMe3)2, [i-PrNON]Zr-
(CH2CHMe2)2(PMe3), and [i-PrNON]Zr(H2CdCMe2)(PMe3)2. The reaction between [i-PrNON]Ti(CH2CMe3)2
and Me2PCH2CH2PMe2 in the absence of dinitrogen yields (i-PrNC6H4)(i-PrNC6H4O)Ti(dmpe), a pseudo-
octahedral species in which one aryl-oxygen bond has been cleaved. In all structurally characterized com-
plexes in which the [i-PrNON]2- ligand is intact, it adopts a mer configuration in which the donor oxygen
atom is planar. Analogous dialkyl complexes [CyNON]MR2 (Cy ) cyclohexyl; M ) Zr, R ) Me, Et, i-Bu,
CH2CMe3, allyl; M ) Ti, R ) Me, CH2CMe3, i-Bu) have also been prepared. Decomposition of [i-PrNON]-
Zr(CH2CH3)2 in the absence of phosphine has been found to proceed in a first-order manner to yield {[i-
PrNON]Zr(CH2CH3)}2(µ-C2H4) via rate-limiting â-hydrogen abstraction to give transient [i-PrNON]Zr(CH2CH2)
followed by either intermolecular selective â-hydrogen transfer or ethyl group transfer from [i-PrNON]Zr-
(CH2CH3)2 to [i-PrNON]Zr(CH2CH2), as suggested by 2H and 13C labeling studies. Decompositions of [i-PrNON]-
ZrR2 complexes are dramatically accelerated in the presence of PMe3. One equivalent of PMe3 is proposed to
bind to give a pseudooctahedral adduct, [i-PrNON]ZrR2(PMe3), in which the two alkyl (R) groups are pushed
close to one another and â-hydrogen abstraction is thereby accelerated. Consistent with the relatively uncrowded
coordination sphere in [i-PrNON]2- or [CyNON]2- complexes, activated dialkyls will only oligomerize 1-hexene.
Compounds whose structures have been determined include [i-PrNON]Ti(CH2CHMe2)2, [i-PrNON]ZrMe2,
[i-PrNON]Zr(H2CdCHMe2)2(PMe3), [i-PrNON]Zr(CH2CMe2)(PMe3)2, [i-PrNON]Ti(PMe3)2]2(µ-N2), [i-PrNON]-
Ti(CHCMe3)(PMe3)2, and (i-PrNC6H4)(i-PrNC6H4O)Ti(dmpe).
Introduction
is the higher energy species when R ) t-Bu because of steric
interaction between the tert-butyl group and the two equatorial
R′ groups, that a twisted fac structure for the dimethyl complex
is a predictor of steric crowding in a pseudotetrahedral cationic
monoalkyl complex, and that a four-coordinate cationic monoalkyl
species will more likely behave as a living catalyst for
polymerization of ordinary olefins if the twisted fac structure
is the lower energy species. We set out to test this theory by
preparing analogous complexes in which the R group is
isopropyl or cyclohexyl. We have found that only mer structures
are observed in a variety of five-coordinate (and six-coordinate)
titanium and zirconium complexes that contain the [(i-PrN-o-
C6H4)2O]2- ([i-PrNON]2-) ligand, consistent with a natural
preference for twisted mer structures in the sterically less
demanding circumstance. We have also found that “{[i-PrNON]-
ZrMe}[B(C6F5)4]” will only oligomerize 1-hexene, consistent
with the metal center not being sufficiently crowded, and
perhaps far from pseudotetrahedral.
We recently reported the synthesis of titanium, zirconium,
and hafnium complexes that contain the [(t-BuN-o-C6H4)2O]2-
([t-BuNON]2-) ligand, and the living polymerization of up to
∼1000 equiv of 1-hexene by activated zirconium dimethyl
complexes in chlorobenzene at 0 °C.1-3 All five-coordinate
compounds, e.g., [t-BuNON]ZrMe2 and the B(C6F5)3-activated
zirconium dimethyl complex, were shown to be “twisted”,
approximately trigonal bipyramidal species in which the amido
nitrogens occupy equatorial positions and the two R′ groups
are inequivalent (a “twisted fac” structure; eq 1). However, the
(1) Baumann, R.; Davis, W. M.; Schrock, R. R. J. Am. Chem. Soc. 1997,
119, 3830.
(2) Baumann, R.; Schrock, R. R. J. Organomet. Chem. 1998, 557, 69.
(3) Schrock, R. R.; Baumann, R.; Reid, S. M.; Goodman, J. T.; Stumpf,
R.; Davis, W. M. Organometallics, in press.
two R′ groups equilibrate rapidly on the NMR time scale in
solution, even at -80 °C, presumably via a mer structure with
C2 or C2V symmetry. We began to suspect that the mer structure
10.1021/ja983549v CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/11/1999

Ti and Zr Complexes Containing Diamido Ligands
J. Am. Chem. Soc., Vol. 121, No. 34, 1999 7823
Results
Synthesis of Ti and Zr [(i-PrNC6H4)2O]2- Complexes. The
isopropyl derivative of (NH2-o-C6H4)2O, [(i-PrNHC6H4)2O] (H2-
[i-PrNON]), can be prepared readily from acetone and zinc in
acetic acid, and isolated as a pale yellow oil in high yield on a
∼10 g scale. This methodology is derived from that reported
for the synthesis of a variety of substituted anilines.4
Addition of 2 equiv of LiBu to a solution of H2[i-PrNON] in
ether, followed by addition of (NMe2)2TiCl2, yielded the
dimethylamido complex [i-PrNON]Ti(NMe2)2, an orange oil that
was used in crude form for further reactions. Proton NMR
spectra of [i-PrNON]Ti(NMe2)2 suggest that it has C2V symmetry
on the NMR time scale; only one singlet at ∼3.1 ppm for the
dimethylamido groups is observed, and the isopropyl methyl
groups are equivalent. The dichloride complex [i-PrNON]TiCl2
was obtained quantitatively in 24 h by treating [i-PrNON]Ti-
(NMe2)2 with Me3SiCl (eq 2). [i-PrNON]TiCl2 is a deep purple-
Li2[i-PrNON]
(NMe2)2TiCl2
8
ether, -25 °C
2Me3SiCl
[i-PrNON]Ti(NMe2)2 toluene, 110 °C8 [i-PrNON]TiCl2 (2)
black solid that is moderately soluble in toluene and ether. A
possible dimeric structure for [i-PrNON]TiCl2 with bridging
chlorides cannot be ruled out on the basis of the observed
dimeric structure for [t-BuNON]ZrCl2.3
Alkylation of [i-PrNON]TiCl2 with 2 equiv of MeMgI,
Me2CHCH2MgCl, or Me3CCH2MgCl yielded the corresponding
dialkyl complexes (eq 3) as orange microcrystalline solids in
Figure 1. (a, top) ORTEP diagram (35% probability level) of
[i-PrNON]Ti(CH2CHMe2)2. (b, bottom) Chem 3D drawing of [i-PrNON]-
Ti(CH2CHMe2)2.
[i-PrNON]TiCl2 2RMgX, -25 °C8 [i-PrNON]TiR2
(3)
to note that [i-PrNON]Zr(CH2CHMe2)2 can be prepared directly
from ZrCl4 in good yield (eq 5). We believe that the product of
R ) Me, CH2CHMe2, or CH2CMe3
1
51-69% yields. At room temperature the H NMR spectra of
[i-PrNON]TiR2 complexes are consistent with their having C2V
symmetry on the NMR time scale in solution. [i-PrNON]Ti-
(CH2CHMe2)2 can be stored as a solid at -25 °C, but it slowly
decomposes at room temperature within ∼1 day to a mix-
ture of unidentified products. [i-PrNON]Ti(CH2CMe3)2 is
relatively stable in the solid state at 22 °C, but it is best
stored at -25 °C. In C6D6 solution [i-PrNON]Ti(CH2CMe3)2
decomposes completely within 12 h at 50 °C to unidenti-
fied products. Attempts to prepare and isolate [i-PrNON]Ti-
(CH2CH3)2 failed.
1. H2[i-PrNON],
toluene, 22 °C, 5 h
ZrCl4 2. 4i-BuMgCl, -25 °C, 30 min8
[i-PrNON]Zr(CH2CHMe2)2 (5)
the reaction between ZrCl4 and H2[i-PrNON] is an adduct,
subsequent dehydrochlorination and alkylation of which by
the Grignard reagent proceed smoothly to give [i-PrNON]Zr-
(CH2CHMe2)2 in a 60% isolated yield on a 1-2 g scale.
The structure of [i-PrNON]Ti(CH2CHMe2)2, as determined
in a single-crystal X-ray study, is a distorted trigonal bipyra-
mid, as shown in Figure 1. (See also Tables 1 and 3.) The
[i-PrNON]2- ligand is coordinated in a meridional manner with
oxygen in the equatorial position and the planar amido nitrogens
in the “axial” positions with a N(1)-Ti-N(2) angle of 145.53-
(14)°. The mer configuration is best characterized by an angle
between the N(1)/Ti/O and N(2)/Ti/O planes of 180°. The two
isobutyl groups are found in equatorial positions, with the C(1)-
Ti-C(5) angle being 104.6(2)°. The [i-PrNON]2- ligand
backbone is twisted (Figure 1b), as shown by O-Ti-N-Ci-Pr
dihedral angles of 172.0° and 168.7°, in part to minimize
unfavorable steric interactions between the hydrogen atoms ortho
to the oxygen atom, but also in part as a consequence of the
inflexible nature of the o-phenylene backbone. The Ti-O
distance is 2.149(3) Å, and the oxygen atom is planar. The Ti-
The zirconium dichloride complex [i-PrNON]ZrCl2 was
prepared via the route shown in eq 4. Dialkyl complexes
H2[i-PrNON] + Zr(NMe2)4 -2Me NH8
2
[i-PrNON]Zr(NMe2)2 excess Me SiCl8
3
[i-PrNON]ZrCl2 alkylation8 [i-PrNON]ZrR2 (4)
R ) Me, Et, i-Bu, CH2CMe3
[i-PrNON]ZrR2, where R ) Me, Et, CH2CHMe2, or CH2CMe3,
could all be prepared in a straightforward manner. The [i-
PrNON]ZrEt2 complex is the least stable toward â-hydrogen
abstraction, as one would predict on the basis of studies
involving zirconocene dialkyl complexes;5,6 decomposition
studies are described in detail in a later section. It is interesting
O
donor distance contrasts with the relatively long Ti-Odonor bond
(5) Negishi, E.; Takahashi, T. Acc. Chem. Res. 1994, 27, 124.
(4) Micovic, I. V.; Ivanovic, M. D.; Piatak, D. M.; Bojic, V. D. Synthesis
1991, 1043.
(6) Negishi, E.; Nguyen, T.; Maye, J. P.; Choueiri, D.; Suzuki, N.;
Takahashi, T. Chem. Lett. 1992, 2367.

7824 J. Am. Chem. Soc., Vol. 121, No. 34, 1999
Baumann et al.
Table 1. Crystal Data and Structure Refinement for [i-PrNON]Ti(CH2CHMe2)2, [i-PrNON]Ti(PMe3)2]2(µ-N2),
(i-PrNC6H4)(i-PrNC6H4O)Ti(dmpe), and [i-PrNON]Ti(CHCMe3)(PMe3)2 (L ) [i-PrNON]2-
)
compound
crystals obtained from
formula
formula weight
crystal size (mm)
crystal system
space group
a (Å)
LTi(CH2CHMe2)2
pentane
C26H40N2OTi
444.50
[LTi(PMe3)2]2(µ-N2)
ether
C48H80N6O2P4Ti2
992.86
0.23 × 0.16 × 0.12
orthorhombic
Pbcn
22.2673(7)
13.6674(5)
19.6101(6)
90
90
90
5968.1(3)
4
1.105
(i-PrNC6H4)(i-PrNC6H4O)Ti(dmpe)
LTi(CHCMe3)(PMe3)2
toluene/pentane
C29H50N2OP2Ti
552.55
0.38 × 0.12 × 0.12
monoclinic
P21/c
16.3753(8)
17.2240(9)
11.5092(6)
90
96.9710(10)
90
toluene/pentane
C24H38N2OP2Ti
480.40
0.32 × 0.28 × 0.25
monoclinic
C2/c
36.7261(13)
10.3024(4)
14.7015(5)
90
109.88
90
5231.2(3)
8
1.220
0.18 × 0.12 × 0.08
triclinic
P1h
9.14700(10)
10.0436(3)
14.2372(4)
79.552(2)
88.5520(10)
77.869(2)
1257.43(5)
2
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
3222.2(3)
4
1.139
0.387
1192
Z
density calcd (Mg/m3)
1.174
0.359
480
183(2)
µ (mm-1
F (000)
T (K)
)
0.412
2120
183(2)
0.467
2048
190(2)
193(2)
θ range (ω scans)
1.45-23.24°
1.75-20.00°
2.06-23.25°
1.25-23.26°
no. of reflections collected
no. of independent reflections
no. of data/restr/parameters
goodness-of-fit on F2
R1/wR2 [I > 2σ(I)]
5146
16332
10398
12888
3527
3524/0/290
1.063
0.0561/0.1408
0.0832/0.1846
0.008(3)
2785
2781/0/281
1.125
0.0598/0.1690
0.0660/0.1829
0.0036(6)
0.499/-0.357
3754
3748/0/272
1.046
0.0443/0.1060
0.0594/0.1213
0.00003(8)
0.420/-0.264
4613
4603/0/344
0.900
0.0521/0.1374
0.0786/0.1762
0.0089(12)
0.350/-0.355
R1/wR2 (all data)
extinction coefficient
max/min peaks (e/Å3)
0.332/-0.345
Table 2. Crystal Data and Structure Refinement for [i-PrNON]ZrMe2, [i-PrNON]Zr(CH2CHMe2)2(PMe3), and
[i-PrNON]Zr(η2-CH2CMe2)(PMe3)2 ([i-PrNON]2- ) L)
compound
crystals obtained from
formula
formula weight
crystal size (mm)
crystal system
space group
a (Å)
LZrMe2
pentane
C20H28N2OZr
403.66
0.38 × 0.12 × 0.12
orthorhombic
P212121
8.415(2)
13.889(3)
16.544(4)
90
90
90
1933.6(8)
4
1.387
LZr(CH2CHMe2)2(PMe3)
pentane
C29H49N2OPZr
563.89
LZr(CH2CMe2)(PMe3)2
toluene/pentane/PMe3
C28H48N2OP2Zr
581.84
0.32 × 0.08 × 0.08
monoclinic
P21/n
10.5432(3)
21.6042(5)
13.2855(2)
90
93.5560(10)
90
0.38 × 0.25 × 0.22
triclinic
P1h
9.8443(2)
9.91370(10)
16.40460(10)
90.0120(10)
97.0780(10)
101.6160(10)
1555.73(4)
2
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
3020.31(12)
4
1.280
0.492
1232
Z
density calcd (Mg/m3)
1.204
0.426
600
293(2)
µ (mm-1
F (000)
T (K)
)
0.577
840
183(2)
183(2)
θ range (ω scans)
1.91-23.26°
1.25-23.24°
1.80-20.00°
no. of reflections collected
no. of independent reflections
data/restraints/parameters
goodness-of-fit on F2
R1/wR2 [I > 2σ(I)]
7753
6414
8975
2771
2771/0/218
1.164
0.0323/0.0655
0.0363/0.0667
0.0027(5)
0.409/-0.533
4354
4350/0/308
1.010
0.0409/0.1097
0.0529/0.1345
0.008(2)
2815
2622/0/308
1.047
0.0409/0.0787
0.0658/0.1120
0.0007(2)
0.453/-0.343
R1/wR2 (all data)
extinction coefficient
max/min peaks (e/Å3)
0.385/-0.614
length (2.402(4) Å) found in [t-BuNON]TiMe2, which has the
fac geometry in the solid state.3
quence of the lower degree of steric crowding in the zirconium
complex between the amido isopropyl groups and the equatorial
alkyl groups. The Zr-Odonor distance (2.309(2) Å) is ∼0.1 Å
shorter than in [t-BuNON]ZrMe2 (2.418(3) Å),3 consistent with
more efficient M-O bonding in the mer structure, perhaps in
part as a consequence of some Zr-O π bonding.
Both titanium and zirconium dialkyl complexes appear to be
sensitive to light, even fluorescent room light, especially in the
presence of phosphines. In several cases (see below) reactions
involving phosphines that were not protected from room light
were compromised to a significant degree by side reactions.
The solid-state structure of [i-PrNON]ZrMe2 is similar to that
of [i-PrNON]Ti(CH2CHMe2)2 (Figure 2, Tables 2 and 3). The
larger size of Zr accounts for the longer metal-ligand bonds
in [i-PrNON]ZrMe2 than in [i-PrNON]Ti(CH2CHMe2)2, and also
for the slightly smaller N-Zr-N angle (137.70(12)°) relative
to the N-Ti-N angle (145.53(14)°). The M-N-Ci-Pr angle
in [i-PrNON]ZrMe2 is also 6-7° smaller than it is in [i-PrNON]-
Ti(CH2CHMe2)2. The O-M-N-Ci-Pr dihedral angles in [i-
PrNON]ZrMe2 are close to 180°, perhaps largely as a conse-

Ti and Zr Complexes Containing Diamido Ligands
Table 3. Selected Bond Lengths (Å) and Angles (deg) in [i-PrNON]Ti(CH2CHMe2)2, [i-PrNON]ZrMe2, and [i-PrNON]Zr(CH2CHMe2)2(PMe3)
[i-PrNON]Ti(CH2CHMe2)2 [i-PrNON]ZrMe2 [i-PrNON]Zr(CH2CHMe2)2(PMe3)
J. Am. Chem. Soc., Vol. 121, No. 34, 1999 7825
Ti-O
2.149(3)
1.993(3)
1.993(3)
2.106(4)
2.099(4)
145.53(14)
72.66(12)
97.2(2)
103.8(2)
72.87(12)
103.8(2)
97.0(2)
127.9(2)
104.6(2)
127.5(2)
125.8(3)
117.6(3)
117.9(3)
180
Zr-O
2.309(2)
2.087(3)
2.087(3)
2.243(4)
2.237(4)
137.70(12)
69.07(10)
101.53(13)
101.40(14)
68.80(11)
101.99(13)
102.42(14)
129.5(2)
110.7(2)
119.81(13)
125.3(3)
111.5(2)
110.3(2)
176
Zr-O
2.329(2)
2.107(3)
2.113(3)
2.280(3)
2.268(4)
137.27(12)
69.51(10)
113.83(12)
93.95(12)
68.76(10)
104.62(13)
99.13(13)
154.69(12)
97.33(13)
107.77(12)
125.9(3)
112.1(2)
114.8(2)
180
Ti-N(1)
Zr-N(1)
Zr-N(2)
Zr-C(2)
Zr-C(1)
Zr-N(1)
Ti-N(2)
Zr-N(2)
Ti-C(5)
Zr-C(1)
Ti-C(1)
Zr-C(5)
N(1)-Ti-N(2)
N(1)-Ti-O
N(1)-Zr-N(2)
N(1)-Zr-O
N(1)-Zr-C(2)
N(1)-Zr-C(1)
N(2)-Zr-O
N(2)-Zr-C(2)
N(2)-Zr-C(1)
C(2)-Zr-O
C(2)-Zr-C(1)
C(1)-Zr-O
C(16)-O-C(26)
Zr-N(1)-C(17)
Zr-N(2)-C(27)
N(1)/Zr/O/N(2)a,b
O-Zr-N(1)-C(17)a
O-Zr-N(2)-C(27)a
N(1)-Zr-N(2)
N(1)-Zr-O
N(1)-Zr-C(1)
N(1)-Zr-C(5)
N(2)-Zr-O
N(2)-Zr-C(1)
N(2)-Zr-C(5)
C(1)-Zr-O
N(1)-Ti-C(5)
N(1)-Ti-C(1)
N(2)-Ti-O
N(2)-Ti-C(5)
N(2)-Ti-C(1)
C(5)-Ti-O
C(5)-Ti-C(1)
C(1)-Ti-O
C(16)-O-C(26)
Ti-N(1)-C(17)
Ti-N(2)-C(27)
N(1)/Ti/O/N(2)a,b
O-Ti-N(1)-C(17)a
O-Ti-N(2)-C(27)a
C(5)-Zr-C(1)
C(5)-Zr-O
C(12)-O-C(22)
Zr-N(1)-C(17)
Zr-N(2)-C(27)
N(1)/Zr/O/N(2)a,b
O-Zr-N(1)-C(17)a
O-Zr-N(2)-C(27)a
Zr-P
172.0
168.7
178.8
176.0
177.5
166.4
3.0326(11)
174.44(10)
77.78(9)
P-Zr-C(5)
P-Zr-C(1)
a Obtained from a Chem 3D model. b The external angle between the planes.
Figure 2. ORTEP diagram (35% probability level) of [i-PrNON]-
ZrMe2.
Figure 3. ORTEP diagram (35% probability level) of [i-PrNON]Zr-
(CH2CHMe2)2(PMe3).
All attempts to characterize the decomposition product or
products of photochemical decomposition reactions failed, even
in the presence of a phosphine such as PMe3 (see below).
and temperature dependent and consistent with ready loss of
PMe3 to yield [i-PrNON]Zr(CH2CHMe2)2.
An X-ray study of a single crystal of [i-PrNON]Zr-
(CH2CHMe2)2(PMe3) (Tables 2 and 3) revealed the pseudooc-
tahedral structure shown in Figure 3. The PMe3 has attacked
the metal from “outside” the C-Zr-C wedge and in the ZrC2
plane (eq 6). The Zr-P bond length is quite long (3.0326(11)
Decomposition of [i-PrNON]Zr(CH2CHMe2)2 in the Pres-
ence of PMe3. Upon addition of PMe3 to a C6D6 solution of
[i-PrNON]Zr(CH2CHMe2)2 a downfield shift of the isobutyl
1
methyl resonances is observed in the H NMR spectrum. A
variable-temperature 31P NMR study of a toluene-d8 solution
of [i-PrNON]Zr(CH2CHMe2)2 in the presence of PMe3 (∼3
equiv) reveals one sharp singlet at -59.3 ppm at room temper-
ature. At -80 °C two broad resonances appear, one at -49.2
ppm for coordinated PMe3, and the other at -61.7 ppm for free
PMe3. Upon cooling a pentane solution of [i-PrNON]Zr-
(CH2CHMe2)2 to -25 °C in the presence of excess PMe3, pale
yellow crystals formed whose elemental analysis is consistent
with the formulation [i-PrNON]Zr(CH2CHMe2)2(PMe3). Proton
and 31P NMR spectra of [i-PrNON]Zr(CH2CHMe2)2(PMe3)
confirm that proposal. However, the spectra are concentration
Å), consistent with the relatively crowded pseudooctahedral
environment and lability of the PMe3 ligand. The [i-PrNON]2-
ligand is coordinated in a twisted mer fashion (N(1)/Zr/O/N(2)

7826 J. Am. Chem. Soc., Vol. 121, No. 34, 1999
Baumann et al.
Table 4. Selected Bond Lengths (Å) and Angles (deg) in [i-PrNON]Zr(CH2CMe2)(PMe3)2, {[i-PrNON]Ti(PMe3)2}2(µ-N2),
[i-PrNON]Ti(CHCMe3)(PMe3)2
[i-PrNON]Zr(CH2CMe2)(PMe3)2
{[i-PrNON]Ti(PMe3)2}2(µ-N2)
[i-PrNON]Ti(CHCMe3)(PMe3)2
Zr-O
2.367(3)
2.170(4)
2.161(4)
68.48(14)
68.26(14)
136.7(2)
118.8(3)
120.7(3)
126.8(4)
178
Ti-O
2.198(3)
2.068(4)
2.123(4)
72.40(13)
72.44(13)
144.6(2)
121.2(3)
130.4(4)
126.5(4)
176
Ti-O
2.202(3)
2.080(3)
2.074(3)
72.81(11)
72.44(10)
145.22(13)
122.8(3)
122.7(2)
128.3(3)
179
Zr-N(1)
Ti-N(1)
Ti-N(1)
Zr-N(2)
Ti-N(2)
Ti-N(2)
N(1)-Zr-O
N(2)-Zr-O
N(1)-Zr-N(2)
Zr-N(1)-C(17)
Zr-N(2)-C(27)
C(11)-O-C(21)
N(1)/Zr/O/N(2)a,b
O-Zr-N(1)-C(17)b
O-Zr-N(2)-C(27)b
Zr-C(2)
N(1)-Ti-O
N(1)-Ti-O
N(2)-Ti-O
N(2)-Ti-N(1)
C(17)-N(1)-Ti
C(27)-N(2)-Ti
C(11)-O-C(26)
N(1)/Ti/O/N(2)a,b
O-Ti-N(1)-C(17)b
O-Ti-N(2)-C(27)b
Ti-C(1)
N(2)-Ti-O
N(1)-Ti-N(2)
Ti-N(1)-C(111)
Ti-N(2)-C(210)
C(11)-O-C(21)
N(1)/Ti/O/N(2)a,b
O-Ti-N(1)-C(111)b
O-Ti-N(2)-C(210)b
Ti-N(3)
162.8
159.6
165.9
171.1
162.6
166.1
2.291(6)
2.790(2)
2.803(2)
1.457(8)
160.55(5)
111.8(2)
110.2(2)
108.1(2)
111.7(2)
159.7(2)
36.8(2)
1.811(4)
2.650(2)
2.668(2)
177.7(4)
166.69(5)
175.5(2)
1.264(8)
1.884(4)
2.6368(12)
2.6396(12)
179.3(3)
169.42(4)
106.83(14)
107.95(15)
94.85(12)
178.20(14)
87.01(9)
89.72(9)
86.78(8)
95.73(12)
89.87(9)
87.09(9)
82.64(8)
Zr-P(2)
Ti-P(2)
Ti-P(3)
Ti-P(1)
Ti-P(2)
Zr-P(3)
C(1)-C(2)
Ti-N(3)-N(3A)
P(2)-Ti-P(3)
N(3)-Ti-O
C(2)-C(1)-Ti
P(1)-Ti-P(2)
C(1)-Ti-N(2)
C(1)-Ti-N(1)
C(1)-Ti-P(1)
C(1)-Ti-O
P(2)-Zr-P(3)
N(1)-Zr-C(2)
N(1)-Zr-C(1)
N(2)-Zr-C(2)
N(2)-Zr-C(1)
C(2)-Zr-O
N(3)-N(3A)
N(2)-Ti-P(1)
N(1)-Ti-P(1)
O-Ti-P(1)
C(1)-Ti-P(2)
N(2)-Ti-P(2)
N(1)-Ti-P(2)
O-Ti-P(2)
C(2)-Zr-C(1)
C(1)-Zr-O
Zr-C(1)
163.5(2)
2.327(5)
a The external angle between the planes. b Obtained from a Chem 3D model.
) 180°), similar to what is found in [i-PrNON]Ti(CH2CHMe2)2
and [i-PrNON]ZrMe2. Metal-ligand bond lengths are all slightly
longer (by ∼0.02-0.04 Å) than they are in [i-PrNON]ZrMe2,
consistent with more steric crowding and greater electron
donation to the metal in the six-coordinate species. Coordination
of PMe3 forces the two isobutyl groups together (C(1)-Zr-
C(5) ) 97.33(13)°), but the weak coordination of PMe3 and
repulsion between the two isobutyl groups prevent a further
decrease in the C-Zr-C angle. The isobutyl group cis to PMe3
turns away from PMe3, while the isobutyl group containing C(5)
turns away from that containing C(1), so that C(6) is oriented
toward the relatively open face of the [i-PrNON]2- ligand. It is
tempting to propose that in the â-hydrogen abstraction reaction
described below the proton on C(2) is transferred to carbon atom
C(5), which is 2.93 Å away from it. The Zr-C(1) and Zr-
C(5) distances are essentially equal, giving no hint as to which
is the weaker Zr-C bond.
When a solution of [i-PrNON]Zr(CH2CHMe2)2 in neat PMe3
Figure 4. ORTEP diagram (35% probability level) of [i-PrNON]Zr-
(η2-CH2CMe2)(PMe3)2.
is heated to 37 °C, the isobutylene adduct [i-PrNON]Zr(η2-
CH2CMe2)(PMe3)2 is formed and can be isolated in ∼59% yield
(eq 7). The reaction must be performed in the absence of light;
reotopic, whereas the isopropyl methine protons are equivalent,
indicating that the molecule has Cs symmetry on the NMR time
scale. While C6D6 solutions of [i-PrNON]Zr(η2-CH2CMe2)-
(PMe3)2 are stable in solution at room temperature for several
hours, significant decomposition is observed at 40 °C. Solutions
of [i-PrNON]Zr(η2-CH2CMe2)(PMe3)2 that contain additional
PMe3 are significantly more stable. Apparently [i-PrNON]Zr-
(η2-CH2CMe2)(PMe3) is not stable enough to be isolated.
An X-ray study of [i-PrNON]Zr(η2-CH2CMe2)(PMe3)2 (Tables
2 and 4) confirms that the phosphines are coordinated trans to
one another (P(2)-Zr-P(3) ) 160.55(5)°) and the isobutylene
ligand is coordinated trans to the oxygen donor of the
[i-PrNON]2- ligand (Figure 4). The Zr-P bonds are shorter
than the Zr-P bond in [i-PrNON]Zr(CH2CHMe2)2(PMe3),
although still relatively long, consistent with a crowded pseu-
otherwise the sample decomposes to unidentified products.
Room temperature 31P NMR spectra (in C6D6) of [i-PrNON]-
Zr(η2-CH2CMe2)(PMe3)2 show two broad PMe3 resonances at
-26.0 ppm (coordinated PMe3) and -60.1 ppm (free PMe3),
consistent with ready loss of one PMe3 from [i-PrNON]Zr(η2-
CH2CMe2)(PMe3)2. The isopropyl methyl groups are diaste-

Ti and Zr Complexes Containing Diamido Ligands
J. Am. Chem. Soc., Vol. 121, No. 34, 1999 7827
dooctahedral complex. The [i-PrNON]2- ligand adopts the mer
conformation with bond lengths and angles similar to those
found in [i-PrNON]Zr(CH2CHMe2)2(PMe3). The Zr-C(1) dis-
tance (2.327(5) Å) is slightly longer than Zr(1)-C(2) (2.291-
(6) Å), which is what one would expect on steric grounds, and
what has been observed in Cp2Hf(η2-CH2CMe2)(PMe3).7 The
C(1)-C(2) bond length (1.457(8) Å) is consistent with a
substantial degree of back-bonding into the olefin. It should be
noted that the two olefinic carbon atoms of the coordinated
isobutylene in [i-PrNON]Zr(η2-CH2CMe2)(PMe3)2 lie almost in
the ZrP2 plane, in contrast to the twisting of the ethylene in
[t-BuNON]Zr(η2-CH2CH2)(PMe3)2 ∼15° out of the ZrP2 plane.3
These facts are consistent with a greatly reduced steric influence
by the isopropyl groups in comparison with tert-butyl groups
in complexes of this general type. NMR spectra of [i-PrNON]-
Zr(η2-CH2CMe2)(PMe3)2 are consistent with no ready rotation
of the isobutylene ligand about the Zr-O axis by ∼90°, which
would generate a second mirror plane in the molecule and
equilibrate the isopropyl methyl groups; the more facile process
is loss of one PMe3 ligand. The PMe3 ligand next to C(1) on
steric grounds would appear to be the one that is lost most
readily.
Addition of trimethylphosphine to [i-PrNON]ZrPr2 and [i-
PrNON]ZrEt2 led to quantitative formation of [i-PrNON]Zr-
(η2-CH2CHCH3)(PMe3)2 and [i-PrNON]Zr(η2-CH2CH2)(PMe3)2,
respectively, according to NMR studies. The isopropyl methyl
groups are all inequivalent in the former, as they must be
regardless of whether propylene rotates rapidly about the Zr-
propylene bond axis or not.
Decomposition Reactions of Titanium Dialkyl Complexes.
Decomposition of [i-PrNON]Ti(CH2CHMe2)2 in the presence
of PMe3 (∼4 equiv) under dinitrogen (1 atm) yields the bridging
dinitrogen complex {[i-PrNON]Ti(PMe3)2}2(µ-N2) in 61%
isolated yield (eq 8). We propose that trimethylphosphine
Figure 5. ORTEP diagram (35% probability level) of {[i-PrNON]Ti-
(PMe3)2}2(µ-N2).
a “hydrazido4-” type,8,9 according to the N-N bond distance
of 1.264(8) Å and a Ti-N(3) distance of 1.811(4) Å. The Ti-
(µ-N2) bond distance (Ti(1)-N(3) ) 1.811(4) Å) and the N-N
bond distance (N(3)-N(3A) ) 1.264(8) Å) are similar to
distances in titanium dinitrogen complexes that contain nitrogen-
based ancillary ligands.10-12 Titanium(II) dinitrogen complexes
stabilized by cyclopentadienyl ligands13-15 have longer Ti-N
bonds (1.920-2.033 Å) and shorter N-N (1.155-1.191 Å)
bonds, consistent with less reduction of dinitrogen and less
electron delocalization across the dinitrogen bridge. The
[i-PrNON]2- ligand adopts the same twisted mer configuration
that is found in other complexes reported here.
When [i-PrNON]Ti(CH2CHMe2)2 was allowed to decompose
in benzene in the presence of DMPE (Me2PCH2CH2PMe2) and
in the absence of dinitrogen in the dark at 22 °C, the color of
the reaction mixture changed from bright orange to red-black.
Black crystals of the new diamagnetic complex appeared to have
the formula [i-PrNON]Ti(dmpe), according to elemental analy-
ses. However, the complex proton and carbon NMR spectra
suggested that the molecule had no symmetry. An X-ray study
revealed that “[i-PrNON]Ti(dmpe)” is formally the product of
an oxidative addition of an aryl-oxygen bond of the ligand
backbone (C(26)-O(1)) to the metal center (eq 9, Tables 1 and
induces decomposition of [i-PrNON]Ti(CH2CHMe2)2 to yield
isobutane and an intermediate isobutene complex, [i-PrNON]-
Ti(η2-CH2CMe2)(PMe3). The isobutene complex then loses
isobutene and binds dinitrogen, and another equivalent of PMe3
adds to yield the observed product.
An X-ray structure of {[i-PrNON]Ti(PMe3)2}2(µ-N2) (Tables
1 and 4, Figure 5) confirms that two identical pseudooctahedral
titanium fragments are linearly bridged by one dinitrogen
molecule (Ti(1)-N(3)-N(3A) ) 177.7(4)°) and twisted with
respect to one another by 90°. The geometry about the one
unique titanium center can be described as distorted octahedral
with two PMe3 molecules mutually trans and the titanium atom
lying in the N-O-N plane. The dinitrogen could be said to be
5, Figure 6); O(1) and C(26) end up approximately trans to one
(8) Hidai, M.; Mizobe, Y. Chem. ReV. 1995, 95, 1115.
(9) Turner, H. W.; Fellmann, J. D.; Rocklage, S. M.; Schrock, R. R.;
Churchill, M. R.; Wasserman, H. J. J. Am. Chem. Soc. 1980, 102, 7809.
(10) Hagadorn, J. R.; Arnold, J. J. Am. Chem. Soc. 1996, 118, 893.
(11) Duchateau, R.; Gambarotta, S.; Beydoun, N.; Bensimon, C. J. Am.
Chem. Soc. 1991, 113, 8986.
(7) Buchwald, S. L.; Kreutzer, K. A.; Fisher, R. A. J. Am. Chem. Soc.
1990, 112, 4600.
(12) Beydoun, N.; Duchateau, R.; Gambarotta, S. J. Chem. Soc., Chem.
Commun. 1992, 244.

7828 J. Am. Chem. Soc., Vol. 121, No. 34, 1999
Baumann et al.
Table 5. Bond Lengths (Å) and Angles (deg) for
(i-PrNC6H4)(i-PrNC6H4O)Ti(dmpe)
Distances
Ti-N(2)
Ti-N(1)
Ti-C(26)
Ti-C(21)
1.943(3)
1.996(2)
2.146(4)
2.510(3)
Ti-O(1)
Ti-P(l)
Ti-P(2)
1.961(2)
2.6238(11)
2.6138(10)
Angles
N(2)-Ti-O(1)
O(1)-Ti-N(1)
O(1)-Ti-C(26)
N(2)-Ti-P(2)
N(1)-Ti-P(2)
O(1)-Ti-P(1)
C(26)-Ti-P(1)
P(2)-Ti-P(1)
116.54(11) N(2)-Ti-N(1)
80.71(10) N(2)-Ti-C(26)
162.45(12) N(1)-Ti-C(26)
109.04(11)
68.64(14)
114.31(12)
80.78(7)
88.19(8)
158.98(8) C(26)-Ti-P(2)
85.89(7)
84.51(11) C(11)-O(1)-Ti
O(1)-Ti-P(2)
82.70(9)
N(1)-C(17)-C(19) 111.0(3)
114.5(2)
111.7(2)
150.64(9)
92.37(8)
141.2(2)
76.43(3)
C(16)-N(1)-Ti
N(2)-Ti-P(1)
N(1)-Ti-P(l)
C(27)-N(2)-Ti
C(16)-N(1)-C(17) 116.3(2)
C(17)-N(1)-Ti
C(21)-N(2)-Ti
C(21)-N(2)-C(27) 122.6(3)
131.6(2)
95.9(2)
Figure 6. ORTEP diagram (35% probability level) of (i-PrNC6H4)-
(i-PrNC6H4O)Ti(dmpe).
another (O(1)-Ti-C(26) ) 162.45(12)°). We propose that the
C(26)-O(1) bond is cleaved in intermediate [i-PrNON]Ti-
(dmpe). The Ti-O bond in (i-PrC6H4)(i-PrC6H4O)Ti(dmpe)
(Ti-O(1) ) 1.961(2) Å) is shorter than dative Ti-O bonds
found in other compounds reported here, consistent with a
covalent Ti-O bond. Trans to the oxygen atom is a phenyl
group [O(1)-Ti-C(26) ) 162.45(12)°] that is part of a strained
azatitanabenzocyclobutene ring, a fact that accounts for the acute
Ti-C(26)-C(21) angle [86.5(2)°]. The sum of the angles about
C(26) and N(2) is ∼360°. The Ti-N and Ti-P distances are
in the expected range.
Heating [i-PrNON]Ti(CHCMe3)2 in the presence of excess
PMe3 produces the neopentylidene complex [i-PrNON]Ti-
(CHCMe3)(PMe3)2, which was isolated as green-black crystals
1
in ∼50% yield (eq 10). The H and 13C NMR resonances for
Figure 7. ORTEP diagram (35% probability level) of [i-PrNON]Ti-
(CHCMe3)(PMe3)2). (Only one orientation of the tert-butyl group is
shown; see the Experimental Section.)
data are consistent only with a highly distorted neopentyli-
dene ligand in which the R hydrogen is not localized on the
NMR time scale. In many neopentylidene complexes, including
the first of the high oxidation state type (Ta(CHCMe3)-
(CH2CMe3)318), the alkylidene HR is not localized on the NMR
time scale, even though the agostic interaction is significant.
the neopentylidene HR and CR are slightly concentration
dependent, consistent with ready dissociation of one PMe3, as
found in other pseudooctahedral complexes reported here. The
alkylidene proton appears at ∼3 ppm, which is upfield relative
to its chemical shift in all other titanium alkylidene complexes
(δ ≈ 12 ppm),16 and characteristic of an alkylidene in which
there is a significant agostic interaction17 between the CHR
electron pair and the metal.18 The gated decoupled 13C NMR
spectrum shows a doublet at δ ≈ 229 ppm with a low JCH
coupling constant (80 Hz). The isopropyl methyl groups are
equivalent, which suggests that [i-PrNON]Ti(CHCMe3)(PMe3)2
is C2V symmetric in solution on the NMR time scale. These
An X-ray structure (Tables 1 and 4; Figure 7) revealed that
the neopentylidene ligand in [i-PrNON]Ti(CHCMe3)(PMe3)2 is
indeed linear [Ti(1)-C(1)-C(2) ) 179.3(3)°] with a short
Ti-C(1) bond [1.884(4) Å], characteristic of a “distorted”
alkylidene in which the CHR agostic interaction is significant.18
No electron density could be located in the vicinity of C(1)
that could be ascribed to HR. Therefore, we cannot say whether
HR is localized in the solid state or disordered over two or more
bonding sites with no detectable disorder in the surrounding
ligand set. For comparison, in the only other structurally
characterized titanium neopentylidene complex, {Me2PCH2C-
(O)[o-(CMe2)2C6H4]}CpTi(CHCMe3), a Ti-C bond length of
1.911(3) Å and a Ti-CR-Câ bond angle of 158.7(2)° were
found.19 The ligand configuration and Ti-P bond lengths in
[i-PrNON]Ti(CHCMe3)(PMe3)2 are similar to those of other
species reported here. The tert-butyl group was found in two
orientations, although the disorder could be solved readily.
(13) de Wolf, J. M.; Blaauw, R.; Meetsma, A.; Teuben, J. H.; Gyepes,
R.; Varga, V.; Mach, K.; Veldman, N.; Spek, A. L. Organometallics 1996,
15, 4977.
(14) Berry, D. H.; Procopio, L. J.; Carroll, P. J. Organometallics 1988,
7, 570.
(15) Sanner, R. D.; Duggan, D. M.; McKenzie, T. C.; Marsh, R. E.;
Bercaw, J. E. J. Am. Chem. Soc. 1976, 98, 8358.
(16) Beckhaus, R. Angew. Chem., Int. Ed. Engl. 1997, 36, 687.
(17) Brookhart, M.; Green, M. L. H.; Wong, L. Prog. Inorg. Chem. 1988,
36, 1.
(18) Schrock, R. R. In Alkylidene Complexes of the Earlier Transition
Metals; Braterman, P. R., Ed.; Plenum: New York, 1986.
(19) van Doorn, J. A.; van der Heijden, H.; Orpen, A. G. Organometallics
1995, 14, 1278.

Ti and Zr Complexes Containing Diamido Ligands
J. Am. Chem. Soc., Vol. 121, No. 34, 1999 7829
Table 6. A Summary of the Rate Constants for Decomposition of
[i-PrNON]ZrR2 Complexesa
A Study of the Decomposition of [i-PrNON]Zr(CH2CH3)2.
[i-PrNON]Zr(CH2CH3)2 is the only â-hydrogen-containing
dialkyl complex discussed in this paper that decomposes in the
absence of PMe3 cleanly to a single identifiable diamagnetic
species. In pentane the product crystallizes out in ∼75% yield
as small, pale yellow needles that so far have not been suitable
for a single-crystal X-ray study. NMR data and combustion
analyses are consistent with the product having the composi-
tion {[i-PrNON]Zr(CH2CH3)}2(µ-CH2CH2). Since a proton
resonance in the same position as that of ethane is observed
during decomposition, we propose that the reaction is that
shown in eq 11. The isopropyl methyl groups in {[i-PrNON]-
rate constant
(min-1
R
factor
compound
Zr(CH2CH3)2
temp (°C)
)
22
0.0078
0.0083
0.020
0.058
0.025
0.080
0.0025
0.012
0.0036
0.052
0.020
0.9998
0.9993
0.9936
0.9913
0.9974
0.9865
0.9965b
0.9978b
0.9865b
0.9933c
0.9942d
35
45
35
45
35
45
45
22
22
Zr(CD2CH3)2
Zr(CD2CD3)2
Zr(CH2CH2CH3)2
Zr(CH2CH3)2/∼2.5PMe3
pentane
Zr(CH2CH2CH3)2/∼2.5PMe3
2[i-PrNON]ZrEt2 22 °C, 21 h, -ethane8
a Determined by proton NMR following the disappearance of the
dialkyl. b Followed for approximately 1 half-life. c [Zr]o ) 31 mM; [P]o
) 85 mM. d [Zr]o ) 33 mM; [P]o ) 85 mM.
{[i-PrNON]ZrEt}2(µ-C2H4) (11)
Zr(CH2CH3)}2(µ-CH2CH2) are diastereotopic, which would be
consistent with the expected mer configuration of the ligand,
but no mirror plane exists that relates the two methyl groups in
one isopropyl group. A singlet resonance for the bridging
rate constants were 0.0078 min-1 (R ) 0.9998) and 0.0083
min-1 (R ) 0.9993; Table 6). The observed rate constant for
decomposition of [i-PrNON]Zr(CH2CH3)2 was found to be 0.020
min-1 (R ) 0.9936; >2 half-lives) at 35 °C. This value should
be compared with kobs for decomposition of [i-PrNON]Zr-
(CD2CH3)2 (0.025 min-1; R ) 0.9974; >2 half-lives) and
[i-PrNON]Zr(CD2CD3)2 (0.0025 min-1; R ) 0.9965; ∼1 half-
life) at 35 °C. These data suggest that the mechanism of
decomposition of [i-PrNON]Zr(CH2CH3)2 consists of a unimo-
lecular â-hydride abstraction to give ethane and “[i-PrNON]-
Zr(CH2CH2)”, which then reacts rapidly with [i-PrNON]Zr-
(CH2CH3)2 to give the observed product (eqs 12 and 13). The
1
ethylene is found at 1.09 ppm in the H NMR spectrum. The
µ-ethylene complex decomposes upon heating in solution.
The dimeric formulation of {[i-PrNON]Zr(CH2CH3)}2(µ-
CH2CH2) is based upon the characterization of complexes
that contain bridging ethylene ligands in the literature. The
reported complexes include [MX3(PEt3)2]2(µ-η2:η2-C2H4) (M )
Zr, Hf; X ) Cl, Br),20-22 which have type A structures,21,22
and (Cp2ZrR)2(µ-η2:η2-C2H4) (R ) ClAlEt3,23 Me24) and
{[(SiMe2)2(η5-C5H3)2]ZrEt}2(µ-η2:η2-C2H4),25 which have type
B structures, perhaps in part as a consequence of greater steric
k1
[i-PrNON]ZrEt2
98 [i-PrNON]Zr(CH2CH2) + ethane (12)
k2
[i-PrNON]Zr(CH2CH2) + [i-PrNON]ZrEt2
98
{[i-PrNON]ZrEt}2(µ-C2H4) (13)
difference between the rate constant for decomposition of
[i-PrNON]Zr(CD2CH3)2 (0.025 min-1) and that for [i-PrNON]-
Zr(CH2CH3)2 (0.0200 min-1) at 35 °C is evidence for a typically
small inverse secondary isotope effect. However, the difference
in the observed rate constant for decomposition of [i-PrNON]-
Zr(CD2CH3)2 (0.025 min-1) and [i-PrNON]Zr(CD2CD3)2 (0.0025
min-1) is clearly indicative of a primary isotope effect of
approximately 10 for the â-hydride abstraction step. At 45 °C
the decompositions of [i-PrNON]Zr(CH2CH3)2, [i-PrNON]Zr-
(CD2CH3)2, and [i-PrNON]Zr(CD2CD3)2 are not as well-behaved
as they are at 35 °C, with some curvature in the first-order plot
being observed beyond 2 half-lives for [i-PrNON]Zr(CH2CH3)2
and [i-PrNON]Zr(CD2CH3)2, and beyond 1 half-life for [i-PrNON]-
Zr(CD2CD3)2. However, the data are useful, if not as accurate,
and suggest again an inverse isotope effect of ∼0.8 for
decomposition of [i-PrNON]Zr(CH2CH3)2 versus [i-PrNON]Zr-
(CD2CH3)2, and a primary isotope effect of ∼7 for decomposi-
tion of [i-PrNON]Zr(CH2CH3)2 versus [i-PrNON]Zr(CD2CD3)2
at 45 °C (Table 6). The rate constant for the rate-limiting
formation of “[i-PrNON]Zr(CH2CH2)” (k1; eq 12) would be half
the observed rate constant, if the subsequent bimolecular reaction
(eq 13) were rapid.
crowding in zirconocenes. The JCH coupling constants for the
ethylene bridge in (Cp2ZrMe)2(µ-η2:η2-C2H4) and {[(SiMe2)2-
(η5-C5H3)2]ZrEt}2(µ-η2:η2-C2H4) are 146 and 141 Hz, respec-
tively. The JCH for the “olefinic” 13C resonances in a propylene
complex of type A were found to be in the range 127-129
Hz.20 In {[i-PrNON]Zr(CH2CH3)}2(µ-CH2CH2) JCH ) 143 Hz.
Therefore, on the basis of JCH values, we propose that
{[i-PrNON]Zr(CH2CH3)}2(µ-CH2CH2) is of type B. It should
be noted that {[(SiMe2)2(η5-C5H3)2]ZrEt}2(µ-η2:η2-C2H4) is
formed via intermediate [(SiMe2)2(η5-C5H3)2]ZrEt2, while (Cp2-
ZrMe)2(µ-η2:η2-C2H4) was obtained upon treatment of [Cp2-
ZrEt(CH2dCH2)]MgBr with [Cp2ZrMe2], a reaction that would
require transfer of a methyl group from one Zr to the other.
Decomposition of [i-PrNON]Zr(CH2CH3)2 to {[i-PrNON]Zr-
(CH2CH3)}2(µ-CH2CH2) was followed by proton NMR. The
decomposition was found to be first order in Zr over a period
of more than 2 half-lives for two samples in C6D6 at 22 °C
with initial concentrations of 0.07 and 0.015 M. The observed
(20) Wengrovius, J. H.; Schrock, R. R.; Day, C. S. Inorg. Chem. 1981,
20, 1844.
(21) Cotton, F. A.; Kibala, P. A. Polyhedron 1987, 6, 645.
(22) Cotton, F. A.; Kibala, P. A. Inorg. Chem. 1990, 29, 3192.
(23) Kaminsky, W.; Kopf, J.; Sinn, H.; Vollmer, H.-J. Angew. Chem.,
Int. Ed. Engl. 1976, 15, 629.
(24) Takahashi, R.; Kasai, K.; Suzuki, N.; Nakajima, K.; Negishi, E.
Organometallics 1994, 13, 3413.
(25) Fernandez, F. J.; Gomez-Sal, P.; Manzanero, A.; Royo, P.; Jacobsen,
H.; Berke, H. Organometallics 1997, 16, 1553.
Two plausible ways to form the final product in an intermo-
lecular fashion are â-hydrogen transfer or alkyl transfer (Scheme
1). The product of decomposition of [i-PrNON]Zr(CD2CH3)2
should be {[i-PrNON]Zr(CD2CH3)}2(CD2CH2) if the alkyl
transfer mechanism prevails, whereas intermolecular â-hydro-
gen abstraction should lead to a 1:1 mixture of {[i-PrNON]Zr-

7830 J. Am. Chem. Soc., Vol. 121, No. 34, 1999
Baumann et al.
Scheme 1
cantly slower rate of decomposition of [i-PrNON]ZrPr2 is at
least consistent with what has been observed in zirconocene
systems, namely, the greater stability of zirconocene dipropyl
intermediates compared to zirconocene diethyl intermediates.5
The rate of decomposition of [i-PrNON]ZrEt2 at 22 °C in
C6D6 in the presence of ∼2.5 equiv of PMe3 is also first order
in Zr through >2 half-lives (kobs ) 0.052 min-1), which is much
faster than decomposition of [i-PrNON]ZrEt2 in the absence of
PMe3 (kobs ) 0.0078 min-1, Table 6). Under the same conditions
the observed rate constant for decomposition of [i-PrNON]ZrPr2
is 0.020 min-1. We conclude that the decomposition of the
dialkyl complexes is induced by phosphine. On the basis of the
structure of isolated [i-PrNON]Zr(CH2CHMe2)2(PMe3), we
propose that [i-PrNON]ZrR2 and PMe3 are in equilibrium with
[i-PrNON]ZrR2(PMe3), and that the equilibrium constant is large
enough that a first-order decay in zirconium is observed even
as phosphine is consumed to form the product. The equilibrium
constant for formation of an adduct in which the alkyls are
smaller than isobutyl (Et and Pr) should be even larger than it
is in isolated [i-PrNON]Zr(CH2CHMe2)2(PMe3). It should be
noted that one phosphine is labile in the observed olefin
products, so the concentration of phosphine actually may change
by a factor of only 2-3 under the stated conditions. A full
treatment of the decomposition to establish the true rate constant
for decomposition of [i-PrNON]ZrR2(PMe3), as well as equi-
librium constants for formation of the monophosphine adducts,
is beyond the scope of this paper.
(CD2CH3)}2(CD2CH2) and [i-PrNON]2Zr2(CH2CD2H)(CD2CH3)-
(CD2CH2), assuming no isotope effect of any kind and an equal
probability of adding a proton to each end of the ethylene in
intermediate [i-PrNON]Zr(CD2CH2). In the latter case the ratio
of D attached to CR of an ethyl group, Câ of an ethyl group,
and an ethylene carbon should be 3:1:2; i.e., ∼17% of the
deuterium should be in the â position of an ethyl group. This
2
labeling study was carried out, and by H NMR no deuterium
could be detected in the â position of an ethyl group. Therefore,
these data suggest either that an alkyl group transfers from one
Zr to the other in the second (fast) step of the reaction or that
the Hâ in an ethyl group on one zirconium transfers virtually
exclusively to the CH2 end of the η2-CH2CD2 ligand in the
intermediate.
Synthesis of Titanium and Zirconium Complexes That
Contain [CyNON]2-. Condensation between cyclohexanone
and O(o-C6H4NH2)2 in acetic acid in the presence of excess
zinc at 65 °C cleanly afforded H2[CyNON] in ∼90% yield (eq
15). H2[CyNON] is an air-stable, yellow viscous oil that is
soluble in common solvents; it occasionally crystallizes at room
temperature after several weeks.
A
13C labeling study also was undertaken to eliminate any
possibility of a dramatic isotope effect altering the reaction
pathway. [i-PrNON]Zr(13CH2CH3)2 was prepared by treating
[i-PrNON]ZrCl2 with CH313CH2MgI, and its decomposition in
C6D6 was followed by 13C NMR at 22 °C. The reaction was
allowed to proceed 2/3 of the way to completion, i.e., until the
ratio of [i-PrNON]ZrEt2 to {[i-PrNON]ZrEt}2(µ-C2H4} was 1:1.
In the 13C NMR spectrum of this mixture the weak intensity of
the resonance for ZrCH213CH3 in {[i-PrNON]ZrEt}2(µ-C2H4}
was entirely consistent with naturally abundant 13C; i.e., there
is no evidence for the 13C-labeled carbon being in the â position
of an ethyl group in the product. These results are again
consistent either with transfer of an ethyl group from ZrEt2 to
the Zr(η2-C2H4) intermediate or with proton transfer selectively
to a 12C carbon atom in the Zr(η2-C2H4) intermediate that has
no time or opportunity to equilibrate with the 13C-labeled end
of the ethylene fragment. We favor the ethyl transfer proposal
(eq 14), since there is precedent in the literature for a methyl
transfer in zirconocene systems to give an analogous µ-
CH2CH2 complex.24
The syntheses of [CyNON]ZrCl2 (eq 16) and [CyNON]TiCl2
(eq 17) are entirely analogous to those for [i-PrNON]MCl2. The
Decomposition of [i-PrNON]ZrPr2 does not yield any char-
acterizable product, e.g., an analogous bridging propylene
complex, although such species have been observed in other
systems.20 At 45 °C [i-PrNON]ZrPr2 decomposes in a first-order
manner in C6D6 at a rate that is approximately 16 times slower
than the rate at which [i-PrNON]ZrEt2 decomposes (Table 6).
Since the mechanism of decomposition of [i-PrNON]ZrPr2 may
not be analogous to that of [i-PrNON]ZrEt2, the relative rates
of â abstraction in the two species are uncertain. The signifi-
dichloride complexes react with lithium reagents in toluene or
magnesium reagents in diethyl ether to give the dialkyl com-
plexes [CyNON]MR2 (M ) Zr, R ) Me, Et, i-Bu, CH2CMe3,
allyl; M ) Ti, R ) Me, CH2CMe3, i-Bu) in moderate to
excellent yields. Proton and carbon NMR spectra of the dialkyl
complexes are all consistent with a structure that has C2 or C2V
symmetry. We presume that they are all mer structures by
analogy with crystallographically characterized [i-PrNON]2-
structures discussed above. The reaction between H2[CyNON]

Ti and Zr Complexes Containing Diamido Ligands
J. Am. Chem. Soc., Vol. 121, No. 34, 1999 7831
and Zr(CH2SiMe3)4 in toluene or benzene at room temperature
gave [CyNON]Zr(CH2SiMe3)2 in high yield. In contrast, at-
tempts to react H2[CyNON] with Zr(CH2Ph)4 or Zr(CH2CMe3)4
(70 °C for 3.5 days and 70 °C for 7 days, respectively) did not
yield the expected dialkyl complexes; no products could be
identified.
circumstances, and by what mechanism, and whether oligo-
merization of some monomers (e.g., ethylene) might produce
oligomers with a molecular weight in an unusual range. The
main point is that the difference in behavior between cationic
[i-PrNON]2- initiators and [t-BuNON]2- initiators1-3 for the
polymerization of 1-hexene is dramatic.
The dimethyl, dineopentyl, and bis(trimethylsilylmethyl)
complexes are stable in solution at room temperature for days,
according to their proton NMR spectra. For example, a toluene-
d8 solution of [CyNON]Zr(CH2SiMe3)2 (∼0.16 M) showed
<5% decomposition after being heated to 100 °C for 3 days.
In contrast, orange [CyNON]Ti(i-Bu)2 (∼8 mM in C6D6)
decomposed at room temperature over a period of several hours.
Solutions of [CyNON]Zr(i-Bu)2 (∼7 mM in C6D6) show no
signs of decomposition at room temperature after 24 h, but
solutions decompose readily at 100 °C with an accompanying
rapid color change from colorless to deep red. [CyNON]ZrEt2
decomposes over a period of 1 h at room temperature.
Decomposition of [CyNON]ZrEt2 in neat trimethylphosphine
yielded orange [CyNON]Zr(η2-C2H4)(PMe3)2 in ∼70% isolated
yield (eq 18), while a similar reaction between [CyNON]Zr-
(i-Bu)2 and neat PMe3 requires a temperature of 50 °C over a
period of 3 days in the dark to yield [CyNON]Zr(H2CdCMe2)-
(PMe3)2.
Discussion
The results reported here suggest that complexes that contain
the [i-PrNON]2- ligand are significantly less crowded than those
that contain the [t-BuNON]2- ligand, and that dialkyl complexes
that contain the [t-BuNON]2- ligand3 are less stable than
[i-PrNON]2- complexes, especially for titanium. For example,
[i-PrNON]Ti(CH2CHMe2)2 and [i-PrNON]Ti(CH2CMe3)2 can
both be prepared, while only [t-BuNON]TiMe2 could be
prepared. While the tert-butyl group is a slightly better electron
donor than an isopropyl group, that difference alone would not
appear to be significant enough to produce a dramatically lower
stability for [t-BuNON]TiR2 complexes. It should be noted,
however, that the greater instability of [t-BuNON]TiR2 com-
plexes cannot be ascribed automatically to decomposition
processes that involve â-hydrogen abstraction or homolytic
Ti-C bond cleavage; alternatives such as N-t-Bu bond
cleavage cannot be eliminated.
It also seems clear now that the mer structure (twisted) is
the lowest energy structure for [i-PrNON]MR2 structures, at least
those whose structures we have determined, and that the
(twisted) fac structures found for [t-BuNON]MR2 complexes
can be ascribed to steric crowding between the tert-butyl group
and the MR2 groups. Although most metal-ligand bond
distances are not dramatically different in the two species, there
is some tendency for M-Odonor bonds to be shorter in mer
complexes, perhaps as a consequence of some M-O π bonding,
which is possible only in mer species. In mer-[i-PrNON]ZrMe2
the isopropyl hydrogen atoms point directly between the methyl
groups, while the isopropyl methyl groups are located on each
side of the phenylene ring. The ligand backbone is still twisted
to some extent as a consequence of an interaction between the
two aryl protons next to the oxygen donor. If [t-BuNON]ZrMe2
were to adopt the same mer structure as [i-PrNON]ZrMe2, the
additional methyl group in each amido substituent would point
between the two zirconium methyl groups. To minimize that
steric interaction, the ligand therefore twists further into the fac
form.
The issues concerning a mer versus a fac structure have been
in the literature for some time for similar dianionic/donor ligands
bound to a main group element such as tin.27 For example,
Holmes28 has reported the structures of the cyclic stannanes
(t-Bu)2Sn[(OCH2CH2)2NMe], a mer structure, and Me2Sn-
[(SCH2CH2)2NMe], a fac structure with a relatively long Sn-S
bond. NMR data29-31 support interconversion of mer and fac
structures, and actual dissociation of the amine donor at higher
temperatures, especially in the sulfur case. In rare cases both
mer and fac isomers have been observed in solution. However,
steric factors would not play the same role in these systems as
in the systems reported here.
excess PMe3
2
(CyNON)ZrR2
R ) Et, i-Bu
8
(CyNON)Zr(η -C2H2R′2)(PMe3)2
R′ ) H, Me
(18)
All of the [CyNON]2- chemistry that we have explored
appears to be entirely analogous to the [i-PrNON]2- chemistry.
Evidently the increased steric bulk of the cyclohexyl group
compared to the isopropyl group is of no consequence, as one
might have surmised on the basis of the X-ray structures of the
[i-PrNON]2- complexes reported here. Since the [CyNON]2-
chemistry is so similar to the [i-PrNON]2- chemistry, the
experimental details of the synthesis of metal complexes are
provided as Supporting Information.
Generation of Cations and Their Reactions with 1-Hexene.
The reaction between [CyNON]ZrMe2 and [HNMe2Ph][B(C6F5)4]
in bromobenzene-d5 at -35 °C led to a single species whose
1H NMR spectrum at 0 °C exhibited a singlet at 0.65 ppm of
area 3, consistent with the formation of a cationic complex
{[CyNON]ZrMe(NMe2Ph)}+. {[CyNON]ZrMe(NMe2Ph)}+ ap-
pears to be stable in solution at temperatures below 10 °C, but
decomposes slowly at room temperature, according to 1H NMR
spectra.
Addition of 50 equiv of 1-hexene to a bromobenzene-d5
solution of {[CyNON]ZrMe(NMe2Ph)}[B(C6F5)4], generated as
above at ∼0 °C, led to slow oligomerization of 1-hexene. New
olefinic resonances were observed in the range 5.42-5.34 ppm,
a region that is consistent with them being internal olefinic
protons, e.g., the product of â elimination from a 2,1 insertion
product.26 The intensity suggests that approximately 10 equiv
of 1-hexene was consumed for every two olefinic protons in
an end group. Analogous reactions involving [i-PrNON]ZrMe2
activated by [Ph3C][B(C6F5)4] in bromobenzene-d5 at 0 °C
produced similar results.
(27) Holmes, R. R. Prog. Inorg. Chem. 1984, 32, 119.
(28) Swisher, R. G.; Holmes, R. R. Organometallics 1984, 3, 365.
(29) Zschunke, A.; Tzschach, A.; Jurkschat, K. J. Organomet. Chem.
1976, 112, 273.
(30) Jurkschat, K.; Mu¨gge, C.; Tzschach, A.; Zschunke, A.; Larin, M.
F.; Pestunovich, V. A.; Voronkov, M. G. J. Organomet. Chem. 1977, 139,
279.
These data suggest that activated [CyNON]2- or [i-PrNON]2-
complexes of the type described here are not successful
polymerization catalysts. Further studies will be necessary to
assess exactly what oligomeric products are formed in such
(26) Resconi, L.; Piemontesi, F.; Camurati, I.; Sudmeijer, O.; Nifant’ev,
I. E.; Ivchenko, P. V.; Kuz’mina, L. G. J. Am. Chem. Soc. 1998, 120, 2308.
(31) Mu¨gge, C.; Jurkschat, K.; Tzschach, A.; Zschunke, A. J. Organomet.
Chem. 1979, 164, 135.

7832 J. Am. Chem. Soc., Vol. 121, No. 34, 1999
Baumann et al.
The [i-PrNON]MR2 and [CyNON]MR2 complexes are much
more stable than analogous Cp2MR2 complexes, especially when
M ) Ti. For example, Cp2Ti(CH2CMe3)2 can be isolated only
below 0 °C.32 Also, it is known that Cp2TiEt2, prepared in situ
from Cp2TiCl2 and EtMgCl, cleanly decomposes in the presence
of PMe3 to yield Cp2Ti(PMe3)2.33 Cp2ZrEt2 is also not isolable,
although Cp2HfEt2 is isolable, consistent with a greater stability
toward â-hydride elimination in the order R ) Et < n-Pr <
i-Bu,5,6 at least for zirconocene complexes of the type Cp2ZrR2.
(For a more extensive discussion of group 4 metal dialkyl
complexes, especially of nitrogen-based ligands, see a related
paper concerning [t-BuNON]2- complexes.3)
Dialkyl metallocenes appear to be inherently more crowded
than the dialkyl complexes discussed here, which could help
explain the generally wider C-M-C angles in the complexes
reported here compared to, for example, the C-Zr-C angle in
Cp2ZrMe2 (95.6(12)°34). Therefore, when a donor adds to a
Cp2ZrR2 complex, it generally does so between the two alkyl
groups. In contrast, in [i-PrNON]Zr(CH2CHMe2)2 a phosphine
can more readily approach the metal from outside the C-Zr-C
wedge, a process that consequently pushes the alkyl groups
toward one another. Therefore, phosphines do not accelerate
â-hydrogen abstraction in Cp2MR2 complexes,5,6 in contrast to
the accelerated â-hydrogen abstraction in [i-PrNON]MR2
complexes found here. In Cp2ZrH2 a central coordination of
PMe3 to zirconium has been observed.35 In short the orbital
between the two alkyl groups in a d0 metallocene is believed to
be the most Lewis acidic one,36 as well as the one that is
sterically most accessible. There are exceptions, e.g., the reaction
of CO with zirconocene dialkyl and diaryl complexes. Low-
temperature NMR and IR studies by Erker showed that initially
CO inserts into one of the two zirconium carbon bonds to give
a lateral η2-acetyl complex that rearranges at higher temperature
to the thermodynamically more stable central product.37,38
significant factor in generating and stabilizing a linear neopen-
tylidene ligand with a significant agostic interaction. Similar
arguments were put forth to explain the formation of essen-
tially linear alkylidenes with significant agostic interactions in
[(Me3SiNCH2CH2)3N]Ta(CHR) complexes,40 in which the
coordination pocket is 3-fold symmetric.
Reports of cleavage of an aryl-oxygen bond are rare in the
organometallic literature. Milstein and co-workers41 have docu-
mented C-O cleavage of alkyl- and aryl-oxygen bonds by
Rh(I), Pd(II), and Ni(II), while Dehnicke and co-workers have
observed activation of a C-O bond in THF by what is believed
to be a Ti(II) intermediate,42 similar to what we propose in this
paper. We were surprised by cleavage of the aryl-oxygen bond
in the presence of isopropyl amido substituents, which at one
time we presumed would be susceptible to CH bond activation
at carbon either R or â to the amido nitrogen. Zirconium com-
plexes that contain the [(i-PrNC6H4)2S]2- or [(t-BuNC6H4)2S]2-
ligands also have been prepared recently, but only [(i-PrNC6H4)2S]-
ZrR2 complexes were stable, [(i-PrNC6H4)2S]Zr(CH2CHMe2)2
even at elevated temperatures (80 °C). The reason for the ther-
mal instability of [(t-BuNC6H4)2S]ZrMe2 could not be deter-
mined; C-S bond cleavage must now be considered, along with
loss of a tert-butyl group.43
Zirconium olefin complexes of the type Cp2Zr(η2-CH2CHR)-
(PMe3) (R ) H, Me, Et) have been known for some time.33,44,45
Nevertheless, [i-PrNON]Zr(η2-CH2CHMe2)(PMe3)2 appears to
be the first example of a stable zirconium complex bearing a
1,1-disubstituted olefin. The first group 4 isobutylene complex
Cp2Hf(η2-CH2CMe2)(PMe3) was reported by Buchwald and co-
workers.7 In the zirconium system, activation of the Cp ligand
was observed.
A limited number of titanium dinitrogen complexes have been
reported that contain nitrogen-based ligands, i.e., {[PhC-
(NTMS)2]2Ti}2(µ-N2)10 (Ti-(µ-N2) ) 1.771(5), 1.759(5) Å;
N-N ) 1.275(6) Å), [(TMS2N)TiCl(tmeda)]2(µ-N2)11 (Ti-
(µ-N2) ) 1.762(5) Å; N-N ) 1.289(9) Å), and [(TMS2N)-
TiCl(pyridine)2]2(µ-N2)12 (Ti-(µ-N2) ) 1.759(3) Å; N-N )
1.263(7) Å). The N-N bond lengths suggest that the dinitrogen
is more reduced in these complexes than in metallocene systems,
consistent with a greater reducing ability of a Ti(II) metal center
supported by nitrogen-based ancillary ligands. As noted above,
complexes that contain the [i-PrNON]M(PMe3)2 core appear
to be poised to form two strong orthogonal π bonds, much as
complexes that contain a triamidoamine ligand system,46 and
are strongly reducing. Therefore, formation of a µ-dinitrogen
complex is perhaps not surprising. However, we were somewhat
surprised that dinitrogen binding to “[i-PrNON]Ti(PMe3)2”
competes with aryl C-O bond cleavage.
The formation of stable alkylidene complexes of group 4
metals by R-hydrogen abstraction is rare, although more
examples are known for titanium16 than for zirconium.39 It is
interesting to note that if only one π bond between Ti and the
amido nitrogens is invoked (involving the unsymmetric com-
bination of p orbitals on amido nitrogens), then six orbitals are
involved in binding the [i-PrNON]2- ligand and two phosphines
to the metal, leaving two π orbitals (e.g., dxz and dyz) and a σ
2
orbital (e.g., dz ) for binding to the neopentylidene ligand. In
the presence of dxz and dyz orbitals that are close in energy and
an approximately 4-fold symmetric sterically demanding coor-
dination pocket, the neopentylidene ligand adopts a near linear
configuration with apparently little energy difference between
an agostic C-HR interaction along the N-Ti-N axis or along
the P-Ti-P axis. It seems surprising that the energy difference
between the two d orbitals of π type symmetry would be so
small. Therefore, steric interactions between the tert-butyl group
and the isopropyl or phosphine methyl groups may be a
One of the main goals of this research was to attempt to
determine the extent to which the tert-butyl groups create a
coordination sphere in “{[t-BuNON]ZrR}+” that leads to a living
polymerization of 1-hexene. Although this aspect of the study
is not finished, it is clear that “{[i-PrNON]ZrR}+” will only
oligomerize 1-hexene, we presume at this stage because of more
(32) van der Heijden, H.; Hessen, B. J. Chem. Soc., Chem. Commun.
1995, 145.
(33) Binger, P.; Mu¨ller, P.; Benn, R.; Rufinska, A.; Gabor, B.; Kru¨ger,
C.; Betz, P. Chem. Ber. 1989, 122, 1035.
(34) Hunter, W. E.; Hrncir, D. C.; Vann Bynum, R.; Penttila, R. A.;
Atwood, J. L. Organometallics 1983, 2, 750.
(35) Lee, H.; Desrosiers, P. J.; Guzei, I.; Rheingold, A. L.; Parkin, G. J.
Am. Chem. Soc. 1998, 120, 3255.
(40) Freundlich, J. S.; Schrock, R. R.; Davis, W. M. J. Am. Chem. Soc.
1996, 118, 3643.
(41) van der Boom, M. E.; Liou, S.-Y.; Ben-David, Y.; Shimon, L. J.
W.; Milstein, D. J. Am. Chem. Soc. 1998, 120, 6531.
(42) Mommertz, A.; Leo, R.; Massa, W.; Harms, K.; Dehnicke, K. Z.
Anorg. Allg. Chem. 1998, 624, 1647.
(36) Crabtree, R. H. The Organometallic Chemistry of the Transition
Metals, 2nd ed.; John Wiley & Sons: New York, 1994; p 121.
(37) Erker, G.; Rosenfeldt, F. Angew. Chem., Int. Ed. Engl. 1978, 17,
605.
(38) Erker, G.; Rosenfeldt, F. J. Organomet. Chem. 1980, 188, C1.
(39) Fryzuk, M. D.; Mao, S. S. H.; Zaworotko, M. J.; MacGillivray, L.
R. J. Am. Chem. Soc. 1993, 115, 5336.
(43) Schrock, R. R.; Seidel, S. W.; Schrodi, Y.; Davis, W. M. Organo-
metallics 1999, 118, 428-437.
(44) Alt, H. G.; Denner, C. E.; Thewalt, U.; Rausch, M. D. J. Organomet.
Chem. 1988, 356, C83.
(45) Buchwald, S. L.; Watson, B. T.; Huffman, J. C. J. Am. Chem. Soc.
1987, 109, 2544.
(46) Schrock, R. R. Acc. Chem. Res. 1997, 30, 9.

Ti and Zr Complexes Containing Diamido Ligands
J. Am. Chem. Soc., Vol. 121, No. 34, 1999 7833
4.01). The resonance for the NCHMe2 proton of the starting material
was monitored in all cases, since it does not overlap with resonances
of other products or byproducts.
facile â elimination. We hypothesize that the more crowded
coordination sphere in “{[t-BuNON]ZrR}+” encourages 1,2-
insertion and slows â-hydride elimination. In contrast, the much
more open coordination sphere of “{[i-PrNON]ZrR}+” allows
for a significant degree of 2,1 insertion to give a less reactive
insertion product that consequently builds up during an oligo-
merization or polymerization reaction, and that is also likely to
be much less stable toward â elimination. Therefore, only
oligomers are formed. This proposal is analogous to that put
forth recently to account for the oligomerization or polymeri-
zation of propylene to give relatively low molecular weight
polymers under a variety of circumstances.47-51
In future studies we hope to outline further the importance
of steric factors in directing insertion reactions in cationic
diamido/donor Zr or Hf metal complexes, and in correlating
the efficacy of polymerization with steric hindrance, the nature
of the donor, and the nature of the connecting link between the
donor and the amido ligand. Diamido/donor complexes of the
type that have yielded observable, if not always well-behaved,
cationic complexes of zirconium or hafnium utterly fail to yield
stable Ti cations. The reasons are not known at this time.
(i-PrNHC6H4)2O (H2[i-PrNON]). A 250 mL one-neck flask was
charged with (2-NH2C6H4)2O (10.0 g, 50 mmol), acetone (15 mL),
activated Zn dust (25.0 g, 382 mmol), and glacial acetic acid (100 mL).
The flask was capped with a rubber septum, connected to an oil-bubbler
via a needle, and then heated under rapid stirring to 60 °C for 24 h.
The reaction mixture was poured onto a mixture of ice (200 mL),
concentrated aqueous NH3 (200 mL), and methylene chloride (150 mL).
The layers were separated, and the aqueous layer was extracted with
methylene chloride (2 × 100 mL). The combined methylene chloride
layers were dried over MgSO4. Removal of the methylene chloride in
vacuo afforded an orange oil. The oil was dissolved in acetone (150
mL), and concentrated HCl (10 mL) was added. Within 1 min colorless
crystals began to form. The mixture was allowed to stand overnight,
and the colorless crystalline solid was then filtered off, washed with
acetone, and dried. A mixture of aqueous NaOH (100 mL, 10%) and
ether (100 mL) was added to this solid. The mixture was stirred until
the solid had dissolved. The layers were separated, and the aqueous
layer was extracted with ether (3 × 50 mL). The combined organic
layers were dried over MgSO4. Activated charcoal was added prior to
the layers being filtered through a bed of Celite. Ether was removed in
vacuo, leaving a pale yellow oil, yield 13.2 g (93%). The product slowly
darkens in air and is therefore best stored under dinitrogen: 1H NMR
δ 6.98 (t, 2), 6.88 (d, 2), 6.63 (d, 2), 6.55 (t, 2), 4.14 (br s, 2, NH),
3.37 (br m, 2, CHMe2), 0.89 (d, 12, CHMe2); 13C NMR δ 144.8, 140.0,
125.1, 118.8, 117.1, 112.5, 44.4 (CHMe2), 23.2 (CHMe2); HRMS (EI)
calcd for C18H24N2O 284.18886, found 284.18875. Anal. Calcd for
C18H24N2O: C, 76.02; H, 8.51; N, 9.85. Found: C, 76.13; H, 8.39; N,
9.81.
[i-PrNON]TiCl2. A solution of LiBu in hexane (34 mL, 2.6 M)
was added to a solution of H2[i-PrNON] (12.48 g, 43.9 mmol) in ether
(250 mL) at -25 °C. The solution was allowed to warm to room
temperature and was stirred for 10 h. The solution was then cooled to
-25 °C, and TiCl2(NMe2)2 (9.10 g, 44.0 mmol) was added. The mixture
was allowed to warm to room temperature and stirred for 9 h. All
volatile components were removed in vacuo, and the residue was
extracted with pentane (250 mL) for 15 min. An off-white solid was
filtered off and the pentane removed in vacuo. Proton and carbon NMR
spectra of the residue, a dark orange oil, were consistent with it being
(i-PrNON)Ti(NMe2)2: 1H NMR δ 7.30 (d, 2), 7.00 (t, 2), 6.80 (d, 2),
6.48 (t, 2), 4.29 (sept, 2, CHMe2), 3.07 (s, 12, NMe2), 1.40 (d, 12,
CHMe2); 13C NMR δ 148.5, 146.3, 125.5, 115.2, 114.8, 114.0, 54.7
(CHMe2), 46.8 (NMe2), 23.7 (CHMe2).
Experimental Section
General Procedures. Unless otherwise noted all manipulations were
performed under rigorous exclusion of oxygen and moisture in a
dinitrogen-filled glovebox or using standard Schlenk procedures. Ether,
THF, and pentane were sparged with dinitrogen followed by passage
through two 1 gal columns of activated alumina. Toluene and benzene
were distilled from benzophenone ketyl. 31P spectra are referenced
versus an external standard of 85% H3PO4 (δ ) 0). All NMR spectra
are recorded at room temperature in C6D6 unless otherwise noted.
Temperatures during variable-temperature NMR studies were not
1
calibrated. Assignments of aromatic ligand resonances in H and 13C
spectra are not given. Coupling constants are usually not given.
TiCl2(NMe2)2,52 (2-NO2C6H4)2O,53 (2-NH2C6H4)2O,54 and Me3CCH2-
MgCl55 were prepared according to literature procedures. 2,4-Dimethyl-
6-nitrophenol56 was prepared according to a procedure reported for
2-tert-butyl-4-methylnitrophenol.57 PMe3, DMPE (Strem Chemicals),
and 1,4-dioxane (anhydrous, Aldrich) were stored under dinitrogen over
4 Å molecular sieves. All other reagents were used as received. Zinc
dust (97.5%) was purchased from Strem Chemicals and activated with
5% aqueous HCl prior to use.4 C6D6 and toluene-d8 (Cambridge Isotope
Laboratories) were degassed with dinitrogen and dried over 4 Å
molecular sieves for ∼1 day prior to use. NMR spectra were recorded
in C6D6 at 22 °C, unless otherwise noted. CDCl3 (Cambridge Isotope
Laboratories) was used as received. Elemental analyses were performed
in our laboratories on a Perkin-Elmer 2400 CHN analyzer or by H.
Kolbe, Mikroanalytisches Laboratorium (Mu¨hlheim an der Ruhr,
Germany).
Toluene (150 mL) and Me3SiCl (30 g, 276 mmol) were added to
the (i-PrNON)Ti(NMe2)2 obtained above. This mixture was heated to
110 °C in a sealed heavy-wall 500 mL one-neck flask behind a
blast shield. During the reaction the color changed from orange to
deep purple-black. After 21 h the solution was allowed to cool to
room temperature. The reaction flask was then brought into a dry-
box and stored at -25 °C overnight to complete formation of black
crystals. The supernatant was decanted off, and the black needles
were washed liberally with pentane and dried in vacuo: yield
All kinetic studies were carried out by proton NMR techniques at
either 300 or 500 MHz in C6D6. Temperatures did not change more
than 0.1 °C. Sublimed ferrocene was used as an internal standard (δ )
1
14.70 g (83%); H NMR δ 7.09 (d, 2), 6.72 (t, 2), 6.42 (m, 4), 5.91
(sept, 2, CHMe2), 1.45 (d, 12, CHMe2); 13C NMR δ 150.6, 145.0, 125.5,
119.7, 114.3, 111.4, 54.6 (CHMe2), 18.6 (CHMe2). Anal. Calcd for
C18H22Cl2N2OTi: C, 53.89; H, 5.53; N, 6.98. Found: C, 53.95; H, 5.59;
N, 6.94.
(47) Resconi, L.; Piemontesi, F.; Balboni, D.; Sironi, A.; Moret, M.;
Rychlicki, H.; Zeigler, R. Organometallics 1996, 15, 5046.
(48) Busico, V.; Cipullo, R.; Chadwick, J. C.; Modder, J. F.; Sudmeijer,
O. Macromolecules 1994, 27, 7538.
(49) Busico, V.; Cipullo, R.; Talarico, G.; Segre, A. L.; Caporaso, L.
Macromolecules 1998, 31, 8720.
(50) Busico, V.; Cipullo, R.; Talarico, G. Macromolecules 1998, 31, 2387.
(51) Guerra, G.; Longo, P.; Cavallo, L.; Corradini, P.; Resconi, L. J.
Am. Chem. Soc. 1997, 119, 4394.
(52) Benzing, E.; Kornicker, W. Chem. Ber. 1961, 94, 2263.
(53) Wilshire, J. F. K. Aust. J. Chem. 1988, 41, 995.
(54) Randall, J. J.; Lewis, C. E.; Slagan, P. M. J. Org. Chem. 1962, 27,
4098.
(55) Schrock, R. R.; Sancho, J.; Pederson, S. F. Inorg. Synth. 1989, 26,
44.
[i-PrNON]TiMe2. A solution of MeMgI in ether (420 µL, 3.0 M)
was added to a suspension of [i-PrNON]TiCl2 (250 mg, 623 µmol) in
toluene (10 mL) at -25 °C. The solution was allowed to warm to room
temperature and stirred for 2 h. Dioxane (135 mg, 1.53 mmol) was
added, and the solution was filtered through Celite. The filtrate was
concentrated to ∼1 mL, layered with pentane (2 mL), and stored at
-25 °C overnight. Orange microcrystals formed: yield 129 mg (57%);
1H NMR δ 7.36 (d, 2), 6.91 (t, 2), 6.83 (d, 2), 6.45 (t, 2), 5.97 (sept,
2, CHMe2), 1.61 (d, 12, CHMe2), 0.94 (s, 6, TiMe2); 13C NMR δ 148.8,
145.3, 125.1, 116.2, 113.6, 112.1, 56.8 (TiMe2), 49.5 (CHMe2), 19.6
(CHMe2). Anal. Calcd for C20H28N2OTi: C, 66.67; H, 7.83; N, 7.77.
Found: C, 66.52; H, 7.85; N, 7.64.
(56) Hodgkinson, W. R.; Limpach, L. J. Chem. Soc. 1893, 63, 104.
(57) Stegmann, H. B.; Dumm, H. V.; Scheffer, K. Phosphorus Sulfur
1978, 5, 159.

7834 J. Am. Chem. Soc., Vol. 121, No. 34, 1999
Baumann et al.
[i-PrNON]Ti(i-Bu)2. A solution of i-BuMgCl in ether (2.5 mL, 2.0
M) was added to a suspension of [i-PrNON]TiCl2 (1.00 g, 2.49 mmol)
in toluene (20 mL) at -25 °C. The mixture was rapidly stirred for 20
min without further cooling. 1,4-Dioxane (0.46 g, 5.2 mmol) was added,
and the mixture was filtered through Celite. The filtrate was concen-
trated to ∼5 mL. Orange microcrystals began to form during the
evaporation of the toluene. Pentane (5 mL) was added, and the mixture
removed in vacuo. The residue was extracted with pentane (40 mL).
The mixture was filtered through Celite, and the pale yellow filtrate
was quickly concentrated to yield a light orange oil (∼1 mL). At -25
1
°C an off-white amorphous solid formed: yield 312 mg (69%); H
NMR δ 7.34 (d, 2), 6.93 (t, 2), 6.75 (d, 2), 6.41 (t, 2), 4.62 (br sept, 2,
NCHMe2), 1.54 (d, 12, NCHMe2), 1.29 (t, 6, ZrCH2CH3), 0.91 (q,
4, ZrCH2CH3); 13C NMR (tol-d8, 0 °C) δ 147.0, 144.7, 115.1, 114.2,
113.1 (only 5 aromatic resonances were observed), 52.8 (t, JCH ) 115,
ZrCH2CH3), 46.8 (NCHMe2), 21.0 (NCHMe2, overlapped with
toluene signal), 11.1 (q, JCH ) 125, ZrCH2CH3). Anal. Calcd for
C22H32N2OZr: C, 61.21; H, 7.47; N, 6.49. Found: C, 61.28; H, 7.46;
N, 6.51.
1
was stored at -25 °C overnight: yield 767 mg (69%); H NMR δ
7.39 (d, 2), 6.92 (m, 4), 6.44 (t, 2), 5.99 (sept, 2, NCHMe2), 2.03 (sept,
2, CH2CHMe2), 1.72 (d, CH2CHMe2), 1.70 (d, NCHMe2, this and the
previous resonance are not entirely resolved, overall integration 16),
0.66 (d, 12, CH2CHMe2); 13C NMR (C6D6, 10 °C) δ 147.8, 145.3, 125.3,
115.7, 113.5, 112.8, 94.7 (CH2CHMe2), 50.7 (NCHMe2), 32.2
(CH2CHMe2), 27.3 (CH2CHMe2), 20.2 (NCHMe2). Anal. Calcd for
C26H40N2OTi: C, 70.26; H, 9.07; N, 6.30. Found: C, 70.12; H, 8.95;
N, 6.38.
[i-PrNON]ZrPr2. A 2.0 M solution of n-PrMgCl in ether (1.16 mL,
2.32 mmol) was added to a cooled (-30 °C) suspension of (i-PrNON)-
ZrCl2 (0.62 g, 1.16 mmol). The solution turned colorless, and a white
precipitate formed. After 5 min the ether was removed in vacuo, and
the residue was extracted with precooled (-30 °C) pentane. The ex-
tract was filtered through Celite, and the pentane was removed in
vacuo. The residue was redissolved in 1-2 mL of pentane, and the
solution was stored at -30 °C overnight to give large colorless crys-
[i-PrNON]Ti(CH2CMe3)2. A solution of Me3CCH2MgCl in ether
(3.8 mL, 1.35 M) was added to a suspension of [i-PrNON]TiCl2 (1.01
g, 2.52 mmol) in toluene (20 mL) at -25 °C. The solution was allowed
to warm to room temperature. After 12 h 1,4-dioxane (0.46 g, 5.2 mmol)
was added, and the mixture was filtered through Celite. The filtrate
was concentrated to ∼5 mL, layered with pentane (10 mL), and stored
at -25 °C. An orange-red microcrystalline solid was isolated: yield
1
tals: yield 0.44 g (62%); H NMR δ 7.34 (d, 2), 6.94 (t, 2), 6.78 (d,
2), 6.42 (t, 2), 4.72 (sept, 2, NCHMe2), 1.69 (m, 4, ZrCH2CH2CH3),
1.59 (d, 12, NCHMe2), 1.01 (m, 4, ZrCH2CH2CH3), 0.92 (t, 6,
ZrCH2CH2CH3); 13C NMR δ 147.0, 144.8, 125.7, 115.1, 114.2, 113.3
(Aryl-C), 66.4 (ZrCH2CH2CH3), 47.0 (NCHMe2), 21.7 (NCMe2),
21.2 (ZrCH2CH2CH3), 20.5 (ZrCH2CH2CH3). Anal. Calcd for
C24H36N2OZr: C, 62.70; H, 7.89; N, 6.09. Found: C, 62.59; H, 8.06;
N, 6.05.
[i-PrNON]Zr(i-Bu)2. Method A. A solution of i-BuMgCl in ether
(1.9 mL, 2.0 M) was added to a suspension of [i-PrNON]ZrCl2 (1.00
g, 1.86 mmol) in toluene (20 mL) at -25 °C. The mixture was allowed
to warm to room temperature. Over a period of 30 min the yellow
solid dissolved and was replaced by a fine white precipitate. 1,4-Dioxane
(0.34 g, 3.86 mmol) was added, and the mixture was filtered through
Celite. The pale yellow filtrate was concentrated to ∼5 mL, layered
with pentane (10 mL), and stored at -25 °C. A colorless crystalline
solid (713 mg) was isolated in two crops. Recrystallization from ether
at -25 °C produced analytically pure material: yield 459 mg (51%);
1H NMR δ 7.30 (d, 2), 6.93 (t, 2), 6.78 (d, 2), 6.40 (t, 2), 4.74 (br m,
2, NCHMe2), 2.19 (sept, 2, CH2CHMe2), 1.61 (d, 12, NCHMe2), 1.03
(d, 4, CH2CHMe2), 0.86 (d, 12, CH2CHMe2); 13C NMR δ 146.8, 144.9,
125.8, 115.2, 114.2, 113.5, 76.6 (CH2CHMe2), 47.3 (NCHMe2), 30.1
(CH2CHMe2), 28.5 (CH2CHMe2), 21.3 (NCHMe2). Anal. Calcd for
C26H40N2OZr: C, 64.01; H, 8.26; N, 5.74. Found: C, 64.08; H, 8.14;
N, 5.73.
Method B. A solution of H2[i-PrNON] (1.39 g, 4.89 mmol) in
toluene (10 mL) was added to a suspension of ZrCl4 (1.14 g, 4.89 mmol)
in toluene (40 mL) at room temperature. The mixture was stirred rapidly
for 5 h. The thick slurry was cooled to -25 °C, and a solution of
i-BuMgCl in ether (9.8 mL, 2.0 M) was added. After 30 min dioxane
(1.73 g, 19.7 mmol) was added, and the mixture was filtered through
Celite. All volatile components were removed in vacuo, leaving a dark
solid residue. The solid was recrystallized twice from ether at -25 °C
to give off-white microcrystals, yield 1.41 g (59%). The compound
was identical in all respects to that prepared by method A.
[i-PrNON]Zr(CH2CMe3)2. A solution of Me3CCH2MgCl in ether
(2.8 mL, 1.35 M) was added to a suspension of [i-PrNON]ZrCl2 (1.00
g, 1.86 mmol) in toluene (20 mL) at -25 °C. The mixture was allowed
to warm to room temperature and stirred for 8 h. Dioxane (0.34 g,
3.86 mmol) was added, and the mixture was filtered through Celite.
The pale yellow filtrate was concentrated to ∼3 mL and layered with
pentane (8 mL). Pale yellow microcrystals began to form. The mixture
was stored at -25 °C to complete crystallization: yield 673 mg (70%);
1H NMR δ 7.20 (d, 2), 6.94 (t, 2), 6.78 (d, 2), 6.38 (t, 2), 4.70 (br m,
2, CHMe2), 1.63 (d, 12, CHMe2), 1.21 (s, 4, CH2CMe3), 0.97 (s, 18,
CH2CMe3); 13C NMR δ 146.3, 145.1, 126.0, 115.1, 114.3, 113.9, 84.6
(CH2CMe3), 48.1 (CHMe2), 36.5 (CH2CMe3), 35.4 (CH2CMe3), 21.2
(CHMe2). Anal. Calcd for C28H44N2OZr: C, 65.19; H, 8.60; N, 5.43.
Found: C, 65.36; H, 8.47; N, 5.41.
1
606 mg (51%); H NMR δ 7.31 (d, 2), 6.93 (m, 4), 6.43 (t, 2), 5.84
(sept, 2, CHMe2), 2.04 (s, 4, CH2CMe3), 1.73 (d, 12, CHMe2), 0.83 (s,
18, CH2CMe3); 13C NMR δ 147.1, 145.9, 125.5, 115.7, 113.6, 113.4,
103.9 (CH2CMe3), 51.6 (CHMe2), 38.7 (CH2CMe3), 34.5 (CH2CMe3),
21.0 (CHMe2). Anal. Calcd for C28H44N2OTi: C, 71.17; H, 9.39; N,
5.93. Found: C, 71.11; H, 9.32; N, 6.05.
[i-PrNON]ZrCl2. H2[i-PrNON] (3.02 g, 10.6 mmol) and Zr(NMe2)4
(2.84 g, 10.6 mmol) were dissolved in pentane (40 mL). The solution
was stirred at room temperature for 3 h. All volatile components were
removed in vacuo. Traces of HNMe2 were removed by repeated
trituration with pentane and gentle heating of the reaction flask. Proton
and carbon NMR spectra of the pale yellow oil were consistent with it
being (i-PrNON)Zr(NMe2)2: 1H NMR δ 7.30 (d, 2), 7.02 (t, 2), 6.77
(d, 2), 6.43 (t, 2), 3.91 (sept, 2, CHMe2), 2.78 (s, 12, NMe2), 1.31 (d,
12, CHMe2); 13C NMR δ 148.0, 144.7, 126.4, 115.0, 113.8, 113.7, 51.2,
42.3, 23.9 (CHMe2).
Toluene (40 mL) and Me3SiCl (2.9 g, 26.7 mmol) were added to
the oil. The solution quickly turned bright orange. After 14 h a small
amount of solid was filtered off, and pentane (40 mL) was added to
the filtrate. The layers were mixed, and the solution was then stored at
-25 °C. A yellow, sometimes crystalline solid precipitated. The solid
was filtered off and recrystallized from a mixture of toluene and pentane
1
at -25 °C; yield 4.11 g (72%). According to H NMR 1 equiv of
toluene was present: 1H NMR (CD2Cl2, toluene resonances not given)
δ 7.67 (d, 2), 7.08 (t, 2), 6.83 (d, 2), 6.77 (t, 2), 4.66 (sept, 2, CHMe2),
1.52 (d, 12, CHMe2); 13C NMR (CD2Cl2, toluene resonances not given)
δ 148.2, 143.4, 126.1, 117.7, 114.7, 113.8, 48.9 (CHMe2), 20.0
(CHMe2). Anal. Calcd for C25H30Cl2N2OZr: C, 55.95; H, 5.63; N, 5.22.
Found: C, 55.84; H, 5.61; N, 5.27.
[i-PrNON]ZrMe2. A solution of MeMgI in ether (1.25 mL, 3.0 M)
was added to a suspension of [i-PrNON]ZrCl2 (1.01 g, 1.88 mmol) in
toluene (20 mL) at -25 °C. The mixture was allowed to warm to room
temperature and was stirred for 30 min. Dioxane (0.35 g, 4 mmol)
was added, and the mixture was filtered through Celite. The pale yellow
solution was concentrated to ∼5 mL and layered with pentane (∼10
mL). The mixture was stored at -25 °C overnight, and an off-white
microcrystalline solid was isolated (565 mg, 74%): 1H NMR δ 7.35
(d, 2), 6.92 (t, 2), 6.73 (d, 2), 6.41 (t, 2), 4.65 (m, 2, CHMe2), 1.53 (d,
12, CHMe2), 0.43 (s, 6, ZrMe2); 13C NMR δ 147.4, 144.8, 125.7, 115.3,
114.2, 113.0, 46.6 (CHMe2), 41.6 (ZrMe2), 21.0 (CHMe2). Anal. Calcd
for C20H28N2OZr: C, 59.51; H, 6.99; N, 6.94. Found: C, 59.49; H,
6.87; N, 6.87.
An X-ray-quality crystal was obtained from a pentane solution at
-25 °C.
[i-PrNON]ZrEt2. A solution of EtMgBr in ether (700 µL, 3.0 M)
was added to a solution of [i-PrNON]ZrCl2 (562 mg, 1.05 mmol) in
ether (30 mL) at -25 °C. After 5 min the yellow color disappeared,
and a fine white precipitate formed. All volatile components were
{[i-PrNON]Ti(PMe3)2}2(µ-N2). Inside a dinitrogen-filled drybox
[i-PrNON]Ti(i-Bu)2 (202 mg, 454 µmol) was dissolved in ether (20

Ti and Zr Complexes Containing Diamido Ligands
J. Am. Chem. Soc., Vol. 121, No. 34, 1999 7835
mL) containing PMe3 (145 mg, 1.91 mmol). The solution was
transferred to a 100 mL one-neck flask which was then capped and
allowed to stand at room temperature. The color changed gradually
from bright orange to green black, and black crystals began to form.
After 24 h the mixture was cooled to -25 °C to complete crystallization;
[i-PrNON]Zr(i-Bu)2(PMe3). PMe3 (93 mg, 1.22 mmol) was added
to a suspension of [i-PrNON]Zr(i-Bu)2 (209 mg, 428 µmol) in pentane
(5 mL) at room temperature. The solid rapidly dissolved, and the color
changed to pale yellow. The solution was concentrated to ∼1 mL. Pale
yellow microcrystals began to form. More PMe3 (100 mg, 1.32 mmol)
was added, and the mixture was stored at -25 °C overnight. The
supernatant was decanted off, and the solid was rinsed with a little
pentane and dried in vacuo: yield 115 mg (48%); 1H NMR (∼0.013M
in C6D6) δ 7.31 (d, 2), 6.92 (t, 2), 6.78 (d, 2), 6.40 (t, 2), 4.78 (br s, 2,
NCHMe2), 2.20 (sept, 2, CH2CHMe2), 1.61 (d, 12, NCHMe2), 1.03 (d,
4, CH2CHMe2), 0.88 (d, 12, CH2CHMe2), 0.74 (d, 9, PMe3, JPH ) 2);
1H NMR (∼0.13 M in C6D6) δ 7.32 (d, 2), 6.90 (t, 2), 6.76 (d, 2), 6.39
(t, 2), 4.89 (br s, 2), 2.24 (sept, 2), 1.61 (d, 12), 1.01 (d, 4), 0.95 (d,
12, CH2CHMe2), 0.63 (s, 9, PMe3); 13C NMR (∼0.13 M in C6D6) δ
146.7, 145.1, 125.7, 115.0, 114.1, 113.6, 76.1 (CH2CHMe2), 46.8
(NCHMe2), 30.5 (CH2CHMe2), 28.8 (CH2CHMe2), 21.3 (NCHMe2),
15.8 (d, PMe3, JPC ) 8); 31P NMR (∼0.13 M in C6D6) δ -57.2. Anal.
Calcd for C29H49N2OPZr: C, 61.77; H, 8.76; N, 4.97. Found: C, 61.82;
H, 8.62; N, 5.01.
1
yield 147 mg (61%). According to H NMR 0.9 equiv of ether was
present: 1H NMR (C6D6, ether resonances not given) δ 7.14 (partially
overlapped with C6D5H resonance), 6.99 (t, 4), 6.82 (d, 4), 6.35 (t, 4),
4.77 (br sept, 4, CHMe2), 1.70 (d, 24, CHMe2), 0.88 (d, 36, PMe3, JPH
) 3.5); 13C NMR (C6D6, ether resonances ignored) δ 149.3, 145.7,
125.8, 114.9, 112.8, 112.4, 54.2 (CHMe2), 25.2 (CHMe2) 15.3 (d, PMe3,
JPC ) 4.5); 31P NMR δ -37.6. Anal. Calcd for C51.6H89N6O3P4Ti2: C,
58.49; H, 8.47; N, 7.93. Found: C, 58.35; H, 8.37; N, 7.86.
(i-PrNC6H4)(i-PrNC6H4O)Ti(dmpe). Inside a glovebox DMPE (71
mg, 473 µmol) was added to a suspension of [i-PrNON]Ti(i-Bu)2 (201
mg, 452 µmol) in benzene (5 mL). The solid rapidly dissolved, and
the solution began to darken within a few minutes. The solution was
transferred to a 25 mL Schlenk tube which was then subjected to two
freeze-pump-thaw cycles (120 mTorr of residual pressure). The
reaction mixture was allowed to stand in the dark at room temperature.
The color slowly changed to dark red-black. After 24 h the Schlenk
tube was brought back into the glovebox, and all volatile components
were removed in vacuo. The black residue was redissolved in a
minimum of toluene (∼3 mL), concentrated to ∼1 mL, and layered
with pentane (2 mL). The mixture was stored at -25 °C overnight to
yield 137 mg (63%) of black crystals: 1H NMR δ 7.32 (t, 1), 7.14
(partially overlapped with C6D5H), 6.90-6.85 (m, 3), 6.81 (t, 1), 6.74
(m, 1), 6.32 (m, 1), 4.05 (sept, 1, NCHMe2), 3.92 (sept, NCHMe2),
1.58 (d, 3), 1.3-0.8 (br m), 1.31 (d), 1.06 (d), 1.03 (d), 0.89 (d), total
integration of previous peaks 22, 0.12 (br s, 3); 13C NMR δ 173.9,
159.6, 154.3, 148.1, 130.3, 120.7, 119.5, 116.6 (t, JCP ≈ 2), 113.5,
108.3, 105.6 (11 aromatic resonances were observed), 53.3, 51.2, 27.2
(br s, DMPE), 26.3 (br s, DMPE), 25.1, 24.6, 24.3, 23.1, 12.7 (br s,
DMPE), 11.8 (br s, DMPE), 10.8 (br s, DMPE); 31P NMR δ -0.3 (br
s), -8.6 (br s). Anal. Calcd for C24H38N2OP2Ti: C, 60.00; H, 7.97; N,
5.83. Found: C, 59.86; H, 7.91; N, 5.75.
[i-PrNON]Ti(CHCMe3)(PMe3)2. A suspension of [i-PrNON]Ti-
(CH2CMe3)2 (300 mg, 635 µmol) in toluene (6 mL) and PMe3 (0.60 g,
7.9 mmol) was stirred in a sealed 50 mL Schlenk tube at 45 °C. The
solid quickly dissolved, and the initially bright orange solution turned
green-black within a few hours. After 12 h the solution was concentrated
to ∼1 mL. Black crystals began to form during the evaporation of the
toluene. More PMe3 (∼150 mg) was added, and the mixture was stored
at -25 °C overnight to afford black crystals; yield 178 mg (51%). Shifts
in 1H and 13C NMR spectra vary slightly depending on the concentration
of the sample: 1H NMR (0.06 M in C6D6) δ 7.12 (d, 2), 6.90 (t, 2),
6.67 (d, 2), 6.33 (t, 2), 4.97 (br m, 2, NCHMe2), 3.00 (s, 1, CHCMe3),
1.62 (d, 12, NCHMe2), 1.20 (s, 9, CHCMe3), 0.90 (d, JPH ) 4, 18,
PMe3); 13C NMR (0.06 M in C6D6) δ 230.1 (in gated decoupled 13C,
d, JCH ) 80, CHCMe3), 149.1, 144.9, 125.1, 113.8, 113.5, 112.6, 53.2
(NCHMe2), 47.6 (CHCMe3), 33.8, 24.3, 16.5 (PMe3); 31P NMR (0.06
M in C6D6) δ -30.9 (br s). Anal. Calcd for C29H50N2OP2Ti: C, 63.04;
H, 9.12; N, 5.07. Found: C, 63.12; H, 8.98; N, 5.20.
[i-PrNON]Zr(η2-CH2CMe2)(PMe3)2. A solution of [i-PrNON]Zr-
(i-Bu)2 (303 mg, 621 µmol) in PMe3 (2.5 g, 32.9 mmol) was heated in
a sealed 25 mL Schlenk tube to 37 °C in the dark. After 63 h the
solution had turned red and was transferred into a 20 mL vial. Needles
began to form immediately. Ether (∼1 mL) was added, and the mixture
was stored at -25 °C overnight. The supernatant was decanted off,
and the solid was washed with pentane to give pink-brown needles
(185 mg). The washings were combined, concentrated, and stored at
-25 °C to give a second crop (29 mg) as a powder: overall yield 214
1
mg (59%); H NMR δ 7.38 (d, 2), 6.82 (t, 2), 6.61 (d, 2), 6.40 (t, 2),
3.43 (sept, 2, NCHMeAMeB), 1.94 (s, 6, CH2CMe2), 1.23 (d, 6,
NCHMeAMeB), 1.17 (d, 6, NCHMeAMeB), 0.96 (s, CH2CMe2), 0.93
(br s, PMe3), 0.86 (br s, PMe3 previous 3 peaks overlapped, overall
integration 20); 13C NMR δ 147.0, 146.1, 124.6, 115.8, 113.9, 113.6,
63.6 (CH2CMe2), 52.3 (CH2CMe2), 44.2 (NCHMeAMeB), 34.2
(CH2CMe2), 22.6 (NCHMeAMeB), 22.4 (NCHMeAMeB), 16.3 (br s,
PMe3), 15.5 (br s, PMe3); 31P NMR δ -26.0 (br s), -60.1 (br s). Anal.
Calcd for C28H48N2OP2Zr: C, 57.80; H, 8.31; N, 4.81. Found: C, 57.63;
H, 8.24; N, 4.72.
[i-PrNON]Zr(η2-C2H4)(PMe3)2. [i-PrNON]ZrEt2 (13 mg, 0.03
mmol) was dissolved in 0.5 mL of C6D6, and PMe3 (5 mg, 0.07 mmol)
1
was added. The solution was transferred to a NMR tube, and the H
and 13C NMR spectra were recorded: 1H NMR δ 7.42 (d, 2), 6.88 (t,
2), 6.47 (d, 2), 6.43 (t, 2), 3.07 (sept, 2, NCHMe2), 1.20 (s, C2H4),
0.97 (d, 12, NCHMe2), 0.90 (s, 18, PMe3); 13C NMR δ 147.7, 145.7,
125.1, 113.9, 113.7, 113.3 (CAr), 45.7 (C2H4), 44.0 (NCHMe2), 21.6
(NCHMe2), 15.1 (br, PMe3). No further products besides C2H6 (δ )
0.80) could be observed.
[i-PrNON]Zr(η2-CH2CHMe)(PMe3)2. [i-PrNON]ZrPr2 (32 mg, 0.07
mmol) was dissolved in 0.5 mL of C6D6, and PMe3 (11 mg, 0.14 mmol)
1
was added. The solution was transferred to a NMR tube, and a H
NMR spectrum was recorded. After 20 min, starting material, product,
and propane resonances could be observed. After the reaction was
complete (3 h), the solution was frozen and all volatile components
were removed in vacuo: 1H NMR δ 7.40 (m, 2), 6.86 (m, 2), 6.59 (d,
1), 6.48 (d, 1), 6.41 (m, 2). 3.16 (sept, 1, NCHMe2), 3.00 (sept, 1,
{[i-PrNON]ZrEt}2(µ-C2H4). [i-PrNON]ZrEt2 (239 mg, 554 µmol)
was dissolved in pentane (6 mL). The sometimes slightly cloudy
solution was filtered and allowed to stand at room temperature. During
the reaction the color changed from very pale yellow to red-orange.
After a few hours pale yellow needles began to form. After 21 h the
supernatant was decanted off, and the solid was washed liberally with
pentane and dried in vacuo; yield 173 mg (75%). The product is
3
3
NCHMe2), 2.18 (d, J ) 6.5, 3, CH2dCHMe), 1.56 (dd, J ) 9 and
13, 1, CH2CHMe), 1.17 (d, 1, NCHMe2), 1.12 (d, 1, NCHMe2), 1.06
(d, 1, NCHMe2), 1.02 (d, 1, NCHMe2), 0.88 (s, br, ∼18, PMe3), 0.75
(dd, 3J ) 9 and 11, 1, CH2CHMe); the resonance of one olefinic proton
could not be detected and is probably buried under the broad PMe3
resonance; 13C NMR δ 147.4, 147.3, 146.1, 145.5, 125.0, 124.8, 114.8,
114.5, 114.2, 113.3, 113.2, 113.0 (CAr), 53.3 (CH2)CHMe), 52.0
(CH2dCHMe), 46.4 (NCHMe2), 44.1 (NCHMe2), 26.5 (CH2dCHMe),
24.1 (NCHMe2), 22.7 (NCHMe2), 21.6 (NCHMe2), 21.5 (NCHMe2),
15.4 (br, PMe3). No other products were observed.
1
contaminated with traces (∼2% by H NMR) of [i-PrNON]ZrEt2. A
sample free of [i-PrNON]ZrEt2 may be obtained as a yellow powder
from a saturated toluene solution layered with ether: 1H NMR δ 7.47
(d, 4), 6.95 (t, 4), 6.76 (d, 4), 6.49 (t, 4), 3.75 (sept, 4, NCHMeAMeB),
1.45 (t, 6, ZrCH2CH3), 1.38 (d, 12, NCHMeAMeB), 1.29 (d, 12,
NCHMeAMeB), 1.16 (q, 4, ZrCH2CH3), 1.09 (s, 4, µ-C2H4); 13C NMR
δ 146.7, 144.8, 125.5, 115.6, 114.5, 114.2, 46.3 (NCHMeAMeB), 44.7
(t, JCH ) 117, ZrCH2CH3), 39.4 (t, JCH ) 143, µ-C2H4), 21.7 (NCHMeA-
MeB), 20.9 (NCHMeAMeB), 13.0 (q, JCH ) 124, ZrCH2CH3). Anal.
Calcd for C42H58N4O2Zr2: C, 60.53; H, 7.01; N, 6.72. Found: C, 60.19;
H, 6.95; N, 6.73.
H2[CyNON]. To a 250 mL round-bottom one-necked flask equipped
with a condenser and charged with a Teflon-sealed stir bar were added
O(o-C6H4NH2)2 (5.03 g, 0.025 mol), cyclohexanone (4.93 g, 0.050 mol,
2 equiv), zinc (16.42 g, 0.25 mol, 10 equiv), and acetic acid (100 mL).
The mixture was heated to 65 °C under nitrogen for 29 h. The gray-
white suspension containing residual zinc was cooled to room tem-

7836 J. Am. Chem. Soc., Vol. 121, No. 34, 1999
Baumann et al.
perature, and methanol (100 mL) was added. The white precipitate was
filtered off, and the filter cake was washed with MeOH (100 mL). The
filtrate and combined washings were concentrated on a rotary evaporator
to ∼10 mL, and crushed ice (∼100 g) and CH2Cl2 (100 mL) were
added. Ammonium hydroxide was then added until pH > 10. The
organic layer was separated from the aqueous layer, which was further
washed with CH2Cl2 (100 mL). The combined organic solutions were
dried over MgSO4, and the solvent was removed under reduced pressure
to give the product as a yellow oil which was pure by NMR and used
for subsequent organometallic chemistry without further purification:
yield 8.03 g (88%); 1H NMR δ 6.99 (t, 2, Ar), 6.93 (d, 2, Ar), 6.71 (d,
2, Ar), 6.59 (t, 2, Ar), 4.29 (s, 2, NH), 3.15 (m, 2, Cy), 1.88 (m, 4,
Cy), 1.49 (m, 4, Cy), 1.39 (m, 2, Cy), 1.12 (m, 4, Cy), 0.95 (m, 6, Cy);
1H NMR (CDCl3) δ 6.99 (t, 2, Ar), 6.76 (m, 4, Ar), 6.59 (t, 2, Ar),
4.19 (br s, 2, NH), 3.32 (m, 2, Cy), 2.08 (m, 4, Cy), 1.77-1.63 (m, 6,
Cy), 1.44-1.16 (m, 10, Cy); 13C NMR (CDCl3) δ 143.98 (C, Ar),
139.12 (C, Ar), 124.33 (CH, Ar), 118.02 (CH, Ar), 116.27 (CH, Ar),
11.92 (CH, Ar), 51.54 (CH, Cy), 33.46 (CH2, Cy), 26.07 (CH2, Cy),
25.09 (CH2, Cy); HRMS (EI, 70 eV) 364.25132, calcd for C24H32N2O
364.251464.
by the above method. To a solution of {[CyNON]ZrMe(NMe2Ph)}-
{B(C6F5)4} in bromobenzene-d5 at 0 °C was added 50 equiv of
1-hexene. The solution was transferred to an NMR tube and kept at 0
1
°C. The reaction was monitored by H NMR.
X-ray Structures. A Siemens SMART/CCD area detection system
with Mo KR radiation (λ ) 0.710 73 Å) was used for all determinations.
Cell determination, data collection, and structure solution and refinement
were performed with the SMART SAINT and SHELXTL 5.0 software
packages. All non-hydrogen atoms were refined anisotropically, and
all hydrogen atoms were placed in calculated positions. {[i-PrNON]-
Ti(PMe3)2}2(µ-N2) contained disordered ether which was removed
during the refinement procedure using the SQUEEZE58 routine. During
the refinement of [i-PrNON]Ti(CH2CHMe2)2 it appeared that the
isobutyl methyne carbons (C(2) and C(6)) were disordered over two
positions close to one another. Subsequent refinement led to collapse
of the two positions to the one position, which is the one that is reported.
The tert-butyl group in [i-PrNON]Ti(CHCMe3)(PMe3)2 was found to
be disordered rotationally over two equally populated positions and
was refined as such. All six methyl carbon atom positions are listed in
the table of positional parameters in the Supporting Information.
The synthetic procedures for [CyNON]2- complexes are analogous
to those for [i-PrNON]2- complexes. Syntheses and spectroscopic de-
tails can be found in the Supporting Information for [CyNON]Ti-
(NMe2)2, [CyNON]TiCl2, [CyNON]TiMe2, [CyNON]Ti(CH2CMe3)2,
[CyNON]Ti(i-Bu)2, [CyNON]Zr(NMe2)2, [CyNON]ZrCl2, [CyNON]Zr-
(CH2SiMe3)2, [CyNON]ZrMe2, [CyNON]Zr(CH2CMe3)2, [CyNON]-
ZrEt2, [CyNON]Zr(i-Bu)2, [CyNON]Zr(CH2CHdCH2)2, [CyNON]-
Zr(η2-C2H4)(PMe3)2, and [CyNON]Zr(η2-H2CCMe2)(PMe3)2.
{[CyNON]ZrMe(NMe2Ph)}{B(C6F5)4} Cation Observation. A
cold solution (-35 °C) of [CyNON]ZrMe2 (10 mg, 0.021 mmol) in
bromobenzene-d5 (0.7 mL) was added to solid [HNMe2Ph][B(C6F5)4]
(17 mg, 0.021 mmol) at -35 °C. The color changed from colorless to
yellow immediately, and then turned yellow-orange in ∼30 s. The
solution was stirred at room temperature briefly, and was then cooled
to -35 °C again before being transferred to a Teflon-sealed NMR tube,
which was then frozen in liquid nitrogen. Proton spectra were acquired
at several temperatures starting at -30 °C. The spectra were essentially
the same between -30 and +20 °C: 1H NMR (C6D5Br, 0 °C) δ 7.24
(d, 2, Ar), 7.19 (t, 2, Ar), 7.00 (m, 3, Ar), 6.75 (d, 2, Ar), 6.68 (m, 4,
Ar), 3.37 (tt, 2, Cy), 2.49 (s, 6, NMe2Ph), 1.85 (m, 4, Cy), 1.73 (m, 4,
Cy), 1.58 (m, 4, Cy), 1.23 (m, 2, Cy), 1.07 (m, 6, Cy), 0.65 (s, 3,
ZrMe).
Acknowledgment. R.R.S. is grateful to the National Insti-
tutes of Health (Grant GM 31978) and the National Science
Foundation (Grant CHE-9700736) for research support. We also
thank Dr. David McConville for disclosing results prior to
publication and for fruitful discussions. R.B. thanks the Studi-
enstiftung des Deutschen Volkes for a predoctoral fellowship,
while R.S. thanks the Alexander von Humboldt Foundation for
a Feodor-Lynen-Fellowship.
Supporting Information Available: Experimental details
for the synthesis of [CyNON]2- complexes, along with ORTEP
drawings, crystal data, atomic coordinates, bond lengths and
angles, and anisotropic displacement parameters for [i-PrNON]-
Ti(CH2CHMe2)2 (1), [i-PrNON]ZrMe2 (2), [i-PrNON]Zr-
(CH2CHMe2)2(PMe3) (3), [i-PrNON]Zr(CH2CMe2)(PMe3)2 (4),
[i-PrNON]Ti(PMe3)2]2(µ-N2) (5), (i-PrNC6H4)(i-PrNC6H4O)Ti-
(dmpe) (6), and [i-PrNON]Ti(CHCMe3)(PMe3)2 (7) (PDF). This
material is available free of charge via the Internet at
http://pubs.acs.org.
JA983549V
Catalytic Reaction between {[CyNON]ZrMe(NMe2Ph)}{B(C6F5)4}
and 1-Hexene. {[CyNON]ZrMe(NMe2Ph)}{B(C6F5)4} was prepared
(58) van der Sluis, P.; Spek, A. L. Acta Crystallogr. 1990, A46, 194.