Synthesis of Zirconium Complexes Containing the Tridentate Diamido Ligands [(t-Bu-d6-N-o-C6H4)2S]62- And [i-PrN-o-C6H4S]2-

Article abstract of DOI:10.1021/om980934k

The thioethers (t-Bu-d₆-NH-o-C₆H₄)₂S (H₂[t-BuNSN]) and (i-PrNH-o-C₆H₄₂S (H₂[i-PrNSN]) have been prepared in three steps in good yield. Zirconium complexes that contain the [t-BuNSN]_2- ligand ([t-BuNSN]Zr(NMe₂)₂, [t-BuNSN]ZrMe₂, and (t-BuNSN]ZrMe₂) were prepared readily, but the last is unstable, and no higher alkyl homologues could be prepared. In contrast, [i-PrNSN]ZrMe₂ and [i-PrNSN]Zr(CH₂CH₂Me₂)₂ are both stable, even at elevated temperatures. An X-ray study of [t-BuNSN]ZrCl(NMe₂) showed it to have a geometry about zirconium that is approximately trigonal bipyramidal with the chloride and sulfur atoms in the axial positions. An X-ray study of [i-PrNSN]ZrMe₂ showed it to have approximately a square-pyramidal structure with a methyl group in the apical position. Low-temperature solution ¹H and ¹³C NMR spectra of [t-BuNSN]_2- and [i-PrNSN]_2- species are consistent with the solid-state structures, although inversion at sulfur on the NMR time scale results in equilibration of the two metal substituents (e.g., methyl groups). Cationic complexes prepared from [RNSN]ZrMe₂ precursors (R = t-Bu, i-Pr) were neither stable at 22°C nor active for the controlled polymerization of 1-hexene.

Full text of DOI:10.1021/om980934k

Organometallics 1999, 18, 843-852
843
Syn th esis of Zir con iu m Com p lexes Con ta in in g th e
Tr id en ta te Dia m id o Liga n d s [(t-Bu -d 6-N-o-C6H4)2S]2- a n d
[(i-P r N-o-C6H4)2S]2-
David D. Graf, Richard R. Schrock,* William M. Davis, and Ru¨diger Stumpf
Department of Chemistry 6-331, Massachusetts Institute of Technology,
77 Massachusetts Avenue, Cambridge, Massachusetts 02139
Received November 17, 1998
The thioethers (t-Bu-d6-NH-o-C6H4)2S (H2[t-BuNSN]) and (i-PrNH-o-C6H4)2S (H2[i-PrNSN])
have been prepared in three steps in good yield. Zirconium complexes that contain the
[t-BuNSN]2- ligand ([t-BuNSN]Zr(NMe2)2, [t-BuNSN]ZrCl2, and [t-BuNSN]ZrMe2) were
prepared readily, but the last is unstable, and no higher alkyl homologues could be prepared.
In contrast, [i-PrNSN]ZrMe2 and [i-PrNSN]Zr(CH2CHMe2)2 are both stable, even at elevated
temperatures. An X-ray study of [t-BuNSN]ZrCl(NMe2) showed it to have a geometry about
zirconium that is approximately trigonal bipyramidal with the chloride and sulfur atoms in
the axial positions. An X-ray study of [i-PrNSN]ZrMe2 showed it to have approximately a
square-pyramidal structure with a methyl group in the apical position. Low-temperature
1
solution H and 13C NMR spectra of [t-BuNSN]2- and [i-PrNSN]2- species are consistent
with the solid-state structures, although inversion at sulfur on the NMR time scale results
in equilibration of the two metal substituents (e.g., methyl groups). Cationic complexes
prepared from [RNSN]ZrMe2 precursors (R ) t-Bu, i-Pr) were neither stable at 22 °C nor
active for the controlled polymerization of 1-hexene.
In tr od u ction
[t-BuNON]TiMe2 a well-behaved initiator for the po-
lymerization of 1-hexene.
We recently reported that [t-BuNON]ZrMe2, where
[t-BuNON]2- is the tridentate diamido ligand [(t-Bu-
d6-N-o-C6H4)2O]2-, upon activation with [PhNMe2H]-
[B(C6F5)4] or [Ph3C][B(C6F5)4] in chlorobenzene, will
initiate the living polymerization of up to ∼500 equiv
of 1-hexene.1 Through 13C labeling studies we showed
that living poly(1-nonene) is formed in a 1,2-insertion
process, that it is stable and observable by NMR at 22
°C in the absence of excess 1-nonene, and that it
decomposes within minutes at 40 °C.2 The fact that the
living polymer is stable in the absence of excess 1-hex-
ene at 22 °C distinguishes this zirconium system from
the first reported system for the living polymerization
of 1-hexene, one based on titanium complexes that
contain a diamido ligand of the type [ArylNCH2CH2-
CH2NAryl]2- (Aryl ) 2,6-i-Pr2C6H3, 2,6-Me2C6H3),3-5
since in the titanium system a termination step destroys
the catalyst when 1-hexene is depleted.6 There is no
report of [ArylNCH2CH2CH2NAryl]2- zirconium or hafni-
um complexes behaving in a living manner as initiators
for the polymerization of 1-hexene, nor is activated
We are interested in exploring the utility and versa-
tility of diamido/donor ligands that are related to
[t-BuNON]2-. For example, we have recently published
syntheses of titanium, zirconium, and hafnium com-
plexes of the type [(ArylNCH2CH2)2O]MR2 (Aryl ) a 2,6-
disubstituted aryl ring) and the reactivity of activated
versions for the polymerization of 1-hexene.7 We also
briefly explored the synthesis and activation of [(Aryl-
NCH2CH2)2S]ZrMe2 for the polymerization of 1-hexene.
Activated [(ArylNCH2CH2)2S]ZrMe2 complexes appeared
to be relatively well-behaved initiators for the polym-
erization of 1-hexene, although less stable than initia-
tors derived from the analogous [(ArylNCH2CH2)2O]-
ZrMe2 compounds. Therefore, we sought also to prepare
[(t-Bu-d6-NH-o-C6H4)2S] (H2[t-BuNSN]) and zirconium
alkyl complexes containing the [t-BuNSN]2- ligand. We
report the results of this effort here. We also explore
briefly the role of sterics in the amido alkyl group by
preparing [(i-PrNH-o-C6H4)2S] (H2[i-PrNSN]) and com-
plexes that contain the [(i-PrN-o-C6H4)2S]2- ligand, and
attempting to prepare well-behaved initiators that
contain either a [t-BuNSN]2- or a [i-PrNSN]2- ligand.
(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,
Resu lts
69.
(3) Scollard, J . D.; McConville, D. H. J . Am. Chem. Soc. 1996, 118,
10008.
(4) Scollard, J . D.; McConville, D. H.; Vittal, J . J . Organometallics
1995, 14, 5478.
(5) Scollard, J . D.; McConville, D. H.; Payne, N. C.; Vittal, J . J .
Macromolecules 1996, 29, 5241.
Syn th esis of Liga n d s. H2[t-BuNSN] was synthe-
sized in four steps that are entirely analogous to those
employed to prepare H2[t-BuNON] (Scheme 1).1 The
(6) Scollard, J . D.; McConville, D. H.; Rettig, S. J . Organometallics
1997, 16, 1810.
(7) Aizenberg, M.; Turculet, L.; Davis, W. M.; Schattenmann, F.;
Schrock, R. R. Organometallics 1998, 17, 4795.
10.1021/om980934k CCC: $18.00 © 1999 American Chemical Society
Publication on Web 02/09/1999

844 Organometallics, Vol. 18, No. 5, 1999
Graf et al.
Sch em e 1
48 h), although when the temperature is raised to 60
°C it is complete in 18 h as long as the liberated
dimethylamine is periodically removed in vacuo. Re-
crystallization of the crude product gives colorless blocks
of analytically pure [t-BuNSN]Zr(NMe2)2 in 77% yield.
The reaction between H2[i-PrNSN] and Zr(NMe2)4 is
complete within 4 h at room temperature to give
[i-PrNSN]Zr(NMe2)2 as a yellow oil which occasionally
solidifies in cold pentane. The faster reaction between
H2[i-PrNSN] and Zr(NMe2)4 can be ascribed to the
greater ability of the isopropyl-substituted amine to
coordinate to Zr in a crowded environment and activate
the amine proton toward abstraction by a dimethyla-
mido group.
[t-BuNSN]Zr(NMe2)2 reacts with 1 equiv of Me3SiCl
in diethyl ether to give [t-BuNSN]ZrCl(NMe2) (70%) and
with excess (4-5 equiv) Me3SiCl to yield [t-BuNSN]-
ZrCl2 (94%; eq 2). [t-BuNSN]ZrCl2 precipitates as a
bright yellow solid (>95% pure by NMR) that can be
used directly for further reactions. [i-PrNSN]Zr(NMe2)2
also reacts with an excess (4-5 equiv) of Me3SiCl in
toluene to give [i-PrNSN]ZrCl2 in high yield. By analogy
9
with the observed dimeric structure of [t-BuNON]ZrCl2
coupling between o-C6H4(NO2)(SH) and o-C6H4(NO2)F
to give S(o-C6H4NO2)2 and the reduction of S(o-C6H4-
NO2)2 to S(o-C6H4NH2)2 both proceeded in high yield.
The reaction between S(o-C6H4NH2)2 and neat acetone-
d6 (10 equiv) in the presence of activated 4 Å molecular
sieves quantitatively provided S[o-C6H4NdC(CD3)2]2 in
6-10 days. Addition of the diimine to LiMe followed by
hydrolysis gave H2[t-BuNSN] as a light orange solid in
∼50% yield (Scheme 1). As is the case in the synthesis
of H2[t-BuNON], the tert-butyl resonance typically
integrates as ∼7 protons and is broadened on the right
side at the base as a consequence of ∼1 NH proton being
exchanged into the CD3 groups to yield some CD2H
groups at some point during the synthesis of the ligand.
NMR spectra suggest that under the reaction conditions
at least some H/D exchange occurs before the imine is
formed, presumably in acetone-d6 itself.
H2[i-PrNSN] was prepared readily in high yield (92%)
by the reaction between S(o-C6H4NH2)2 and acetone in
the presence of zinc dust in glacial acetic acid (eq 1).
This procedure is based on that reported for the
synthesis of a variety of alkylated anilines.8 H2[i-
PrNSN] was obtained as an oil that sometimes crystal-
lizes after several weeks at room temperature.
and because of the relatively low solubility of the
[RNSN]ZrCl2 complexes, we propose that [t-BuNSN]-
ZrCl2 and [i-PrNSN]ZrCl2 are also dimeric species
containing two bridging chlorides.
Thermally unstable [t-BuNSN]ZrMe2 is formed cleanly
at -30 °C upon addition of a THF solution of MeMgCl
or MeMgI to a diethyl ether solution of [t-BuNSN]ZrCl2.
If diethyl ether is employed as the solvent for the
Grignard reagent, the resulting [t-BuNSN]ZrMe2 is only
∼80% pure. In diethyl ether or in a diethyl ether/THF
mixture LiMe reacts with [t-BuNSN]ZrCl2 to yield
decomposition products. [t-BuNSN]ZrMe2 is stable in
solution at -30 °C for hours, but it begins to decompose
after 30-40 min at 0 °C. Samples are stable in the solid
state at -30 °C, but after about 1 h at room temperature
some decomposition is observed. Because of the thermal
instability of [t-BuNSN]ZrMe2 and our inability to
separate it entirely from decomposition products, the
glassware employed during workup had to be precooled
to -30 °C in order to obtain pure samples. Attempts to
make other alkyl complexes (ethyl, isobutyl, neopentyl,
and benzyl) from [t-BuNSN]ZrCl2 and Grignard and/or
lithium reagents in diethyl ether, THF, or diethyl ether/
THF mixtures did not produce any identifiable species
that were stable for more than ∼0.5 h at -30 °C.
However, we cannot confirm that [t-BuNSN]ZrR2 spe-
cies actually were prepared before decomposition en-
sued; decomposition in the process of preparing the
dialkyl species is also a possibility.
[i-PrNSN]ZrMe2 can be synthesized cleanly by treat-
ing [i-PrNSN]ZrCl2 with MeMgCl or MeMgI in diethyl
ether. In contrast to [t-BuNSN]ZrMe2, [i-PrNSN]ZrMe2
is stable in hydrocarbon solution at room temperature
for many (>10) hours and solutions even can be heated
to 90 °C for 10 min without any observable decomposi-
tion (according to 1H NMR). Surprisingly, the synthesis
of [i-PrNSN]Zr(CH2CHMe2)2 was also straightforward.
Syn th eses of Zir con iu m Com p lexes th a t Con ta in
[t-Bu NSN]2- or [i-P r NSN]2-. The reaction between H2-
[t-BuNSN] and Zr(NMe2)4 proceeds cleanly to give
[t-BuNSN]Zr(NMe2)2 as colorless crystals (eq 2). At room
H2[RNSN] + Zr(NMe2)4 -2Me NH8
2
[RNSN]Zr(NMe2)2 2TMSCl8 [RNSN]ZrCl2 (2)
R ) t-Bu, i-Pr
(8) Micovic, I. V.; Ivanovic, M. D.; Piatak, D. M.; Bojic, V. D.
Synthesis 1991, 11, 1043.
(9) Baumann, R. Ph.D. Thesis, Massachusetts Institute of Technol-
ogy, 1998.
temperature the reaction is slow (<50% conversion in

Zr Complexes Containing Tridentate Amido Ligands
Organometallics, Vol. 18, No. 5, 1999 845
a
Ta ble 1. Cr ysta l Da ta a n d Str u ctu r e Refin em en t Deta ils for [t-Bu NSN]Zr Cl(NMe2) a n d [i-P r NSN]Zr Me2
[t-BuNSN]ZrCl(NMe2) [i-PrNSN]ZrMe2
empirical formula
fw
C22H32ClN3SZrb
C20H28N2SZr
497.24
419.72
temp (K)
crystal syst
space group
a (Å)
183(2)
173(2)
triclinic
orthorhombic
P1h
Pna21
9.2816(5)
15.4215(2)
b (Å)
9.8157(6)
15.4052(3)
c (Å)
14.0988(8)
8.3754(2)
R (deg)
85.7090(10)
90
â (deg)
79.3330(10)
90
γ (deg)
68.5410(10)
90
V (Å3), Z
calcd density (Mg/m3)
1174.74(12), 2
1989.75(7), 4
1.406
1.401
abs coeff (mm-1
F000
cryst size (mm)
)
0.683
0.661
516
872
0.18 × 0.18 × 0.12
0.28 × 0.28 × 0.28
θ range for data collecn (deg)
limiting indices
1.47-23.25
1.87-23.26
-7 e h e 10, -10 e k e 10, -11 e l e 15
4906
-17 e h e 17, -9 e k e 17, -8 e l e 9
7711
no. of reflns collected
no. of indep rflns
3320 (Rint ) 0.0356)
2816 (Rint ) 0.0286)
no. of data/restraints/ params
goodness of fit on F
3319/0/254
2816/1/218
1.206
0.945
final R indices (I > 2σ(I))
R1 ) 0.0372, wR2 ) 0.0988
R1 ) 0.0405, wR2 ) 0.1027
0.0010(10)
R1 ) 0.0205, wR2 ) 0.0551
R1 ) 0.0218, wR2 ) 0.0567
0.0015(4)
R indices (all data)
extinction coeff
largest diff peak and hole (e Å-3
)
0.479 and -0.353
0.306 and -0.307
a
All data were collected on a SMART CCD diffractometer with Mo KR radiation (λ ) 0.710 73 Å). Data were refined using full-matrix
least squares on F2. No absorption correction was applied. In this formula we assume D ) H.
b
[i-PrNSN]Zr(CH2CHMe2)2 is stable in solution at room
temperature for several hours and can be heated to 80
°C for a few minutes before a small degree of decompo-
1
sition is detected (by H NMR). Interestingly, we have
not been able to prepare [i-PrNSN]Zr(CH2CH3)2 by
analogous methods; presumably it is markedly less
stable than [i-PrNSN]Zr(CH2CHMe2)2, since the â pro-
tons in an ethyl complex are relatively accessible and
subject to more rapid â abstraction.10
A single-crystal X-ray study of [t-BuNSN]ZrCl(NMe2)
(Figure 1; Tables 1 and 2) shows that it is approximately
a trigonal bipyramid with the chloride and sulfur atoms
in the axial positions (Cl-Zr-S angle 178.86(4)°), a fac
geometry. In the ideal fac form, the angle between one
N-Zr-S plane and the other would be ∼120°. In
contrast, in an ideal mer form, the angle between one
N-Zr-S plane and the other would be 180°. (As shown
in eq 3, for a dialkyl complex fac and mer forms can be
interconverted relatively easily by varying the N-Zr-N
angle and rotating the MR2 unit in the MR2S plane.) In
[t-BuNSN]ZrCl(NMe2) the angle between the N(1)-
Zr-S and N(2)-Zr-S planes is 117°. The “equatorial”
angles at Zr sum to 343.17° and are not equal, with the
largest (N(1)-Zr-N(3) ) 124.08(13)°) being approxi-
mately 20° larger than the other two. The ligand
backbone also is twisted (in Figure 1b, the twist is
clockwise if viewed down the Cl-Zr-S vector); i.e., the
tert-butyl group that contains C(201) points toward N(1)
F igu r e 1. (a, top) ORTEP drawing of the structure of
[t-BuNSN]ZrCl(NMe2): N(3)-C(302) ) 1.406(6) Å, N(3)-
C(301) ) 1.435(6) Å, Zr-C(301) ) 2.849(5) Å, Zr-N(3)-
C(302) ) 136.4(3)°, Zr-N(3)-C(301) ) 107.9(3)°. (b, bot-
tom) Side view of a Chem 3D drawing of the structure of
[t-BuNSN]ZrCl(NMe2).
and the tert-butyl group that contains C(101) points
toward N(3). Steric interaction between the tert-butyl
group that contains C(101) and the NMe2 ligand is likely
(10) Negishi, E.; Takahashi, T. Acc. Chem. Res. 1994, 27, 124.

846 Organometallics, Vol. 18, No. 5, 1999
Graf et al.
Ta ble 2. Selected Bon d Len gth s (Å) a n d An gles (d eg) in [t-Bu NSN]Zr Cl(NMe2), [i-P r NSN]Zr Me2, a n d
[t-Bu NON]Zr Me[MeB(C6F 5)3]
[t-BuNSN]ZrCl(NMe2)
[i-PrNSN]ZrMe2
[t-BuNON]ZrMe[MeB(C6F5)3]a
Zr-S
2.7273(10)
2.102(3)
2.098(3)
2.4555(11)
2.059(3)
Zr-S
2.7343(9)
2.088(2)
2.133(2)
2.245(3)
2.250(3)
Zr-O
2.256
Zr-N(1)
Zr-N(2)
Zr-Cl(1)
Zr-N(3)
Zr-N(1)
Zr-N(2)
Zr-C(1)
Zr-C(2)
Zr-N(1)
Zr-N(2)
Zr-C(1)
Zr-C(2)
2.049(10)
2.065(10)
2.487(12)b
2.200(13)
N(1)-Zr-N(2)
N(1)-Zr-S
N(1)-Zr-Cl(1)
N(1)-Zr-N(3)
N(2)-Zr-S
109.91(12)
72.22(9)
107.04(9)
124.08(13)
75.12(9)
N(1)-Zr-N(2)
N(1)-Zr-S
N(1)-Zr-C(1)
N(1)-Zr-C(2)
N(2)-Zr-S
135.66(8)
72.20(6)
102.67(10)
99.64(10)
71.26(7)
N(1)-Zr-N(2)
N(1)-Zr-O
N(1)-Zr-C(1)
N(1)-Zr-C(2)
N(2)-Zr-O
113.8(4)
73.5(4)
118.6(5)
113.8(5)
75.2(3)
N(2)-Zr-Cl(1)
N(2)-Zr-N(3)
N(3)-Zr-S
106.00(9)
109.18(12)
81.10(9)
98.70(10)
178.86(4)
106.1(2)
N(2)-Zr-C(1)
N(2)-Zr-C(2)
C(2)-Zr-S
112.71(10)
89.04(11)
135.06(10)
115.11(12)
109.77(8)
103.51(13)
104.1(2)
N(2)-Zr-C(1)
N(2)-Zr-C(2)
C(2)-Zr-O
101.2(5)
115.4(5)
79.2(4)
N(3)-Zr-Cl(1)
Cl(1)-Zr-S
C(2)-Zr-C(1)
C(1)-Zr-S
C(2)-Zr-C(1)
C(1)-Zr-O
91.8(5)
167.3(4)
116.4(11)
129.3(8)
125.8(7)
C(7)-S-C(1)
Zr-N(1)-C(101)
Zr-N(2)-C(201)
C(21)-S-C(11)
Zr-N(1)-C(17)
Zr-N(2)-C(27)
C-O-C
Zr-N(2)-C(27)
Zr-N(1)-C(17)
122.9(2)
123.7(2)
121.3(2)
N(1)/Zr/S/N(2)c,d
S-Zr-N(1)-C(1)d
S-Zr-N(2)-C(2)d
117
140.3
134.9
N(1)/Zr/S/N(2)c,d
154
170.9
164.7
N(1)/Zr/O/N(2)c,d
O-Zr-N(1)-C(1)d
O-Zr-N(1)-C(1)d
121
140.3
129.3
S-Zr-N(1)-C(17)d
S-Zr-N(2)-C(27)d
a
b
d
See ref 1. Methyl group bridging between Zr and B. c The external angle between the planes. Obtained from Chem 3D model.
to give rise to the relatively large N(1)-Zr-N(3) angle
and also to the C(301)-N-C(302) plane being twisted
approximately halfway between an orientation parallel
to the N(1)/N(2)/N(3) plane and one perpendicular to the
N(1)/N(2)/N(3) plane (C(302)-N(3)-Zr-S ) 48°). (The
closest H‚‚‚H contact between a proton attached to
C(302) and one attached to C(102) in the structure is
∼2.4 Å, whereas if the C(301)-N-C(302) plane is
perpendicular to the Cl-Zr-S axis, with no other
changes, the closest H‚‚‚H contact is ∼1.9 Å, according
to a Chem 3D model.) There is also evidence for a weak
agostic11 Zr-C(301) interaction (2.849 Å), according to
bond angles and distances in the Zr-NMe2 group, in
particular Zr-C(301) ) 2.849(5) Å and Zr-N(3)-C(301)
) 107.9(3)°. A weak agostic-like interaction may be
possible as a consequence of steric hindrance that
already distorts the dimethylamido group in the ob-
served direction. The Zr-N bond lengths are similar to
those expected for Zr-amido complexes (2.05-2.12
Å),1,4,7,12-14 and the sum of the angles at each amido
nitrogen atom is 360° (within 3σ). The zirconium-sulfur
bond length (2.7273(10) Å) is similar to that (2.730(4)
Å) reported for [Cp*2ZrMe(THT)]+[BPh4]- (THT ) tet-
rahydrothiophene)15 or that (2.8046(9) Å) reported for
[(2,6-Me2C6H3NCH2CH2)2S]ZrMe2.7 Finally, the angle
at sulfur (106.1(2)°) is slightly larger than the corre-
sponding angle in titanium complexes that contain the
substituted [(O-o-C6H4)2S]2- ligand16-20 (e.g., 100.8(2)°
in [(O-2,4-Me2-o-C6H4)2S]Ti(O-i-Pr)219) but approxi-
mately the same as the angle found in [(2,6-Me2C6H3-
NCH2CH2)2S]ZrMe2 (105.0(2)°).7
A single-crystal X-ray study of [i-PrNSN]ZrMe2 (Fig-
ure 2; Tables 1 and 2) shows it to have a structure that
is approximately halfway between a fac and an ideal
mer geometry, as judged by the N(1)/Zr/S/N(2) value
(angle between the N(1)-Zr-S and N(2)-Zr-S planes)
of 154°. The view in Figure 2b emphasizes the ap-
proximately square-pyramidal core in which N(1)-Zr-
N(2) and C(2)-Zr-S are equal (135°) and the angles
between C(1) and the four “equatorial” atoms vary from
102.67° (N(1)-Zr-C(1)) to 115.11° (C(2)-Zr-C(1)). The
S-Zr-N-C dihedral angles are relatively large com-
pared to what they are in [t-BuNSN]ZrCl(NMe2). How-
ever, the C-S-C angles are similar in the two com-
plexes, as are the Zr-S and Zr-Namido bond distances.
A curious feature of the structure of [i-PrNSN]ZrMe2 is
the fact that although the angles around N(1) sum to
360.0° (Zr-N(1)-C(17) ) 104.1° Zr-N(1)-C(16) )
132.9°, and C(17)-N(1)-C(16) ) 123.0°), as is the case
in all diamido/donor complexes we have examined to
date, the angles around N(2) sum to 346.6° (Zr-N(2)-
C(27) ) 121.3°, Zr-N(2)-C(26) ) 108.1°, and C(27)-
N(1)-C(26) ) 117.2°); i.e., N(2) is markedly distorted
from planarity. The slightly longer Zr-N(2) bond dis-
tance (2.088(2) Å) may be a reflection of a somewhat
weaker π bond between Zr and N(2) as a consequence
of the nonplanarity of N(2). Consistent with a weaker
Zr-N(2) π bond is the relatively large difference be-
tween the two Zr-Namido bond lengths in [i-PrNSN]-
ZrMe2 (0.045 Å) compared to the difference in [t-BuNSN]-
ZrCl(NMe2) (0.004 Å).
(11) Brookhart, M.; Green, M. L. H.; Wong, L. Prog. Inorg. Chem.
1988, 36, 1.
(12) Gue´rin, F.; McConville, D. H.; Vittal, J . J . Organometallics
1996, 15, 5586.
(13) Aoyagi, K.; Gantzel, P. K.; Kalai, K.; Tilley, T. D. Organome-
tallics 1996, 15, 923.
(14) Schrock, R. R.; Seidel, S. W.; Schrodi, Y.; Davis, W. M.
Organometallics 1999, 18, 428.
(15) Eshuis, J . J .; Tan, Y. Y.; Meetsma, A.; Teuben, J . H.; Renkema,
J .; Evens, G. G. Organometallics 1992, 11, 362.
(16) Miyatake, T.; Mizunuma, K.; Seki, Y.; Kakugo, M. Makromol.
Chem. Rapid Commun. 1989, 10, 349.
Solution proton and carbon NMR spectra of [t-BuN-
SN]Zr(NMe2)2 and [t-BuNSN]ZrMe2 suggest that struc-
tures which contain two inequivalent NMe2 and methyl
(17) Miyatake, T.; Mizunuma, K.; Kakugo, M. Makromol. Chem.
Macromol. Symp. 1993, 66, 203.
(18) van der Linden, A.; Schaverien, C. J .; Meijboom, N.; Ganter,
C.; Orpen, A. G. J . Am. Chem. Soc. 1995, 117, 3008.
(19) Fokken, S.; Spaniol, T. P.; Kang, H.-C.; Massa, W.; Okuda, J .
Organometallics 1996, 15, 5069.
(20) Porri, L.; Ripa, A.; Colombo, P.; Miano, E.; Capelli, S.; Meille,
S. V. J . Organomet. Chem. 1996, 514, 213.

Zr Complexes Containing Tridentate Amido Ligands
Organometallics, Vol. 18, No. 5, 1999 847
Ta ble 3. Va r ia ble-Tem p er a tu r e 1H NMR Stu d ies of
F lu xion a l [RNSN]2- Com p lexes
q
∆ν
Tc
∆Gc
((2 Hz)a resonance ((2 K) (kcal/mol)b
[t-BuNSN]Zr(NMe2)2
[t-BuNSN]ZrMe2
152
162
20
100
79
NMe2
t-Bu
Zr-Me
t-Bu
NMe2
CHMe2
Zr-Me
CHMe2
CH2CHMe2 323
CHMe2 319
259
223
243
208
342
337
319
320
12.1
10.3c
12.3
9.8c
16.6
16.7
14.9
14.9
15.3
14.9
[i-PrNSN]Zr(NMe2)2
[i-PrNSN]ZrMe2
52
180
205
134
190
[i-PrNSN]Zr(CH2CHMe2)2
a
b
Estimated from resolved spectra at lower temperature. Cal-
culated using the estimated value of ∆ν. Error (σ) estimated to be
(0.1. c Ascribed to a freezing out of the facC1 conformation.
a single methyl resonance, and four aromatic backbone
resonances are observed, consistent with a mer struc-
ture on the NMR time scale, either the ideal mer
structure (with C2v symmetry) or the more likely
“twisted” form (with C2 symmetry) in which the tert-
butyl carbon atoms do not lie in the MN2S plane. As
the sample is cooled, the average resonance for the two
methyl groups on the metal becomes two at approxi-
mately -30 °C. When the sample of [t-BuNSN]ZrMe2
is cooled further, the two tert-butyl groups become
inequivalent and yield two broad resonances, while
more than four aromatic proton resonances can be seen.
In the -78 °C spectrum only one of the two broad tert-
butyl resonances is visible at 1.4 ppm; the broader of
the two tert-butyl resonances is buried under the
resonances for the two methyl groups near 1 ppm. These
data are consistent with a “twisted” fac (facC1) form
analogous to what is found in the solid state for
[t-BuNSN]ZrCl(NMe2) at the lowest temperature. The
fact that the tert-butyl resonances are broad, and one
is much broader than the other, suggests that rotation
of the tert-butyl groups is being restricted in the facC1
structure, and more so in one tert-butyl group than the
other. The two tert-butyl groups and all eight aryl
protons would be inequivalent in a facC1 structure. All
eight aryl proton resonances cannot be resolved, al-
though more than four different kinds of aryl proton
F igu r e 2. (a, top) ORTEP drawing of the structure of
[i-PrNSN]ZrMe2. (b, bottom) Side view of a Chem 3D
drawing of the structure of [i-PrNSN]ZrMe2.
groups, respectively, presumably “twisted” fac struc-
tures analogous to the structure of [t-BuNSN]ZrCl-
(NMe2), can be observed at low temperatures and that
the inequivalent NMe2 or Me groups interconvert
rapidly on the NMR time scale at higher temperatures.
1
For example, in the H NMR spectrum of [t-BuNSN]-
ZrMe2 (Figure 3) at 0 °C a single tert-butyl resonance,
F igu r e 3. Variable-temperature proton NMR (300 MHz) of [t-BuNSN]ZrMe2 in toluene-d8 (*).

848 Organometallics, Vol. 18, No. 5, 1999
Graf et al.
inversion at sulfur in the [i-PrNSN]2- structures vary
between 15 and 17 kcal/mol, a few kcal/mol higher than
q
the ∆Gc values for inversion at sulfur in the [t-Bu-
q
NON]2- structures. The fact that the ∆Gc values for
inversion at sulfur in the [t-BuNSN]2- and [i-PrNSN]2-
complexes do not differ greatly suggests that inversion
is the rate-limiting process in each case and that steric
differences between tert-butyl and isopropyl groups have
only a small effect.
Cation ic Zir con iu m Com plexes Con tain in g [t-Bu -
NSN]2- a n d [i-P r NSN]2-. Reaction of [t-BuNSN]ZrMe2
and 1 equiv of [Ph3C][B(C6F5)4] in chlorobenzene-d5 at
-30 °C leads to a color change from red-orange to
yellow-orange. The proton and carbon NMR spectra of
the resulting species in solution are indicative of clean
methyl abstraction to give Ph3CMe and “{[t-BuNSN]-
ZrMe}{B(C6F5)4}”. In “{[t-BuNSN]ZrMe}{B(C6F5)4}” are
found a Zr-Me resonance at 1.09 ppm and a broadened
resonance for the 13C-labeled methyl group at 52.78 ppm
(cf. 0.72 and 53.60 ppm in “{[t-BuNON]ZrMe}{B(C6F5)4}”
in C6D5Br9). In chlorobenzene, the cation decomposes
within minutes at room temperature but it is relatively
stable at lower temperatures (∼1 h at 0 °C; 2-3 h at
-10 °C). The stability of the cation in bromobenzene is
lower, so that some decomposition is observed within
30 min at 0 °C. We also attempted to synthesize the
base adduct {[t-BuNSN]ZrMe(PhNMe2)}{B(C6F5)4} in
the hope of making a more stable cation. For example,
1 or 2 equiv of PhNMe2 was added to a -30 °C solution
of “{[t-BuNSN]ZrMe}{B(C6F5)4}” (prepared by methyl
abstraction with [Ph3C]+). The desired base adduct
appeared to form initially, but complete decomposition
was observed within 3 h at -30 °C. In contrast, addition
of THF (3 equiv) to {[t-BuNSN]ZrMe(PhNMe2)}-
{B(C6F5)4} led to clean formation of {[t-BuNSN]ZrMe-
(THF)x}{B(C6F5)4} that has no symmetry on the NMR
time scale at room temperature. The THF adduct is
stable at room temperature for 10-15 min (and for
hours at lower temperatures) before it begins to decom-
pose.
F igu r e 4. Variable-temperature proton NMR (500 MHz)
of [i-PrNSN]ZrMe2 in toluene-d8 (*).
resonances clearly are present in the spectrum at -78
°C. The variable-temperature 13C NMR spectrum of
[t-BuNSN]Zr(13CH3)2 showed 13C resonances at 45.89
and 44.50 ppm at -30 °C that coalesced at -10 °C.
Similar spectra are observed for [t-BuNSN]Zr(NMe2)2,
and activation barriers for two different processes can
be observed. The barriers for each of these processes in
[t-BuNSN]Zr(NMe2)2 and [t-BuNSN]ZrMe2 are listed in
Table 3. The process in which fac and mer forms
interconvert has a value for ∆Gcq of ∼12 kcal/mol at the
q
coalescence temperature, while ∆Gc for the twisting
process within the fac form is ∼10 kcal/mol for both
[t-BuNSN]Zr(NMe2)2 and [t-BuNSN]ZrMe2. It cannot be
determined if the mer transition state has C2v or C2
symmetry, but in either case the sulfur must actually
invert through a planar form.
Attempts to prepare cationic complexes from [i-
PrNSN]ZrMe2 and [i-PrNSN]Zr(CH2CHMe2)2 were analo-
gous to methods used to prepare cations from [t-BuNSN]-
ZrMe2. Reaction of equimolar amounts of [i-PrNSN]ZrMe2
(or [i-PrNSN]Zr(CH2CHMe2)2) and [Ph3C][B(C6F5)4] in
chlorobenzene-d5 at -30 °C led to decomposition (ac-
The NMR spectrum of [i-PrNSN]ZrMe2 at -40 °C
(Figure 4) reveals resonances for two inequivalent
methyl groups on Zr at 0.32 and 0.67 ppm and reso-
nances for two types of isopropyl methyl groups at 1.27
and 1.64 ppm, consistent with the square-pyramidal
structure found in the solid state. On the basis of the
fact that only four types of aryl proton resonances and
one type of methine resonance are observed at -78 °C,
the square-pyramidal structure apparently has a plane
of symmetry on the NMR time scale at that tempera-
ture. However, the isopropyl methyl resonances at 1.27
and 1.64 ppm are not sharp doublets, which could be
taken as evidence that a lower symmetry (e.g., twisted)
form is actually observable on the NMR time scale in
this case also but that differences in chemical shifts are
relatively small. What is clear, however, is that upon
warming the sample, the axial and equatorial methyl
groups interconvert and the isopropyl methyl groups
interconvert, essentially at the same rate (∼15 kcal/mol;
Figure 4 and Table 3). Similar spectra are observed for
[i-PrNSN]Zr(NMe2)2 and [i-PrNSN]Zr(CH2CHMe2)2. All
data are consistent with inversion at sulfur via a mer
1
cording to H NMR spectra) along with a color change
to deep red. Formation of cations via protonation
reactions was only slightly more successful. [PhNHMe2]-
[B(C6F5)4] was added to a -30 °C solution of [i-PrNSN]-
ZrMe2, and when the solution was warmed to -10 °C
for 45 min or ∼23 °C for 15 min, the color changed from
pale yellow to orange. The 1H NMR of the reactions
under various conditions was indicative of at least two
main species (A and B, see Experimental Section)
having been formed, as evidenced by two methine
resonances at 4.0-3.0 ppm and two pairs of isopropyl
methyl resonances at 1.6-0.6 ppm. A bromobenzene-
d5 solution containing an equimolar mixture of [Ph-
NHMe2][B(C6F5)4] and [i-PrNSN]ZrMe2 was warmed
from ∼-30 °C to ∼23 °C as the reaction was monitored
by 1H NMR. Protonation of the methyl group and
production of methane was observed as species A
(∼80%) and species B (∼20%) were formed initially.
q
form. As shown in Table 3, the values for ∆Gc for

Zr Complexes Containing Tridentate Amido Ligands
Organometallics, Vol. 18, No. 5, 1999 849
Over the next 30-40 min, species B increased in
concentration at the expense of A and the solution
turned dark red. Appreciable decomposition was ob-
served within 60 min, with complete decomposition
occurring in ∼2 h. We believe that A is the desired
cation, {[i-PrNSN]ZrMe(PhNMe2)}{B(C6F5)4}; the iden-
tity of species B, into which A apparently is trans-
formed, is unknown. Reaction of [PhNHMe2][B(C6F5)4]
and [i-PrNSN]ZrMe2 in the presence of an extra 1 equiv
of PhNMe2 does not further stabilize the cation; decom-
position is observed within 20 min at room temperature.
Attempts were also made to trap a clean base adduct
of species A by performing the reaction in the presence
of 3-4 equiv of THF. The 1H NMR of the resulting
reaction was indicative of a mixture of THF-solvated
complexes. Finally, reaction of [PhNHMe2][B(C6F5)4]
and [i-PrNSN]Zr(CH2CHMe2)2 under a variety of reac-
tion conditions gave a complex mixture of products (1H
NMR).
When 1-hexene was added to “{[t-BuNSN]ZrMe}-
{B(C6F5)4}” (generated in chlorobenzene at -10 °C from
[t-BuNSN]ZrMe2 and [Ph3C][B(C6F5)4]), oligomerization
of 1-hexene was observed, but no polymerization to any
significant extent. (Approximately 20 equiv of 1-hexene
was consumed within the lifetime (∼3 h) of the initiator
at -10 °C.) It is not clear whether this low degree of
polymerization is due to inherently slow propagation or
to a thermal instability of the cationic initiator and the
intermediate cation that contains the growing polymer
chain. This behavior differs markedly from that found
for “{[t-BuNON]ZrMe}{B(C6F5)4}”, which is a living
catalyst for the polymerization of 1-hexene at 0 °C in
chlorobenzene.2
angle, compared to [t-BuNON]2- complexes. Smaller
Zr-N-C angles in [t-BuNSN]ZrCl(NMe2) (by ∼5°)
would seem likely to exacerbate steric crowding near
the tert-butyl groups.
In contrast, the structures of [i-PrNSN]ZrMe2 and
[i-PrNON]ZrMe223 are significantly different. Compound
[i-PrNON]ZrMe2 has a “twisted” mer (C2) structure with
an “S-shaped” backbone. The “equatorial” methyl groups
therefore are interconverted in a C2 operation. The
oxygen donor is strictly planar. In [i-PrNSN]ZrMe2 the
analogous mer form is higher in energy than the
observed square-planar form as a consequence of the
“fixed” C-S-C angle and the reluctance of sulfur to
adopt sp2 hybridization and therefore the required
trigonal-planar geometry at sulfur. We believe that the
structure of [i-PrNSN]ZrMe2 is not analogous to that
of [t-BuNSN]ZrMe2 as a consequence of a lower degree
of steric interaction between the i-Pr groups and the
methyl groups than between the t-Bu groups and the
methyl groups, and that t-Bu/methyl interactions in the
case of [t-BuNSN]ZrMe2 provides additional impetus for
forming a facC1 geometry. The ground-state structure
of [i-PrNSN]ZrMe2 therefore is approximately a square
pyramid, a structure approximately halfway between
a fac and a mer form.
The difference in the thermal stabilities of [t-BuNSN]-
ZrMe2 and [i-PrNSN]ZrR2 species (R ) Me, CH2CHMe2)
is striking. We hypothesize that five-coordinate com-
plexes that contain the [t-BuNSN]2- ligand are sterically
more crowded in subtle but significant ways than
complexes that contain the [i-PrNSN]2- ligand and that
steric crowding leads to decomposition of [t-BuNSN]2-
complexes. A decomposition pathway unique to a tert-
butyl-substituted neutral complex would be loss of a tert-
butyl radical, a process that has been documented in
diamido/phosphorus donor systems recently.14 Cleavage
of an aryl-sulfur bond in either a [t-BuNSN]2- or a
[i-PrNSN]2- complex also seems plausible in view of the
recently documented cleavage of an aryl-oxygen bond
in a proposed [i-PrNON]TiII intermediate.9 It is also
striking that many more [t-BuNON]ZrR2 complexes
(including R ) CH2CHMe2) can be prepared than
[t-BuNSN]ZrR2 complexes, although there appear to be
steric limitations in the [t-BuNON]ZrR2 system also
(e.g., R apparently cannot be CH2CMe3).9 Although it
is tempting to propose that some modes of decomposi-
tion might be faster as a consequence of sulfur being
more electron donating (in a σ sense) than oxygen, there
are no quantitative or even semiquantitative data that
result from this work that would support such an
assertion.
Discu ssion
The structure of [t-BuNSN]ZrCl(NMe2) is similar to
that of [t-BuNON]ZrMe[MeB(C6F5)3], although there are
some notable, but understandable, differences (Table
2).1 The Zr-S bond is nearly 0.5 Å longer than the Zr-O
bond in [t-BuNON]ZrMe[MeB(C6F5)3], a difference that
is expected on the difference in the van der Waals radii21
of an oxygen in a typical ether (1.4-1.6 Å) and a sulfur
in a typical thioether (1.8-1.9 Å). The external angle
between the N(1)-S-Zr and N(2)-S-Zr planes (117°)
in [t-BuNSN]ZrCl(NMe2) is ∼4° less than the corre-
sponding angle in [t-BuNON]ZrMe[MeB(C6F5)3] (121°).
The Cipso-S-Cipso angle (106.1(2)°) is ∼10° smaller in
[t-BuNSN]ZrCl(NMe2) than the Cipso-O-Cipso angle in
[t-BuNON]ZrMe[MeB(C6F5)3] (116.4(9)°), which is a
typical difference in an angle at sulfur versus oxygen
(e.g., H-S-H ) 92.1° and H-O-H ) 104.5° in the gas
phase22). Finally, the N-M-N angle (109.91(12)°) is ∼4°
smaller in [t-BuNSN]ZrCl(NMe2) than in [t-BuNON]-
ZrMe[MeB(C6F5)3].1 The X-M-N(1)-C(1) dihedral
angles (X ) O, S) are approximately the same in
[t-BuNSN]ZrCl(NMe2) and [t-BuNON]ZrMe[MeB(C6F5)3],
although the M-N-C angles appear to be several
The finding that activated [t-BuNON]ZrMe2 is an
initiator for the living polymerization of 1-hexene
contrasts with the finding that activated [i-PrNON]-
ZrMe2 will only oligomerize 1-hexene.24 Nevertheless,
both initiators appear to be stable at room temperature.
The fact that neither [t-BuNSN]ZrMe2 nor [i-PrNSN]-
ZrMe2 can be activated to yield stable initiators that
will oligomerize or polymerize 1-hexene under analogous
conditions raises suspicion that modes of decomposition
might be operative that can be traced directly to sulfur.
degrees smaller in [t-BuNSN]ZrCl(NMe2). The Cipso
-
S-Cipso angle appears to be restricted, which leads to
some decrease in the external angle between the N(1)-
S-Zr and N(2)-S-Zr planes and the N(1)-M-N(2)
(21) Bondi, A. J . Phys. Chem. 1964, 68, 441.
(22) Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements;
Pergamon: Oxford, U.K., 1984.
(23) Baumann, R.; Davis, W. M.; Schrock, R. R., submitted for
publication.
(24) Liang, L.-C.; Schrock, R. R., unpublished results.

850 Organometallics, Vol. 18, No. 5, 1999
Graf et al.
from Exxon Chemical Corp. Other reagents were purchased
from commercial sources and used as received.
However, we cannot say conclusively that this is the
case. It should be noted that cations prepared by
activation of [(ArylNCH2CH2)2S]ZrMe2 also did not
appear to be as stable as those prepared by activation
of [(ArylNCH2CH2)2O]ZrMe2,7 although the conse-
quences of any decomposition reaction again could not
be documented. Finally, we note that titanium com-
plexes that contain the [(O-o-C6H4)2S]2- ligand16-20 have
been prepared and evaluated as precursors to polym-
erization catalysts, but no well-behaved polymerization
activity has yet been reported.
X-ray-quality crystals of [t-BuNSN]ZrCl(NMe2) were grown
by cooling a concentrated diethyl ether solution. Crystals of
[i-PrNSN]ZrMe2 were obtained by cooling a diethyl ether/
benzene solution. X-ray data were collected on a SMART/CCD
diffractometer as described elsewhere.27 (See Supporting In-
formation for full details.)
S(o-C6H4NO2)2. To a vigorously stirred suspension of K2-
CO3 (71 g, 0.51 mol) in 500 mL of degassed DMSO were added
o-C6H4(NO2)SH28 (40 g, 0.26 mol) and o-C6H4(NO2)F (36 g, 0.26
mol). After 18 h of stirring at room temperature, the solution
was poured into 1.5 L of stirred ice/water and the yellow
precipitate was collected by filtration. The precipitate was
dried in vacuo at 60 °C for 12 h to give 72 g (100%) of S(o-
C6H4NO2)2 as a yellow solid. The crude product was >95% pure
(according to its 1H NMR spectrum) and was used without
further purification: 1H NMR (CDCl3) δ 8.09 (dd, 2), 7.48 (m,
4), 7.28 (d, 2); 13C NMR (CDCl3) δ 149.9, 134.2, 134.1, 132.1,
129.1, 126.0.
S(o-C6H4NH2)2. A 1 L hydrogenation vessel was charged
with 10% Pd on carbon (3.5 g, 3.3 mmol of Pd), S(o-C6H4NO2)2
(70 g, 0.25 mol), and 600 mL of degassed 95% ethanol. The
hydrogen vessel was purged three times with hydrogen gas,
the pressure was adjusted to 60 psi, and the contents were
heated to 80 °C for 48 h. The reaction mixture was cooled to
∼40 °C, and the suspension was filtered through a bed of
Celite. The filtrate was concentrated in vacuo to about 200
mL and poured into 1.5 L of ice water. The white precipitate
was collected by filtration, washed twice with 50 mL of water,
and dried on a frit for ∼30 min. The solid was dissolved in
dichloromethane, and the resulting solution was dried over
MgSO4. The mixture was filtered, and the filtrate was con-
centrated to about 75 mL. Hexane (500 mL) was added to the
suspension, and the off-white solid was filtered off and washed
twice with 10 mL of hexane. Concentration of the filtrate in
vacuo yielded a second crop of solid: yield 49 g (91%); 1H NMR
(CDCl3) δ 7.24 (dd, 2), 7.10 (td, 2), 6.71 (m, 4), 4.20 (br s, 4,
NH2); 13C NMR (CDCl3) δ 147.2, 133.9, 129.7, 119.5, 117.7,
116.0.
Con clu sion s
The thioethers (t-Bu-d6-NH-o-C6H4)2S and (i-PrNH-
o-C6H4)2S can be prepared readily and employed to
prepare some zirconium dialkyl complexes via [RNSN]-
Zr(NMe2)2 and [RNSN]ZrCl2 intermediates. [i-PrNSN]-
ZrR2 (R ) Me, CH2CHCMe2) complexes are significantly
more stable than [t-BuNSN]ZrMe2, perhaps because of
some exacerbated steric hindrance and concomitant
weakness of the t-Bu-N bond or S-Cipso bond in such
circumstances. Inversion at sulfur in five-coordinate
dialkyl complexes is significantly slower than inversion
at oxygen in analogous complexes that contain the
[t-BuNON]2- ligand. Cationic complexes prepared from
[RNSN]ZrMe2 precursors were neither stable at 22 °C
nor active initiators for the controlled polymerization
of 1-hexene.
Exp er im en ta l Section
Gen er a l Con sid er a tion s. Unless noted otherwise, all
experiments were conducted under nitrogen in a Vacuum
Atmospheres drybox or by using standard Schlenk techniques.
Nitrogen (for Schlenk work) and ethylene gases were purified
by passage through an activated, zeolite-supported MnO
catalyst (BOC Gases). Toluene was distilled from sodium/
benzophenone. Pentane, diethyl ether, and tetrahydrofuran
were deoxygenated and dried by first sparging with nitrogen
and then passing through either (a) two large activated
alumina columns (diethyl ether and THF) or (b) one large
activated alumina column and then one large column of
activated Q5 (pentane).25 Dichloromethane and chlorobenzene
(Aldrich HPLC grade, 99.9%) were distilled from CaH2 under
nitrogen. Solvents were stored in the drybox over activated 4
Å sieves. 1-Hexene was refluxed over sodium for 4 days,
distilled, and stored over activated 4 Å sieves. Molecular sieves
and Celite were activated in vacuo (10-3 Torr) for 24 h at 175
and 125 °C, respectively.
S[o-C6H4NHC(CD3)2CH3]2 (H2[t-Bu NSN]). S(o-C6H4NH2)2
(30.0 g, 0.139 mol) was dissolved in acetone-d6 (90 g, 1.4 mo),
and 50 g of activated 4 Å sieves was added. After the diimine
had completely formed (in about 10 days as judged by 1H
NMR), the acetone-d6 was decanted off. The sieves were soaked
in 100 mL of diethyl ether, separated by filtration, and washed
twice with 20 mL of diethyl ether. The acetone and ether
solutions were combined and the volatile components removed
in vacuo to give the diimine as a viscous red oil. A solution of
the diimine in 100 mL of diethyl ether was added dropwise to
300 mL of MeLi (1.4 M in diethyl ether) at -78 °C. After 2 h,
the yellow solution was warmed to room temperature and
stirred for 16 h. The mixture was transferred by cannula into
500 mL of vigorously stirred ice/water. The organic layer was
separated, and the aqueous layer was extracted twice with 150
mL of diethyl ether. The organic layer was dried over MgSO4,
and the solvents were removed in vacuo. The red viscous oil
was dissolved in a minimum amount of pentane. Chromatog-
raphy of the crude product on an alumina column (40 × 4.5
cm) provided H2[t-BuNSN] after elution with 500 mL of
pentane followed by 3:1 pentane/diethyl ether. Recovery of H2-
[t-BuNSN] was monitored by TLC (Rf ) 0.9 on Al2O3 with 3:1
pentane/diethyl ether as eluent). Removal of solvents in vacuo
gave analytically pure H2[t-BuNSN] as a pale orange solid (22
g, 47%): 1H NMR (CDCl3) δ 7.25 (dd, 2), 7.15 (td, 2), 6.98 (dd,
Unless noted otherwise, 1H and 13C NMR spectra were
recorded at 300 and 75 MHz, respectively. Chemical shifts
were referenced using the residual solvent peak and are listed
as parts per million downfield from tetramethylsilane. Routine
coupling constants are not listed. All NMR spectra were
recorded at room temperature unless noted otherwise. All
NMR solvents were deoxygenated and dried over 4 Å sieves.
Elemental analyses were performed with a Perkin-Elmer 2400
CHN analyzer or were carried out by Microlytics of Deerfield,
MA. As this analysis measures the number of moles of water
produced, the percentage of hydrogen for deuterium-containing
compounds was calculated from the actual molecular weight,
2
1
assuming that all H present was actually H.
26
Zr(NMe2)4 was prepared according to the literature pro-
2), 6.62 (td, 2), 4.49 (br s, 2, NH), 1.32 (s, 6, CCH3(CD3)2); 13
C
cedure. [Ph3C][B(C6F5)4] and [PhNMe2H][B(C6F5)4] were gifts
NMR (CDCl3) δ 147.2, 134.0, 129.8, 119.1, 117.7, 114.9, 51.6,
(25) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J . Organometallics 1996, 15, 1518.
(26) Diamond, G. M.; Rodewald, S.; J ordan, R. F. Organometallics
1995, 14, 5.
(27) Rosenberger, C.; Schrock, R. R.; Davis, W. M. Inorg. Chem.
1997, 36, 123.
(28) Overman, L. E.; Smoot, J .; Overman, J . D. Synthesis 1974, 59.

Zr Complexes Containing Tridentate Amido Ligands
Organometallics, Vol. 18, No. 5, 1999 851
30.2, 29.7 (m, CCH3(CD3)2). Anal. Calcd for C20H16D12N2S: C,
70.53; H, 8.29; N, 8.23. Found: C, 70.13; H, 8.01; N, 8.01.
S[o-C6H4NHCH(CH3)2]2 (H2[i-P r NSN]). Zinc dust (15 g,
0.23 mol), S(o-C6H4NH2)2 (5.0 g, 0.023 mol), and acetone (13
g, 0.23 mol) were added to 80 mL of glacial acetic acid. The
mildly exothermic reaction was stirred at room temperature
°C: 1H NMR (C6D6) δ 7.05 (br m, 2), 6.81 (td, 2), 6.63 (dd, 2),
6.51 (td, 2), 1.22 (s, 6, CCH3(CD3)2); 13C NMR (C6D6, 70 °C) δ
153.2, 133.4, 130.9, 125.4, 123.9, 123.4, 59.2, 30.5, 29.9 (m,
CCH3(CD3)2). Anal. Calcd for C20H14D12Cl2N2SZr: C, 48.09; H,
5.25; N, 5.61. Found: C, 48.05; H, 5.34; N, 5.44.
[t-Bu NSN]Zr Me2. To a precooled (-30 °C) suspension of
[t-BuNSN]ZrCl2 (0.85 g, 1.7 mmol) in 35 mL of diethyl ether
was added MeMgCl in THF (3.0 M, 1.13 mL, 3.4 mmol). The
suspension was kept at -30 °C (with occasional shaking), as
the bright yellow solid was gradually replaced by a white
precipitate and yellow solution. After 2 h, 15 mL of precooled
(-30 °C) pentane and 2 drops of 1,4-dioxane were added. The
suspension was filtered through a bed of Celite (the frit, Celite,
and receiving flask had been precooled to -30 °C), the filtrate
concentrated until precipitation was observed (∼20 mL), and
the mixture stored at -30 °C overnight. The fine yellow powder
was filtered (precooled frit), washed with cold pentane, and
dried in vacuo briefly (10 min). The sample was cooled to -30
°C and placed in vacuo again for 10 min. [t-BuNSN]ZrMe2 was
obtained as a fine yellow powder (0.37 g). A second crop (0.18
g) was obtained by concentration of the filtrate to ∼10 mL and
cooling to -30 °C overnight: total yield 0.55 g (70%); 1H NMR
(toluene-d8, 0 °C) δ 7.25 (dd, 2), 6.95 (td, 2), 6.86 (dd, 2), 6.58
(td, 2), 1.29 (s, 6, CCH3(CD3)2), 1.02 (s, 6, Zr(CH3)2); 13C NMR
(toluene-d8, -25 °C) δ 153.4 (br), 134.2 (br), 130.4, 123.4,
122.10 (br), 57.4, 45.9 (Zr-CH3), 44.5 (Zr-CH3), 30.4, 29.8 (m,
CCH3(CD3)2).
[i-P r NSN]Zr (NMe2)2. To a solution of Zr(NMe2)4 (1.08 g,
4.04 mmol) in 5 mL of pentane was added H2[i-PrNSN] (1.28
g, 4.26 mmol) in 3 mL of pentane. The reaction was complete
(by 1H NMR) after 4 h at room temperature. The solvents were
removed in vacuo to leave [i-PrNSN]Zr(NMe2)2 as a thick
yellow oil: yield 1.98 g (100%). The complex was pure (>95%)
by 1H and 13C NMR and was used directly in further reactions.
Analytically pure [i-PrNSN]Zr(NMe2)2 could be obtained oc-
casionally in low yield by cooling a saturated pentane solution
of the complex to give a yellow solid: 1H NMR (C6D6) δ 7.30
(dd, 2), 7.02 (td, 4), 6.65 (dd, 2), 6.49 (td, 2), 3.91 (sept, 2), 2.88
(s, 6, N(CH3)2), 2.79 (s, 6, N(CH3)2), 1.32 (d, 3, NCH(CH3)2),
1.14 (d, 3, NCH(CH3)2); 13C NMR (C6D6) δ 155.7, 133.8, 130.9,
118.1, 117.6, 116.2, 51.2 (NCH(CH3)2), 43.3, 42.1, 23.7 (NCH-
(CH3)2), 23.1 (NCH(CH3)2). Anal. Calcd for C22H34N4SZr: C,
55.30; H, 7.17; N, 11.73. Found: C, 55.29; H, 7.30; N, 11.66.
[i-P r NSN]Zr Cl2. To a solution of [i-PrNSN]Zr(NMe2)2 (1.88
g, 3.9 mmol) in 20 mL of toluene was added an excess of Me3-
SiCl (1.7 g, 16 mmol). The solution turned yellow, and a yellow
precipitate formed within 15 min. After 5 h, 20 mL of pentane
was added and the suspension was cooled to -30 °C. After 2
h the fine yellow solid was collected by filtration, washed twice
with 40 mL of pentane, and dried in vacuo to give 1.62 g (90%)
of analytically pure [i-PrNSN]ZrCl2: 1H NMR (C6D6, 500 MHz)
δ 7.23 (dd, 2), 6.83 (td, 2), 6.45 (m, 4), 4.63 (sept, 2), 1.29 (s,
br, 12, NCH(CH3)2); 13C NMR (C6D6, 125.8 MHz) δ 155.1, 134.0,
130.5, 121.7, 116.3, 48.9 (NCH(CH3)2), 22.3 (NCH(CH3)2), 21.2
(NCH(CH3)2). Anal. Calcd for C18H22Cl2N2SZr: C, 46.94; H,
4.81; N, 6.08. Found: C, 46.93; H, 4.78; N, 6.04.
1
and judged (by H NMR) to be complete in 48 h. The reaction
slurry was poured into a vigorously stirred mixture of 160 mL
of concentrated aqueous NH3, 140 mL of ice/water, and 300
mL of diethyl ether. Once most of the solid had dissolved, the
layers were separated and the aqueous layer was extracted
twice with 100 mL of diethyl ether. The organic extracts were
combined and dried over MgSO4. The solution was concen-
trated to ∼50 mL in vacuo. The product was purified by flash
chromatography on an alumina column (8 × 3 cm) using
diethyl ether as the eluent. The solvent was removed in vacuo
and the dark orange oil dried in vacuo (30 mTorr, 3 h) to give
6.6 g (95%) of analytically pure H2[i-PrNSN]. The oil crystal-
lizes upon standing for several weeks at room temperature:
1H NMR (C6D6) δ 7.39 (dd, 2), 7.02 (td, 2), 6.50 (td, 2), 6.42
(dd, 2), 4.51 (d, 2, NH), 3.26 (sept, 2), 0.83 (d, 12); 13C NMR
(C6D6) δ 147.8, 134.5, 129.9, 117.4, 117.2, 111.6, 44.2 (NCH-
(CH3)2), 22.8 (NCH(CH3)2). Anal. Calcd for C18H24N2S: C,
71.96; H, 8.05; N, 9.32. Found: C, 71.70; H, 8.07; N, 9.33.
[t-Bu NSN]Zr (NMe2)2. A solution of H2[t-BuNSN] (4.00 g,
11.7 mmol) and Zr(NMe2)4 (3.14 g, 11.7 mmol) in 30 mL of
toluene was heated to 65 °C under a slow purge of nitrogen.
After 2 h, the reaction was carefully subjected to vacuum for
∼10 s and backfilled with nitrogen three times. After 6, 10,
and 16 h, the vacuum/backfill procedure was repeated. After
18 h at 65 °C, the toluene solution was cooled to room
temperature, filtered, and concentrated to about 5 mL in
vacuo. Pentane was added (10 mL) and the solution was cooled
to -30 °C overnight. A first crop (3.60 g) of analytically pure
colorless crystals of [t-BuNSN]Zr(NMe2)2 was isolated, washed
with cold pentane, and dried in vacuo. A second crop (0.86 g)
and third crop (0.22 g) of crystals were obtained from the
filtrate by removal of all volatile components in vacuo and
recrystallization of the residue from pentane: total yield (three
crops) 4.68 g (77%); 1H NMR (C6D6) δ 7.31 (dd, 2), 6.97 (m, 4),
6.62 (m, 2), 3.03 (s, 12, N(CH3)2), 1.16 (s, 6, CCH3(CD3)2); 13C
NMR (C6D6) δ 154.7, 133.1, 129.8, 126.9, 126.6, 122.0, 58.3,
44.9, 32.2, 31.6 (m, CCH3(CD3)2). Anal. Calcd for C24H26D12N4-
SZr: C, 55.79; H, 7.40; N, 10.85. Found: C, 55.74; H, 7.10; N,
10.78.
[t-Bu NSN]Zr Cl(NMe2). Me3SiCl (100 mg, 0.92 mmol) was
added to a solution of [t-BuNSN]Zr(NMe2)2 (0.50 g, 0.97 mmol)
in 15 mL of diethyl ether over a period of 5 min. The solution
turned yellow and became cloudy. After 18 h, the suspension
was concentrated to 8 mL and cooled to -30 °C overnight. The
microcrystalline yellow solid was collected by filtration, washed
twice with 1 mL of cold diethyl ether, and dried in vacuo (290
mg). A second crop (100 mg) was obtained by concentrating
the filtrate to 3 mL and cooling (-30 °C) overnight. Total
yield: 0.39 g (74%). Analytically pure samples were obtained
by a second recrystallization from diethyl ether: 1H NMR
(C6D6) δ 7.21 (dd, 2), 6.94 (td, 2), 6.85 (dd, 2), 6.54 (td, 2), 2.97
(s, 6, N(CH3)2), 1.36 (s, 6, CCH3(CD3)2); 13C NMR (C6D6) δ
155.0, 151.9, 133.4, 130.4, 123.2, 121.8, 66.2, 43.3, 31.1, 30.4
(m, CCH3(CD3)2). Anal. Calcd for C22H20D12ClN3SZr: C, 52.00;
H, 6.35; N, 8.27. Found: C, 51.87; H, 6.35; N, 8.01.
[i-P r NSN]Zr Me2. To a precooled (-30 °C) suspension of
[i-PrNSN]ZrCl2 (0.98 g, 2.12 mmol) in 25 mL of diethyl ether
was added MeMgCl in THF (3.0 M, 1.4 mL, 4.2 mmol). The
suspension was kept at -30 °C (with occasional shaking), as
the bright yellow solid was replaced by a white precipitate and
pale yellow solution. After 1 h the suspension was filtered
through a bed of Celite and the filtrate concentrated to ∼3
mL and stored at -30 °C overnight. The light yellow crystals
were filtered and washed with 1 mL of cold diethyl ether and
then 1 mL of cold pentane. The residual solvent was removed
by dissolving the sample in cold (-30 °C) toluene (3 mL) and
removing the solvent in vacuo; three repetitions removed
>95% of the THF and diethyl ether (by 1H NMR). This
procedure also led to a slight darkening of the solution (to
yellow-brown), although no change in purity (>98%) was
[t-Bu NSN]Zr Cl2. Me3SiCl (3.8 g, 35 mmol) was added to a
solution of [t-BuNSN]Zr(NMe2)2 (4.5 g, 8.7 mmol) in 50 mL of
diethyl ether. Immediately a yellow color was observed,
followed by precipitation of a bright yellow solid. After 12 h,
the volatiles were removed in vacuo and 30 mL of pentane
was added to the residue. The solid was filtered off, washed
liberally with pentane, and dried in vacuo to give 4.1 g (94%)
of crude product that was pure enough (>95% by NMR) for
use in subsequent reactions. Analytically pure samples were
obtained by recrystallization from toluene/pentane at -30

852 Organometallics, Vol. 18, No. 5, 1999
Graf et al.
observed according to NMR (1H, 13C): yield 630 mg of tan
powder (71%); 1H NMR (C6D6) δ 7.27 (dd, 2), 7.02 (td, 2), 6.65
(dd, 2), 6.48 (td, 2), 4.61 (sept, 2), 1.69 (br s, 6, NCH(CH3)2),
1.36 (br s, 6, NCH(CH3)2), 0.61 (br s, 3, Zr-CH3), 0.26 (br s, 3,
Zr-CH3); 13C NMR (C6D6) δ 153.3, 133.4, 130.8, 118.8, 116.9,
114.2, 47.5 (NCH(CH3)2), 39.5 (br, Zr-CH3), 22.4 (NCH(CH3)2),
21.5 (NCH(CH3)2).
d5. After 5 min, 10 equiv of 1-hexene (60 µL) was added and
the sample stored at -10 °C. After 3 h, the H NMR spectrum
1
indicated all 1-hexene was converted to poly(1-hexene). Under
similar conditions but with 50-100 equiv of 1-hexene added,
only a small portion (10-20 equiv) was polymerized in 3 h.
The reaction was also conducted on a batch scale. [t-BuNSN]-
ZrMe2 (22 mg, 48 µmol) was dissolved in 5 mL of precooled
(-30 °C) chlorobenzene and added to a -30 °C solution of
[Ph3C][B(C6F5)4] (38 mg, 41 µmol) in 5 mL of chlorobenzene.
After 5 min, 300 equiv (1.0 g, 1.5 mL) of 1-hexene was added
and the reaction mixture was stirred at -10 °C for 3 h. The
reaction was quenched by adding 3 mL of 1.0 M HCl in diethyl
ether, and the solvents were removed in vacuo to leave 200
mg (20%) of a white sticky residue whose NMR is consistent
with poly(1-hexene). Similar yields are obtained if 100 or 500
equiv of 1-hexene is used.
[i-P r NSN]Zr (CH2CHMe2)2. To a precooled (-30 °C) sus-
pension of [i-PrNSN]ZrCl2 (0.50 g, 1.1 mmol) in 25 mL of
diethyl ether was added i-BuMgCl (2.0 M, 1.1 mL, 2.2 mmol).
The suspension was kept at -30 °C for 1.5 h (with occasional
shaking). The suspension was filtered through a bed of Celite.
The filtrate was concentrated to ∼2 mL, 2 mL of pentane was
added, and the solution was stored at -30 °C overnight. The
light yellow crystals were filtered off and washed with 1 mL
of cold diethyl ether and then 1 mL of cold pentane. Residual
diethyl ether and pentane were removed by dissolving the
sample in cold (-30 °C) toluene (3 mL) and removing the
solvent in vacuo several times. This procedure also led to a
slight darkening of the solution (to orange-yellow), although
no change in purity (>98%) was observed by NMR (1H, 13C):
Rea ction of [i-P r NSN]Zr Me2 w ith [P h NMe2][B(C6F 5)4].
[i-PrNSN]ZrMe2 (25 mg, 59 µmol) was dissolved in 0.6 mL of
bromobenzene-d5, the solution was cooled to -30 °CC, and
slightly less than 1 equiv of powdered [PhNMe2][B(C6F5)4] (43
mg, 54 µmol) was added. The reaction mixture was warmed
to -10 °C and stirred for 45 min as the color slowly changed
1
yield 410 mg of an orange powder (75%); H NMR (C6D6, 500
1
MHz) δ 7.32 (dd, 2), 7.04 (td, 2), 6.76 (dd, 2), 6.54 (td, 2), 4.70
(sept, 2), (s, 6, Zr(CH3)2), 2.21 (br s, 1, ZrCH2CHMe2), 2.07 (br
s, 1, ZrCH2CHMe2), 1.78 (br s, 6, NCH(CH3)2), 1.39 (br s, 6,
NCH(CH3)2), 1.01 (br s, 10, (ZrCH2CH(CH3)2), 0.74 (br s, 6,
ZrCH2CH(CH3)2); 13C NMR (C6D6, 125.8 MHz) δ 152.4, 133.3,
130.7, 118.8, 116.9, 114.8, 80.1 (Zr-CH2), 78.1 (Zr-CH2), 47.8
(NCH(CH3)2), 28.3, 22.7 (NCH(CH3)2), 21.7 (NCH(CH3)2).
“{[t-Bu NSN]Zr Me}{B(C6F 5)4}”. [t-BuNSN]ZrMe2 (20 mg,
43 µmol) was dissolved in 0.3 mL of precooled chlorobenzene-
d5 and added to a -30 °C solution of [Ph3C][B(C6F5)4] (39 mg,
42 µmol) in 0.3 mL of chlorobenzene-d5. The initial red-orange
solution immediately turned orange-yellow. The sample was
kept at -10 °C during transfer to a precooled NMR probe. The
1H NMR spectrum was indicative of clean formation of {[t-
BuNSN]ZrMe}{B(C6F5)4}: 1H NMR (C6D5Cl, -10 °C) δ 7.50-
6.98 (m, ∼23), 2.30 (s, 3, Ph3CCH3), 1.35 (s, 6, CCH3(CD3)2),
1.09 (s, 3, ZrCH3).
to orange-red. The H NMR of the sample after ∼50 min was
indicative of two main species, A (75%) and B (25%). 1H NMR
(23 °C, C6D5Br): δ 7.5-6.5 (m, ∼18, A and B), 4.16 (sept, ∼1.5,
A), 4.05 (sept, ∼0.5, B), 3.19 (br s, ∼4, A/B), 3.0 (br s, ∼1, A/B),
2.92 (s, 3, A/B), 1.71 (d, ∼6, A), 1.46 (d, ∼3, B), 1.36 (d, ∼3,
B), 1.11 (d, ∼6, A), 1.04 (s, ∼3, A). A mixture of species A (50%)
and B (50%) is also observed if the reaction mixture is warmed
to room temperature for 10 min instead of -10 °C for 45 min.
The assignment of resonances for A and B is supported by
the observation that, at room temperature, the resonances
assigned to A decrease in intensity as those of B increase until
significant decomposition occurs within 90 min at room
temperature. Conversion of A to B is facilitated by heating
the sample. Attempts to control the reaction conditions to
1
obtain a clean H NMR spectrum of A or B were unsuccessful.
Ack n ow led gm en t. R.R.S. thanks the Department
of Energy (Contract No. DE-FG02-86ER13564) and
Exxon for supporting this research and the National
Science Foundation for funds to help purchase a de-
partmental SMART/CCD diffractometer. R.S. thanks
the Alexander von Humboldt Foundation for a Feodor-
Lynen-Fellowship. D.D.G. thanks R. Baumann for use-
ful discussions.
The above reaction was also carried out by utilizing
[t-BuNSN]Zr(13CH3)2 and [Ph3C][B(C6F5)4]. The resulting 13C
NMR spectrum was indicative of clean formation of [t-BuNSN]-
Zr(13CH3)][B(C6F5)4] and Ph3C(13CH3): 13C NMR (C6D5Cl, -10
°C) 52.8 (br, Zr-13CH3), 31.24 (Ph3C-13CH3).
Addition of an excess of THF (3 equiv) to a solution of {[t-
BuNSN]ZrMe}{B(C6F5)4} led to a pale yellow solution. The 1H
NMR spectrum was indicative of clean formation of unsym-
metric {[t-BuNSN]ZrMe(THF)x}{B(C6F5)4}: 1H NMR (C6D5Cl,
23 °C) δ 7.60-6.80 (m, ∼23), 3.90 (br s, ∼12, THF), 2.29 (s, 3,
Ph3CCH3), 1.74 (br s, ∼12, THF), 1.58 (s, 3, CCH3(CD3)2), 1.34
(s, 3, CCH3(CD3)2), 0.90 (s, 3, ZrCH3).
Su ppor tin g In for m ation Available: Fully labeled ORTEP
drawings and tables of atomic coordinates, bond lengths and
angles, and anisotropic displacement parameters for [t-BuN-
SN]ZrCl(NMe2) and [i-PrNSN]ZrMe2. This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
Attem p ted P olym er iza tion of 1-Hexen e. [t-BuNSN]-
ZrMe2 (21 mg, 46 µmol) was dissolved in 0.3 mL of precooled
(-30 °C) chlorobenzene-d5 and added to a -30 °C solution of
[Ph3C][B(C6F5)4] (38 mg, 41 µmol) in 0.3 mL of chlorobenzene-
OM980934K