Synthesis and characterization of organozirconium complexes bearing a new trianionic multidentate cyclopentadienyl ligand by use of amine elimination

Article abstract of DOI:10.1016/S0022-328X(00)00179-0

The new multidentate ligand o-(NH(C₆H₅)(N′H(Si(CH₃) ₂(C₅(CH₃)₄H))C₆H ₄ (1), was synthesized by the reaction of (C₅(CH₃)₄(H)Si(CH₃)₂Cl) with o-C₆H₄(NH₂)(N′HPh) in the presence of NEt₃. The ligand was bound to Zr using the amine elimination method through the reaction of 1 with Zr(N(CH₃)₂)₄. Because of the incomplete removal of HCl generated in the synthesis of 1, two complexes were isolated and characterized from the reaction of 1 with Zr(N(CH₃)₂)₄, o-C₆H₄{N(C₆H₅)N′[Si(CH ₃)₂(C₅(CH₃) ₄)]}Zr(N(CH₃)₂)(NH(CH₃)₂) (2) and o-C₆H₄{N(C₆H₅)N′[Si(CH ₃)₂(C₅(CH₃) ₄)]}Zr(Cl)(NH(CH₃)₂)₂ (3). The structures of both 2 and 3 were confirmed by X-ray crystallography. Crystals of 2 are triclinic, space group P1?, with a=10.9864(1), b=11.8841(2), c=12.3633(1) ?, and Z=2. Compound 2 has a square-pyramidal geometry with the -NMe₂ group occupying the apical position. Crystals of 3 are monoclinic, space group P2₁/c, with a=12.1925(2), b=18.9677(4), c=12.2218(2) ?, and Z=4. Complex 3 has a distorted octahedral geometry with the cyclopentadienyl and the amido groups of the chelating ligand occupying pseudo-meridional positions.

Full text of DOI:10.1016/S0022-328X(00)00179-0

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 603 (2000) 213–219
Synthesis and characterization of organozirconium complexes
bearing a new trianionic multidentate cyclopentadienyl ligand by
use of amine elimination
P. Doufou, K.A. Abboud, J.M. Boncella *
Department of Chemistry and Center for Catalysis, Uni6ersity of Florida, PO Box 117200, Gaines6ille, FL 32611-7200, USA
Received 24 August 1999; received in revised form 2 February 2000
Abstract
The new multidentate ligand o-(NH(C6H5)(N%H(Si(CH3)2(C5(CH3)4H))C6H4 (1), was synthesized by the reaction of
(C5(CH3)4(H)Si(CH3)2Cl) with o-C6H4(NH2)(N%HPh) in the presence of NEt3. The ligand was bound to Zr using the amine
elimination method through the reaction of 1 with Zr(N(CH3)2)4. Because of the incomplete removal of HCl generated in the
synthesis of 1, two complexes were isolated and characterized from the reaction of
C6H4{N(C6H5)N%[Si(CH3)2(C5(CH3)4)]}Zr(N(CH3)2)(NH(CH3)2) (2) and o-C6H4{N(C6H5)N%[Si(CH3)2(C5(CH3)4)]}Zr(Cl)-
(NH(CH3)2)2 (3). The structures of both 2 and 3 were confirmed by X-ray crystallography. Crystals of 2 are triclinic, space group
1 with Zr(N(CH3)2)4, o-
(
,
P1, with a=10.9864(1), b=11.8841(2), c=12.3633(1) A, and Z=2. Compound 2 has a square-pyramidal geometry with the
ꢀNMe2 group occupying the apical position. Crystals of 3 are monoclinic, space group P21/c, with a=12.1925(2), b=18.9677(4),
,
c=12.2218(2) A, and Z=4. Complex 3 has a distorted octahedral geometry with the cyclopentadienyl and the amido groups of
the chelating ligand occupying pseudo-meridional positions. © 2000 Published by Elsevier Science S.A. All rights reserved.
Keywords: Zirconium; ansa-Metallocene; Cyclopentadienyldiamide; Amine elimination
employed as ancillary ligands in a-olefin polymerization
catalysts [1c,d,4]. The substitution of a cyclopentadienyl
ring with an amido ligand increases the Lewis acidity
and electrophilicity of the metal center in these com-
plexes because of reduced steric crowding and electron
count compared to their bis(cyclopentadienyl) analogs.
For example, scandium complexes such as (h5,h1-
C5Me4SiMe2NCMe3)ScR, synthesized by Bercaw et al.
[1c,d,4a] using the C5Me4SiMe2NCMe3 ligand, have
been shown to have significantly enhanced Lewis acid-
ity compared to ansa-metallocenes.
To date, most of the ‘CpSiNR’ type ligands that have
been synthesized have been dianionic, and with or
without additional neutral donors. We were interested
in synthesizing similar chelating mixed cyclopentadienyl
ligands that are trianionic to examine their ability to
coordinate to early transition metals. This paper re-
ports the synthesis of two zirconium compounds that
have a CpSi(NR)2 hybrid ligand system in which the
diamide piece of the ligand is derived from an o-
phenylenediamine derivative.
1. Introduction
Among the large number of cyclopentadienyl ligands
that have been synthesized to date, there are systems
which contain a cyclopentadienyl ring linked to addi-
tional pendent groups such as amides [1], alkoxides
[1a,2], amines [1a,j–l,3], phosphines [1o,p] and olefins
[1a,h,l]. The pendent donors on the cyclopentadienyl
rings can provide an effective means of altering the
coordination environment of these cyclopentadienyl
metal complexes [1]. In recent years, these chelating
cyclopentadienyl ligands have generated increasing in-
terest with respect to their use as ancillary ligands for
homogeneous olefin polymerization catalysts [1b].
The CpSiNR class of ligands, in which a (dimethyl-
silyl)amide is linked to the cyclopentadienyl moiety,
have been successfully used as alternatives to the
bridged bis(cyclopentadienyl) ligands that have been
* Corresponding author. Fax: +1-352-3928758.
E-mail address: boncella@chem.ufl.edu (J.M. Boncella)
0022-328X/00/$ - see front matter © 2000 Published by Elsevier Science S.A. All rights reserved.
PII: S0022-328X(00)00179-0

214
P. Doufou et al. / Journal of Organometallic Chemistry 603 (2000) 213–219
2. Experimental
2.1. General considerations
118.31, 119.30 (CHC4(CH3)4), 126.98, 127.47, 129.38,
130.04, 132.45, 136.48, 144.84, 146.80 (aromatic C’s).
CIMS/CH4: M+1=363.109 amu.
All procedures were performed using standard
Schlenk techniques or in a nitrogen-filled Vacuum At-
mospheres glovebox. Diethyl ether (Et2O), pentane,
and hexanes were distilled from sodium benzophenone
ketyl; toluene was previously deolefinated by washing
twice with cold H2SO4 followed by water and bicar-
bonate, and was then distilled from potassium metal.
Dichloromethane (CH2Cl2) was distilled from CaH2.
After collection, all solvents were stored under argon
and over molecular sieves. Glassware was oven dried
before use. The compound C5(CH3)4(H)Si(CH3)2Cl
was synthesized according to published procedures [4].
N-phenyl-o-phenylenediamine was purchased from
Aldrich and was used without further purification.
Homoleptic amido complexes Zr(N(CH3)2)4 and
Ta(N(CH3)2)5 were synthesized according to the proce-
dure reported by Jordan and co-workers [6]. All NMR
solvents were purchased from Cambridge Isotope Lab-
oratories and were degassed and stored over molecular
sieves in a drybox prior to use.
2.3. o-C6H4{N(C6H5)N%[Si(CH3)2(C5(CH3)4)]}-
Zr(N(CH3)2)(NH(CH3)2) (2)
To a stirring hexane solution of Zr(N(CH3)2)4 (0.126
g, 0.47 mmol) at r.t., was added dropwise a solution
of 1 (0.18 g, 0.48 mmol) in hexanes. The reaction
mixture immediately became light orange and when
approximately half of the ligand had been added,
compound 2 began precipitating as a yellow solid.
After addition of 1 was completed, the reaction mix-
ture was allowed to stir for an additional 6 h. The
precipitate was allowed to settle, the supernatant was
removed, and the remaining solid was extracted with
pentane until the extracts were clear. Removal of the
pentane from the combined extracts gave crude 2 as a
yellow–orange solid. Compound 2 was recrystallized
1
from a toluene–pentane mixture. The H-NMR of the
first crop of crystals isolated from the toluene–pentane
mixture contained ca. 60% of 2 with the remaining
40% being complex 3 (overall yield of 2: 0.07 g, 25%).
Repeated recrystallizations separated 2 from 3. Similar
results were observed when the reaction was per-
formed in Et2O. X-ray quality crystals formed from a
r.t. C6D6 solution of 2. Anal. Calc. for C27H40N4SiZr:
C, 60.1; H, 7.47; N, 10.4. Found: C, 59.2; H, 7.25; N,
9.93%. 1H-NMR (25°C, C6D6): l 0.79, 0.86 (s, 6H
each, Si(CH3)2), 1.61, 1.77, 2.07, 2.14 (s, 3H each,
C5(CH3)4), 1.62 (6H, N(CH3)2H), 2.71 (s, 6H,
ꢀN(CH3)2), 6.88–7.21 (multiplets, 9H, aromatic H).
The N(CH3)2H is hidden in the baseline. 13C-NMR
(25°C, C6D6): l 3.53, 3.92 (ꢀSi(CH3)2), 12.05, 13.35,
13.49, 14.04 (C5(CH3)4)H), 38.37 (NH(CH3)2), 46.51
(N(CH3)2), 102.70, 110.94, 114.00, 116.17, 118.90,
123.63, 128.81, 129.94, 144.48, 151.97, 155.21
(C5(CH3)4)H and aromatic C’s).
NMR spectra were measured either on a Varian
VXR or Gemini-300 (300 MHz) spectrometer. Chemi-
1
cal shifts for the H spectra were referenced to residual
protons in the deuterated solvents and are reported
relative to tetramethylsilane. Mass spectrometry was
performed by the UF Department of Chemistry Ana-
lytical Services. Elemental analyses were performed by
Atlantic Microlabs, Norcross, GA.
2.2. o-(NH(C6H5)(N%H((C5(CH3)4H)Si(CH3)2)C6H4 (1)
To a −78°C Et2O solution of (C5(CH3)4H)Si-
(CH3)2Cl (1.17 g, 5.44 mmol) an Et2O solution of
N-phenyl-o-phenylenediamine (1.03 g, 5.59 mmol) was
slowly added. The reaction mixture was warmed to
room temperature (r.t.) and stirred for ꢀ6 h. At this
time the reaction mixture was light pink in color and a
small amount of a pink solid had precipitated. To this
reaction mixture excess Et3N (3 ml, 40.9 mmol) was
added at r.t. resulting in a color change from pink to
pale orange. The reaction was stirred for 1 h after
which the Et2O was removed under reduced pressure,
and a waxy pale-orange solid was obtained. Extraction
of the solid with four portions of cold pentane fol-
lowed by filtration and removal of the pentane under
reduced pressure gave 1 as a thick, pale orange, oil
2.4. o-C6H4{N(C6H5)N%[Si(CH3)2(C5(CH3)4)]}Zr-
(Cl)(NH(CH3)2)2 (3)
To a hexane solution of 1 (0.14 g, 0.39 mmol, 2.1
equivalents) at r.t., a hexane solution of Zr(N(CH3)2)4
(0.05 g, 0.18 mmol) was added dropwise. An immedi-
ate color change to orange and precipitation of a
yellow–orange solid was observed. The reaction mix-
ture was stirred for an additional 15 min and was left
to settle. The supernatant was removed and the solid
was washed twice with pentane to remove the excess
ligand (yield: 0.03 g, 28%). X-ray quality crystals were
obtained within 4 days from a r.t. C7D8 solution of 3.
Anal. Calc. for C27H40ClN4SiZr: C, 56.4; H, 7.01; N,
9.74. Found: C, 55.3; H, 6.58; N, 9.52. 1H-NMR
1
(yield: 1.6 g, 82%). H-NMR (25°C, C6D6): l 0.14 (s,
6H, Si(CH3)2), 1.72, 1.87 (s, 6H each, C5(CH3)4)H),
2.91 (s, 1H, C5(CH3)4)H), 4.10, 4.38 (broad s, 1H
each, NꢀH), 6.49–7.09 (phenyl region, 9H). 13C-NMR
(25°C, C6D6):
(C5(CH3)4)H), 54.88 (CHC4(CH3)4), 115.09, 115.98,
l
2.97 (ꢀSi(CH3)2), 11.25 14.38

P. Doufou et al. / Journal of Organometallic Chemistry 603 (2000) 213–219
215
3. Results and discussion
(25°C, C6D6): l 0.70 (s, 6H, ꢀSi(CH3)2), 1.88, 1.93 (s,
6H each, C5(CH3)4)H), 2.00 (broad s, 12H, NH(CH3)2),
3.28 (very broad s, 2H, NH(CH3)2), 6.57–7.41 (multi-
plets, 9H, aromatic protons). 13C-NMR (25°C, C6D6): l
3.46 (ꢀSi(CH3)2), 12.21, 14.05 (C5(CH3)4)H), 40.40
(NH(CH3)2), 112.58, 115.12, 115.92, 121.10, 123.62,
126.24, 127.31, 129.51, 143.51, 155.09 (C5(CH3)4)H and
aromatic C’s).
In order to synthesize a bis-amido cyclopentadienyl
chelating trianionic ligand we chose unsubstituted or
monosubstituted phenyleldiamines as the bis-amido
part of the ligand. Initially, we attempted reaction of
o-C6H4(NH2)2 or o-C6H4(NH2)(N%(H)SiMe3) with
C5Me4HSiMe2Cl, however, these reactions produced
multiple products. In both cases one of the products
were identified by mass spectrometry to be the bis-cy-
clopentadienyl compound o-C6H4[(C5Me4H)SiMe2-
(NH)]2. To avoid the problem of multiple substitution
and/or cleavage of the already existing substituents on
the nitrogens, we pursued the reaction of o-
(PhNH)C6H4(NH2) with C5Me4HSiMe2Cl in which
case the desired product, o-(NHPh)C6H4((C5Me4H)-
SiMe2NH) (1) was obtained in good yield (though
always contaminated with small amounts of an impu-
rity that was tentatively identified as 1·HCl as described
below) (Eq. (1)).
2.5. X-ray data collection and structure refinement for
compounds 2 and 3
Data for both compounds were collected at 173 K on
a Siemens CCD SMART PLATFORM equipped with a
CCD area detector and a graphite monochromator
,
utilizing Mo–Ka radiation (u=0.71073 A). Cell
parameters were refined using 8192 reflections from
each data set. A hemisphere of data (1381 frames) was
collected using the ꢀ-scan method (0.3° frame width).
(1)
Although a number of cyclopentadienyl isomers are
possible via possible proton or silicon sigmatropic shifts
[8], the H-NMR spectrum of 1 indicates the presence
The first 50 frames were remeasured at the end of data
collection to monitor instrument and crystal stability
(maximum correction on I was B1%). Absorption
corrections were applied based on the „-scan using the
entire data sets.
1
of only one isomer with the tetramethylcyclopentadi-
enyl methine proton and ꢀSiMe2 group bound at the 5
position of the Cp ring. The methine proton resonates
at 2.91 ppm, whereas the two secondary amine protons
are observed as two singlets at 4.10 and 4.38 ppm,
respectively. Compound 1 is relatively air and moisture
sensitive, and unstable in chlorinated solvents, with
cleavage of the SiꢀN bond a likely decomposition path-
way. Alternatively, 1 could be isolated in low yield by
initial deprotonation of o-(PhNH)C6H4(NH2) with n-
BuLi and subsequent addition of (C5Me4H)SiMe2Cl.
The usual synthetic approach to Group III and IV
transition metal complexes that contain cyclopentadi-
enylꢀamido ligands employs the metathetical reaction
of the alkali salts of these ligands with the appropriate
metal halides [1c,d,g,h,j,l,n,4]. Therefore, we attempted
the synthesis of alkali salts of 1 by reaction with MeLi,
n-BuLi, LiN(CH3)2 and KH with no success. Addition-
ally, when we attempted in situ deprotonation of 1 and
subsequent addition to ZrCl4 or TaCl5, intractable mix-
tures of compounds were observed.
Both structures were solved by direct methods in
SHELXTL [7] and were refined using full-matrix least-
squares on F2. The non-H atoms were refined with
anisotropic thermal parameters. All of the H atoms
were refined without any constraints. Compound 2 also
has disordered benzene molecules located on centers of
inversion. They were refined as two dependent parts
and their site occupation factors refined to 0.56(3) and
0.44(3), respectively. The benzene H atoms were in-
cluded in the final cycle of refinement in a riding mode
on their parent atoms. For compound 2, 514 parame-
ters were refined in the final cycle of refinement using
6040 reflections with I\2|(I) to yield R1 and wR2 of
2.38 and 5.92, respectively. For compound 3, 472
parameters were refined in the final cycle of refinement
using 5484 reflections with I\2|(I) to yield R1 and
wR2 of 2.37 and 5.79, respectively. Refinement was
performed using F2.

216
P. Doufou et al. / Journal of Organometallic Chemistry 603 (2000) 213–219
The methods of amine [1e,i,5] and alkane elimination
[1k,m] are alternative strategies for the attachment of
cyclopentadienylꢀamido type ligands to transition
metals. We therefore pursued these possibilities in order
to coordinate 1 to zirconium and tantalum. As a model
reaction, we initially investigated the amine elimination
route by addition of one equivalent of 1 to a solution of
Zr(NMe2)4. This resulted in the isolation of
o-C6H4{N(C6H5)N%[Si(CH3)2(C5(CH3)4)]}ꢀZr(N(CH3)2)-
(NH(CH3)2) (2) in low yield (Eq. (2)). We found that
the coordinated NMe2H could not be removed even
when the reaction mixture was heated in an open
system.
Fig. 1. Thermal ellipsoid plot of the molecular structure of o-
C6H4{N(C6H5)N%[Si(CH3)2(C5(CH3)4)]}Zr(N(CH3)2)(NH(CH3)2) (2).
(2)
1
The H-NMR spectrum of 2 exhibits four Cp ring
methyl (1.61, 1.77, 2.07, 2.14 ppm) and two silyl methyl
(0.79, 0.86 ppm) resonances reflecting an asymmetric
geometry. The methyls of the coordinated NMe2H ap-
pear as a broad singlet at 1.62 ppm, whereas the amine
proton is either broadened into the baseline or overlaps
with the Me peaks of the amine or amide groups. The
structure in Eq. (2) is consistent with the NMR spectra
and was confirmed via a single-crystal X-ray diffraction
study (vide infra).
The single-crystal X-ray diffraction study performed
on 2 confirmed the proposed structure of this complex.
A thermal ellipsoid plot of 2 is found in Fig. 1. The
crystallographic data are presented in Table 1, while
selected bond lengths and angles for 2 are summarized
in Table 2.
The geometry around zirconium is best described as
a distorted square pyramid with the dimethylamido
ligand occupying the apical position. The dimethy-
lamine, the centroid of the cyclopentadienyl ligand
(Cp(c)), and the two amido groups of the phenylenedi-
amine define the base of the pyramid. The angles
between both sets of trans ligands are less than 180°.
Therefore, the Zr atom is slightly displaced from the
Table 1
Summary of crystallographic data, collection parameters, and refine-
ment parameters for 2 and 3 a
2
3
Crystal data (173 K)
Empirical
formula
FW
Crystal size (mm) 0.34×0.26×0.11
Crystal system
Space group
C
27H40N4SiZr·1/2(C6H6) C27H41ClN4SiZr
578.99
576.40
0.34×0.26×0.13
Monoclinic
P21/c
12.1925(2)
18.9677(4)
12.2218(2)
Triclinic
P1
(
,
a (A)
10.9864(1)
11.8841(2)
12.3633(1)
77.310(1)
70.304(1)
77.732(1)
1465.68(3)
2
,
b (A)
,
c (A)
h (°)
i (°)
k (°)
91.407(1)
3
,
V (A )
2825.6(9)
4
1.355
1208
Z
Dcalc (Mg m−3
)
1.312
F(000), electrons 610
Structure
refinement
Refinement
method
S, goodness-of-fit 1.092
R1/reflections
wR2/reflections
Rint (%)
,
equatorial plane of the square pyramid. The 2.059 A
Full-matrix least-squares Full-matrix
distance between ZrꢀN(3) is comparable to the average
ZrꢀN distances in compounds such as Zr(NMe2)4 [9],
rac-[C2H4(indenyl)2]Zr(NMe2)2 [6], and [(C5Me4)-
SiMe2(Nꢀt-Bu)]Zr(NMe2)2 [1i], and is consistent with
the presence of p-donation from the nitrogen of the
amido ligand to the Zr center.
on F2
least-squares on F2
1.022
0.0237/5464\2|(I)
0.0579/5464\2|(I)
2.50
0.0238/6040\2|(I)
0.0592/6040\2|(I)
2.25
a R1=S(ꢀꢀFoꢀ−ꢀFcꢀꢀ)/SꢀFoꢀ,
S=[S[w(F2o−F2c)2]/(n−p)]1/2
p=[max(F2o, 0)+2F2c]/3.
wR2=[S[w(F2o−F2c)2]/S[w(Fo2)2]]1/2
w=1/[|2(F2o)+(0.0370p)2+0.31p],
,
The sum of the angles around the chelating nitro-
gens, N(1) and N(2) are approximately 360°, indicating
,

P. Doufou et al. / Journal of Organometallic Chemistry 603 (2000) 213–219
217
Table 2
107.25°, respectively. The bite angle of the phenylenedi-
amido fragment is 71.34°, which is considerably smaller
than the 86.8° angle reported for complex [o-
C6H4(NSiiPr3)2]2Zr [10].
a
,
Selected bond distances (A) and angles (°) for 2
Bond lengths
ZrꢀN(1)
ZrꢀN(2)
ZrꢀN(3)
ZrꢀN(4)
ZrꢀCp(c)
SiꢀN(1)
N(1)ꢀC(8)
N(2)ꢀC(9)
N(2)ꢀC(14)
N(3)ꢀC(20)
N(3)ꢀC(21)
SiꢀC(1)
2.1451(12)
2.2513(12)
2.0599(13)
2.4005(13)
2.332(2)
1.7173(13)
1.385(2)
1.401(2)
1.432(2)
1.462(2)
1.452(2)
1.874(2)
Isolation of a pure sample of 2 was troublesome. Its
synthesis was accompanied by formation of a second
1
zirconium complex that, by H-NMR, appeared to be a
symmetric species. We were able to isolate this second
complex in low yields, by dropwise addition of
1
Zr(NMe2)4 to two or more equivalents of 1. H-NMR
spectroscopy together with X-ray crystallography,
confirmed
that
this
complex
was
o-C6H4-
{N(C6H5)N%[Si(CH3)2(C5(CH3)4)]}Zr(Cl)(NH(CH3)2)2
(3) (Eq. (3)). Consistent with our assignment, the bridg-
ing silyl methyls of 3 appear as a sharp singlet at 0.70
Bond angles
N(1)ꢀZrꢀN(2)
N(1)ꢀZrꢀN(3)
N(1)ꢀZrꢀN(4)
N(1)ꢀZrꢀCp(c)
N(2)ꢀZrꢀN(3)
N(2)ꢀZrꢀN(4)
N(2)ꢀZrꢀCp(c)
N(3)ꢀZrꢀN(4)
N(3)ꢀZrꢀCp(c)
N(4)ꢀZrꢀCp(c)
N(1)ꢀSiꢀC(1)
SiꢀN(1)ꢀZr
C(8)ꢀN(1)ꢀZr
C(8)ꢀN(1)ꢀSi
C(9)ꢀN(2)ꢀZr
C(9)ꢀN(2)ꢀC(14)
C(14)ꢀN(2)ꢀZr
C(20)ꢀN(3)ꢀZr
C(20)ꢀN(3)ꢀC(21)
C(21)ꢀN(3)ꢀZr
SiꢀC(1)ꢀCp(c)
71.34(4)
102.88(5)
145.90(5)
97.4(5)
96.47(5)
77.72(5)
153.0
94.38(5)
109.61
104.32
92.86(6)
107.25(6)
121.87(9)
130.64(10)
117.62(9)
113.35(11)
128.71(9)
117.44(11)
109.49(14)
132.83(12)
159.42
1
ppm in the H-NMR spectrum, whereas the methyls of
the Cp ring appear as two singlets at 1.88 (6H) and 1.93
(6H) ppm, respectively. The methyls of the coordinated
NMe2H groups appear as a sharp singlet at 2.00 ppm.
In this case, the two amine protons were observed as a
very broad singlet at 3.28 ppm.
The isolation of complex 3 was due to presence of
chloride impurity in 1. The source of chloride could be
attributed either to NEt3HCl or 1·HCl, which are both
generated during the preparation of 1. Although we
cannot be absolutely certain, two points support 1·HCl
being responsible for the isolation of 3. First, 1 was
extracted several times with cold pentane, which should
have separated 1 from the NEt3HCl byproduct. Second,
addition of NEt3HCl to pure 2 produced only a mini-
mal amount 3. Therefore, it is more likely that the
formation of 3 arises from the presence of 1·HCl. This
suggests that 1·HCl is a stable adduct that is not
completely converted to 1 even when an excess of NEt3
is present. It also cannot be successfully separated from
1 due to its similar solubility to 1 even in non-polar
solvents.
a Cp(c) denotes the centroid of the cyclopentadienyl ring.
sp2-hybridization. The ZrꢀN(1) and ZrꢀN(2) distances
of the chelating ligand however, are ꢀ0.10–0.20 A
,
longer, respectively, compared to those in [o-
C6H4(NSiiPr3)2]2Zr [10]. This reflects either a weaker
pp–dp interaction between N(1), N(2) and Zr, or exces-
sive steric congestion around the metal center.
The bond lengths between zirconium and the carbon
atoms of the cyclopentadienyl ring range from 2.474 to
,
2.769 A revealing a significant tilt due to the constraints
imposed by the rest of the chelating ligand. The
ZrꢀCp(c) distance (where Cp(c) is the centroid of the
cyclopentadienyl ring) is comparable to values reported
in the literature for similar complexes [1i]. The presence
of the ꢀSiMe2 bridge leads to reduction of the
Cp(c)ꢀZrꢀN(1) angle compared to similar complexes
with unlinked cyclopentadienyl and amido ligands
[1i,11]. The geometry imposed by the chelating ligand is
also characterized by the substantial displacement of
the Si atom from the plane of the cyclopentadienyl ring
(Cp(c)ꢀC(1)ꢀSi=159.4°), and results in the internal
C(1)ꢀSiꢀN(1) and SiꢀN(1)ꢀZr angles being 92.86 and
(3)
Orange crystals suitable for an X-ray diffraction
study were isolated from a C7D8 solution of 3 after four
days at r.t. A thermal ellipsoid plot of 3 is shown in
Fig. 2, whereas crystallographic data are summarized in
Table 1. Table 3 contains selected bond lengths and
angles for 3. The structure of o-C6H4{N(C6H5)-
N%[Si(CH3)2(C5(CH3)4)]}Zr(Cl)(NH(CH3)2)2 (3) is best
described as a distorted octahedron with the centroid of

218
P. Doufou et al. / Journal of Organometallic Chemistry 603 (2000) 213–219
Table 3
the cyclopentadienyl ring, the NPh group of the
phenylenediamido ligand, occupying mutually trans po-
sitions. The silicon atom is displaced from the plane of
the cyclopentadienyl ring as indicated by the
Cp(c)ꢀC(1)ꢀSi=156.5° due to the constraints imposed
by the chelating ligand.
a
,
Selected bond distances (A) and angles (°) for 3
Bond lengths
ZrꢀN(1)
ZrꢀN(2)
ZrꢀN(3)
ZrꢀN(4)
ZrꢀCl
ZrꢀCp(c)
SiꢀN(1)
SiꢀC(1)
2.1245(12)
2.3024(13)
2.4511(14)
2.4190(13)
2.5376(4)
2.300(2)
1.7296(14)
1.873(2)
1.399(2)
The sums of the angles around the nitrogens of the
phenylenediamido fragment of the chelating ligand are
approximately 360° indicating sp2-hybridization at both
N atoms. The ZrꢀN(1) and ZrꢀN(2) distances in 3
N(1)ꢀC(8)
N(2)ꢀC(9)
N(2)ꢀC(14)
,
(2.124 and 2.302 A respectively) are comparable to
1.399(2)
1.432(2)
those of 2. The substitution of the ꢀNMe2 ligand in 2
by chloride which is not as good a p-donor renders the
metal center more electron deficient and allows coordi-
nation of an additional amine group. The zirco-
niumꢀamine bonds ZrꢀN(3) and ZrꢀN(4) in 3 are
similar in length to the ZrꢀN(4) distance in 2 and are
consistent with a ZrꢀN(sp3) single bond.
Bond angles
N(1)ꢀZrꢀN(2)
N(1)ꢀZrꢀN(3)
N(1)ꢀZrꢀN(4)
N(1)ꢀZrꢀCl
N(1)ꢀZrꢀCp(c)
N(2)ꢀZrꢀN(3)
N(2)ꢀZrꢀN(4)
N(2)ꢀZrꢀCl
N(2)ꢀZrꢀCp(c)
N(3)ꢀZrꢀN(4)
N(3)ꢀZrꢀCl
N(3)ꢀZrꢀCp(c)
N(4)ꢀZrꢀCl
N(4)ꢀZrꢀCp(c)
ZrꢀClꢀCp(c)
N(1)ꢀSiꢀC(1)
SiꢀN(1)ꢀZr
C(8)ꢀN(1)ꢀSi
C(8)ꢀN(1)ꢀZr
C(9)ꢀN(2)ꢀC(14)
C(9)ꢀN(2)ꢀZr
C(14)ꢀN(2)ꢀZr
SiꢀC(1)ꢀCp(c)
71.84(5)
92.33(5)
97.92(5)
158.74(4)
97.1
74.95(5)
74.19(5)
87.13(3)
168.9
142.42(5)
78.82(4)
107.1
78.73(4)
107.4
The isolation of complexes 2 and 3 supported the
idea that amine elimination is a viable route for the
synthesis of other transition metal complexes of 1.
However, our efforts to extend this chemistry to tanta-
lum have met with little success. For example, addition
of 1 to Ta(NMe2)5 produced a number of products that
1
had H-NMR spectra that were similar to those of 2
and 3, but these products could not be unequivocally
identified or separated. In contrast, reaction of 1 with
Ta(CH2Ph)3Cl2 did not proceed even under forced
conditions.
The synthesis of the trianionic mixed cyclopentadi-
enylꢀdiamido ligand 1 was achieved in good yield. The
amine elimination method was a viable route for at-
tachment of 1 to zirconium and resulted in the isolation
104.0
92.64(7)
108.93(6)
128.54(11)
122.50(10)
114.71(12)
115.78(10)
128.40(10)
156.5
a Cp(c) denotes the centroid of the cyclopentadienyl ring.
of complexes 2 and 3. The unexpected isolation of the
chloride complex 3 is likely to be due to presence of the
1·HCl adduct rather than NEt3HCl, both of which are
generated during the synthesis of 1. The crystal struc-
tures of both 2 and 3 indicate that the rigid trianionic
cyclopentadienylꢀbisamido ligand is not very sterically
demanding and leaves a large area of the Zr center
exposed. This behavior is evidenced by the strong coor-
dination of amine groups in both 2 and 3.
References
[1] (a) J. Okuda, Comments Inorg. Chem. 16 (1994) 185 and
references therein. (b) M.J. Bochman, J. Chem. Soc. Dalton
Trans. (1996) 255 and references therein. (c) P.J. Shapiro, E.
Bunel, W.P. Schaefer, J.E. Bercaw, Organometallics 9 (1990)
867. (d) P.J. Shapiro, W.D. Cotter, W.P. Schaefer, J.A.
Labinger, J.E. Bercaw, J. Am. Chem. Soc. 116 (1994) 4623. (e)
W. Herrmann, W. Baratta, J.A. Morawietz, J. Organomet.
Fig. 2. Thermal ellipsoid plot of the molecular structure of o-
C6H4{N(C6H5)N%[Si(CH3)2(C5(CH3)4)]}Zr(Cl)(NH(CH3)2)2 (3).

P. Doufou et al. / Journal of Organometallic Chemistry 603 (2000) 213–219
219
Chem. 497 (1995) C4. (f) J. Okuda, F. Schattermann, S.
Wocadlo, W. Massa, Organometallics 14 (1995) 789. (g) A.K.
Hughes, A. Meetsma, J.H. Teuben, Organometallics 12 (1993)
1936. (h) R. Spence, W.E. Piers, Organometallics 14 (1995) 4617.
(i) D.W. Carpenetti, L. Kloppenburg, J.T. Kupec, J.L. Petersen,
Organometallics 5 (1996) 1572. (j) J. Okuda, K.E. du Plooy, W.
Massa, H.C. Kang, U. Rose, Chem. Ber. 129 (1996) 275. (k)
Y. Mu, W.E. Piers, D.C. McQuarrie, M.J. Zaworotko, V.G.
Young, Organometallics 15 (1996) 2720 and references
therein. (l) K.E. du Plooy, U. Moll, S. Wocaldo, W. Massa,
J. Okuda, Organometallics 14 (1995) 3129. (m) Y.X. Chen,
T.J. Marks, Organometallics 16 (1997) 3649. (n) F. Amor,
T.P. Spaniol, J. Okuda, Organometallics 16 (1997) 4765. (o)
M.D. Fryzuk, P.B. Duval, S.H. Mao, S.J. Rettig, M.J. Za-
worotko, L.R. Macgillivray, J. Am. Chem. Soc. 121 (1999) 2478.
(p) M.D. Fryzuk, P.B. Duval, S.H. Mao S.J. Rettig, M.J.
Zaworotko, L.R. Macgillivray, J. Am. Chem. Soc. 121 (1999)
1707.
(b) M. Enders, R. Rudolph, H. Pritzkow, Chem. Ber. 129 (1996)
459. (c) P. Jutzi, J. Dahlhaus, Synthesis (1993) 684.
[4] (a) W.E. Piers, P.J. Shapiro, E. Bunel, J.E. Bercaw, Synlett 1
(1990) 74. (b) J.M. Canich, Eur. Patent Appl. EP-420-436-A1,
1991. (c) J.M. Canich, US Patent Appl. 5,026,798, 1991. (d) J.M.
Canich, G.G. Hlatky, H.W. Turner, PCT Appl. WO92-00333,
1992. (e) J.C. Stevens, F.J. Timmers, D.R. Wilson, G.F.
Schmidt, P.N. Nickias, R.K. Rosen, G.W. Knight, S. Lai, Eur.
Patent Appl. EP-416-815-A2, 1991.
[5] W.A. Hermann, W.J. Baratta, J. Organomet. Chem. 506 (1996)
357.
[6] G.M. Diamond, R.F. Jordan, J.L. Petersen, J. Am. Chem. Soc.
118 (1996) 8024.
[7] G.M. Sheldrick, SHELXTL, Nicolet XRD, Madison, WI, 1986.
[8] P. Jutzi, Chem. Rev. 86 (1986) 983.
[9] K. Hagen, C.J. Holwill, D.A. Rice, J.D. Runnacles, Inorg.
Chem. 27 (1988) 2032.
[10] K. Ayogi, K. Gantzel, K. Kalai, T.D. Tilley, Organometallics 15
(1996) 923.
[11] R.M. Pupi, J.N. Coalter, J.L. Petersen, J. Organomet. Chem.
497 (1995) 172.
[2] S. Ciruelos, T. Cuenca, P. Gomez-Sal, A. Manzreno, P. Royo,
Organometallics 14 (1995) 177.
[3] (a) P. Jutzi, U. Siemeling, J. Organomet. Chem. 500 (1995) 175.
.