Synthesis and characterization of planarly chiral nonbridged bis(1-R-indenyl) zirconium dichloride complexes and X-ray structure of structure fo rac-[(η5-C9H6-1-C(CH3)2-o-C6 H4-OCH3)2ZrCl2].THF

Article abstract of DOI:10.1016/S0022-328X(02)01197-X

Reactions of the lithium salts of 3-substituted indenes 1, 2 with ZrCl₄(THF)₂ gave two series of nonbridged bis (1-substituted)indenyl zircogone dichloride complexes. Fractional recrystallization from THF- petroleum ether furnished the pure racemic and mesomeric isomers of [(η⁵-C₉H₆-1-C(R¹)(R¹ = R² = CH₃, n = 1, rac-1a and meso-1b: R¹ = CH₃, R₂ = C₂H₅; n = 0.5 or O, rac-2a and meso-2b), respectively, Complex 1a was further characterized by X-ray diffraction to have a C₂ symmetrically racemic structure, where the six-member rings of the indenyl parts are oriented laterally and two o-CH₃O-C₆H₄-C(CH₃)₂-substituents are oriented to the open of the metallocene (Ind: bis-laterral, anti; Substituent: bis-centra, syn). The four zirconocene complexes are highly symmetrical in solution as characterized by room temperature ¹H-NMR, however ¹H-¹H NOESY of meso-1b shows that some of the NOE interactions arise from the two separated indenyl parts of the same molecule, which can only be explained well by taking into account the torsion isomers in solution.

Full text of DOI:10.1016/S0022-328X(02)01197-X

Journal of Organometallic Chemistry 650 (2002) 114–122
www.elsevier.com/locate/jorganchem
Synthesis and characterization of planarly chiral nonbridged
bis(1-R-indenyl) zirconium dichloride complexes and X-ray
structure of rac-[(h5-C9H6-1-C(CH3)2ꢀo-C6H4ꢀOCH3)2ZrCl2]·THF
Haiyan Ma, Jiling Huang, Yanlong Qian *
Laboratory of Organometallic Chemistry, East China Uni6ersity of Science and Technology, P.O. Box. 310, 130 Meilong Road,
Shanghai 200237, China
Received 15 November 2001; received in revised form 15 January 2002; accepted 16 January 2002
Abstract
Reactions of the lithium salts of 3-substituted indenes 1, 2 with ZrCl4(THF)2 gave two series of nonbridged bis(1-substi-
tuted)indenyl zirconocene dichloride complexes. Fractional recrystallization from THF–petroleum ether furnished the pure
racemic and mesomeric isomers of [(h5-C9H6-1-C(R1)(R2)ꢀo-C6H4ꢀOCH3)2ZrCl2]·nTHF (R1=R2=CH3, n=1, rac-1a and
meso-1b; R1=CH3, R2=C2H5; n=0.5 or 0, rac-2a and meso-2b), respectively. Complex 1a was further characterized by X-ray
diffraction to have a C2 symmetrically racemic structure, where the six-member rings of the indenyl parts are oriented laterally
and two o-CH3OꢀC6H4ꢀC(CH3)2ꢀ substituents are oriented to the open side of the metallocene (Ind: bis-lateral, anti; Substituent:
bis-central, syn). The four zirconocene complexes are highly symmetrical in solution as characterized by room temperature
1
1H-NMR, however H–1H NOESY of meso-1b shows that some of the NOE interactions arise from the two separated indenyl
parts of the same molecule, which can only be explained well by taking into account the torsion isomers in solution. © 2002
Published by Elsevier Science B.V.
Keywords: Zirconocene; Planar chirality; Nonbridged bis(1-R-indenyl) ligand; Crystal structure
as good examples of this point: bulky substituents
enabled these metallocenes to polymerize sections of
atactic and isotactic polypropylene in the same polymer
chain, which was thought to have been caused by the
interconversion of the molecular structure between
mesomeric and racemic conformers [10].
From features associated with the metal–indenyl tor-
sion potential, nonbridged bis(1-substituted)indenyl zir-
conocenes do not reveal fundamentally stereochemical
differences to ansa-metallocenes [11]; prochiral 1-substi-
tuted indenyl ligands can also yield two diastereoiso-
mers. R,R and S,S attachments give rise to the two
racemic enantiomers, whereas a R,S attachment gives
the meso-diastereoisomer [12]. It could be conceived
that nonbridged bis(1-substituted)indenyl zirconocene
complexes might also have some interesting properties
towards the polymerization of a-olefins. However, re-
search on the synthesis and application of such com-
1. Introduction
The discovery of the extremely active Group 4 bent
metallocene/aluminoxane catalysts has led to remark-
able research activities in homogenous catalytic a-olefin
polymerization [1–5]. The majority of this work has
been carried out with the ansa-metallocene derived
catalyst systems, since such bridged metallocene cata-
lysts produced highly isotactic or syndiotactic
polypropylene with very high activities [6–9]. It is
further proved that the stereoselectivity during the pro-
cess of polymerization is raised from the specific model
of the attachment of the Group 4 metal atom to the
enantiotropic R or S faces of the planarly prochiral
1-substituted indenyl ligand moieties; the intrinsic chi-
rality properties have no direct relation with the bridge-
part. Waymouth’s nonbridged and oscillating
bis(2-substituted) indenyl zirconocene complexes acted
plexes in polymerization is limited; only
a few
publications by Erker and a few other groups appeared
in the open literature when compared to ansa-metal-
locenes [11–16].
* Corresponding author. Tel./fax: +86-21-6470-2573.
E-mail address: qianling@online.sh.cn (Y. Qian).
0022-328X/02/$ - see front matter © 2002 Published by Elsevier Science B.V.
PII: S0022-328X(02)01197-X

H. Ma et al. / Journal of Organometallic Chemistry 650 (2002) 114–122
115
Scheme 1.
In this work we synthesized two bulky-group-substi-
tuted indenes 3-[o-CH3OꢀC6H4ꢀC(R1)(R2)]ꢀC9H6 (R1=
R2=CH3, 1; R1=CH3, R2=C2H5, 2), and used them
in the complexation with ZrCl4(THF)2. Two series of
racemic/mesomeric, nonbridged bis(1-substituted)-
indenyl zirconium dichloride complexes were obtained
and separated successfully through fractional recrystal-
lization. The investigation of their behaviors towards
the polymerization of a-olefin is still in progress.
large amount of THF had to be used. It seemed that
some aggregated structures might be formed during the
process of the rapid precipitation. By using less concen-
trated solution, all of them could be separated as
crystals, thus 1a and 1b were separated as THF-con-
taining complexes, and 2a was characterized as contain-
1
2
ing
solvate THF per molecule, however 2b was
isolated without THF.
It is noticeable that there is a chiral sp3 carbon in
ligand 2; after metallization by ZrCl4, the 1-substituted
indenyl leads to the planar chirality, thus in total there
are four stereogenic elements in the same molecule. The
formation of ten potential diastereoisomers (four pairs
of racemic enantiomers, and two mesomeric isomers) is
conceived. However, from the reaction, only three iso-
mers (rac-2a and meso-2b) were isolated. No other
isomer could be detected. Our previous results [18]
showed the coordination of zirconium to this ligand
was stereoselective, and the chirality in the a-carbon of
the side chain controlled the planar chirality of the
resulting complex1. According to this fact, we propose
the isomers formed in the reaction are those with the
ethyl group of the side chain locating against the six-
member ring of the indenyl ligand.
2. Results and discussion
2.1. Synthesis of zirconocene complexes
Nucleophilic attacks of the lithium reagents at the
positively polarized exo carbon of the corresponding
benzofulvenes prepared according to the literature
methods [17] and sequential hydrolysis gave the 3-subsi-
tuted
indenes,
3-[o-CH3OꢀC6H4ꢀC(R1)(R2)]ꢀC9H6
(R1=R2=CH3, 1; R1=CH3, R2=C2H5, 2) in high
yields, respectively. Then the substituted indenes 1, 2
were treated with n-butyllithium and 0.5 equivalent of
ZrCl4(THF)2 in sequence to afford mixture of rac- and
meso-diastereoisomers 1a/1b (rac:meso, about 55:45) or
2a/2b (rac:meso, about 60:40) in quite high yields
(Scheme 1). Fractional crystallization of the mixtures
from THF and petroleum ether fulfilled the separation
of pure rac- and meso-isomers. However the net yields
were quite low. The many fractional crystallization
steps undertaken for the purification of these zir-
conocenes with quite similar properties incurred a
heavy product-loss.
2.2. Characterization of zirconocene complexes
The rac- and meso-diastereoisomers 1a/1b or 2a/2b
are highly symmetrical in solution as characterized by
1H-NMR. Although torsion isomers were reported for
nonbridged bis(indenyl)zirconium complexes [11–14],
the ligand rotation is generally too fast to be detected
The solubility of the four zirconocene complexes in
THF was initially good, but upon precipitation of the
products from the solution, most of the cases as pow-
der, it became difficult to dissolve them again, and a
1 Zirconium
complex
[h5:h1-C9H6-1-C(CH3)(C2H5)ꢀo-C6H4ꢀ
OCH3]ZrCl3·THF was obtained as single pair of enantiomers ((pR,
S) and (pS, R)). Its crystal structure shows the ethyl group of the side
chain is located against the six-member ring of the indenyl ligand.

116
H. Ma et al. / Journal of Organometallic Chemistry 650 (2002) 114–122
Table 1
doublets through its coupling to protons H-2 and H-7
Selected 1H-NMR spectra data for the bis(1-R-indenyl)zirconium
1
(characterized by H–1H TOCSY); H-2 shows simple
dichloride complexes
doublet through the coupling with H-3.
Comparing with the reported rac- and meso-bis(1-
substituted)indenyl zirconocene complexes in Table 1,
we find the signals of H-2 and H-3 in 1a and 2a have
the same tendency as those in rac-bis(1-tBu-Ind)ZrCl2
(this complex also contains a bulky substituent of t-bu-
tyl); the H-2 signals of them are all upfield from H-3.
Furthermore, the H-2, H-3 signals in 1b, 2b and meso-
bis(1-tBu–Ind)ZrCl2 have the same tendency. Besides
these two points, the result of Grimmer and his co-
workers shows, the ‘‘inner pair’’ of H-2 and H-3 signals
usually belongs to the rac-isomers [13], thus we tenta-
tively assigned complexes 1a, 2a as racemic isomers,
and complexes 1b, 2b as mesomeric isomers. The abso-
lute stereochemical assignment for rac-1a was made
possible by X-ray crystallography (vide infra).
Metallocene
Isomer
H-2 H-3 Dl (H-2–H-3)
6.15 6.54 −0.39
1a a
1b
1c
2a
rac-
meso-
rac-
rac-
meso-
rac-
meso-
rac-
meso-
rac-
6.78 6.00
0.78
6.32 6.48 −0.16
6.14 6.63 −0.49
2b
6.88 6.07
5.91 5.90
6.33 5.37
0.81
0.01
0.96
b
Bis(1-iPr-Ind)ZrCl2
b
Bis(1-tBu-Ind)ZrCl2
5.89 6.48 −0.59
6.43 6.05
6.05 5.83
6.29 5.29
0.38
0.22
1.00
b
Bis(1-Bz-Ind)ZrCl2
meso-
a Assignment of stereochemistry made by X-ray crystallography.
b The data were cited from Ref. [13].
1
by room temperature scale H-NMR, the rac-isomers
In order to get more information about the stereo-
chemical structures, 1H–1H NOESY of meso-1b was
determined at room temperature. Fig. 1 shows clearly
the NOE interactions within the molecule of meso-1b,
H-2 and H-3 come into contact with H-4; the methyl
groups of side chain also come into contact with H-2,
H-3 and H-7. Our unpublished result [19] shows, in the
mixed Cp–Ind zirconocene complex [(h5-C9H6-1-
C(CH3)2ꢀo-C6H4ꢀOCH3)Cp]ZrCl2, only H-3 of the in-
are C2 symmetrical and the meso-isomer has CS sym-
metry in solution. The rac- and meso-diastereoisomeric
complexes of 1a/1b or 2a/2b give rise to two different
1
sets of H-NMR resonances of the same general spec-
troscopic type and appearance, respectively. However,
the rac- or meso-isomers can be readily distinguished
from each other by the differing position of their H-2
and H-3 spectroscopy signals. For all of the metal-
locenes, the H-3 signal was identified as a doublet of
Fig. 1. 1H–1H NOESY of meso-1b (CDCl3, 298 K, 400 MHz).

H. Ma et al. / Journal of Organometallic Chemistry 650 (2002) 114–122
117
Scheme 2.
Scheme 3.
the methyl groups of side chain and H-3%; thus there is
little possibility that the interligand NOE interactions
arise from this structure. However, the other two struc-
tures provide the possibility of these contacts, (1) in the
structure of meso-B, methyl groups of the side chain
could be located near to H-3%; (2) in the structure of
meso-C, H-2 could be located near H-4%, and the methyl
groups of the side chain could be located near H-3%; (3)
at the same time, most of these hydrogen atoms are
located in the narrow side of the bent structure; it is
imaginable that there will be NOE interactions. We
suggest here, although the Cs symmetric structure of
meso-1b is characterized by room temperature 1H-
NMR, the interligand NOE contacts existing possibly
in the torsion isomers could be detected under the same
condition.
When we attempted to characterize complex rac-1a
through the same method, it proved to be unsuccessful.
rac-1a was not stable enough to fulfil the determination
of 1H–1H NOESY in solution. During the measure-
ment, the H-signal of methoxyl groups had a significant
downshift from l3.15 to 3.89 gradually, and this trans-
formation achieved completion after 1 day. It seemed
that the methoxyl groups of the two substituents tended
to coordinate to the zirconium center [21], and formed
a new complex 1c2. The same tendency of coordination
denyl ligand comes into contact with H-4, no NOE
interaction exists between H-2 and H-4; and the methyl
groups of the side chain only come into contact with
H-2 and H-7. No NOE interaction exists between
methyl group and H-3 (adopting the same numbering
method as this work), thus we infer that the interac-
tions of H-2 and H-4, methyl groups and H-3 can only
arise from the two separated indenyl ligands within the
same molecule; they are interligand NOE contacts.
As for the rest of the NOE contacts, we of course
could not simply assign them as intraligand interac-
tions; to a certain extent, some of them may also arise
from the two separated indenyl ligands. Some relevant
results of NOE contacts concerning the interligand
interactions of two sterically hindered Cp-based zir-
conocene complexes were reported recently by Grim-
mond and his co-workers [20].
Only taking into account the Cs symmetric structure
of meso-1b in solution, it is difficult to explain the two
mentioned interligand NOE interactions. Generally, in
the Cs symmetric structure, H-2 is located far away
from H-4%, and methyl groups are also far away from
H-3%; no interaction could be imagined. This inconsis-
tency makes us go back to the torsion isomers. It is
reported that at least three torsion isomers may exist
for
the
meso-diastereoisomer
of
bis(1-substi-
tuted)indenyl metallocene complexes; their idealized
structures could be depicted in Scheme 2 [11,12]. meso-
1b should have similar torsion isomers in solution. In
the structure of meso-A, H-2 is found to be located far
away from the H-4% (and also H-2% and H-4), and so are
2 The coordinated complex 1c can be characterized by 1H-NMR (l,
CDCl3): 7.31 (m, 2H), 7.13 (m, 4H), 7.0 (m, 2H), 6.95 (m, 4H), 6.88
(t, 2H), 6.78 (m, 2H), 6.48 (dd, 2H), 6.32 (d, 2H), 3.89 (s, 6H), 1.49
(s, 6H), 1.08 (s, 6H). It is an average structure.

118
H. Ma et al. / Journal of Organometallic Chemistry 650 (2002) 114–122
was also found in complex rac-2a3. Compared with
other zirconocene complexes in Table 1, the H-2 and
H-3 signals of 1c have the same trend as those in
rac-1a, rac-2a and rac-bis(1-tBuꢀInd)ZrCl2; thus we
assigned its structure as racemic isomer (Scheme 3).
2.3. Crystal structure of rac-1a
Single crystals of rac-1a suitable for X-ray diffraction
were obtained from THF–petroleum ether (Fig. 2). The
crystal structure of rac-1a reveals the existence of two
unique molecules in a 1:1 ratio as racemic enantiomers,
which crystallize in the orthorhombic space group Pbcn
(c60) with a solvate THF molecule. Table 2 summa-
rizes the bonding parameters of it. rac-1a has C2 sym-
metry, with the zirconium atom in a pseudotetrahedral
environment formed by the two chlorine ligands and
two h5-coordinated indenyl ligands. The corresponding
bond lengths and angles of the two indenyl-parts in one
molecule are identical. The ZrꢀC bond lengths of five
,
member rings range between 2.427(5) and 2.696(4) A
,
(D=0.269 A). This scope is wider than the nonbridged
1- or 2-substituted bis(indenyl) zirconocene complexes
,
(D=0.1–0.16 A) reported in the literature [11–14,22–
24] and is closer to the ansa-bis(indenyl) zirconocene
,
complexes (D=0.18–0.25 A) [9]. The ZrꢀC distance to
Fig. 2. Solid state structure of rac-1a (hydrogen atoms are omitted
for clarity).
the substituted carbon is longer than the other Zrꢀ(CH)
distances (ZrꢀC(9)\ZrꢀC(8)\ZrꢀC(7)); but the
,
longest ZrꢀC distance is ZrꢀC(1) (2.696(4) A), where
Table 2
C(1) is adjacent to C(9) and belongs to the six-member
ring. It suggests that the bulky substituent of
ꢀC(CH3)2ꢀo-C6H4ꢀOCH3 group leads to the unsymmet-
rical bonding of Zr atom to the carbons of the five-
member ring significantly.
,
Selected bond lengths (A) and angles (°) for rac-1a
Bond lengths
ZrꢀCl
ZrꢀCl*
ZrꢀC(1)
ZrꢀC(6)
ZrꢀC(7)
ZrꢀC(8)
ZrꢀC(9)
C(9)ꢀC(10)
C(10)ꢀC(11)
C(12)ꢀO(1)
ZrꢀCp(cnt)
2.4183(14)
2.4183(14)
2.696(4)
2.546(4)
2.427(5)
2.519(5)
2.661(4)
1.535(6)
1.534(7)
1.386(7)
2.274
C(1)ꢀC(6)
C(6)ꢀC(7)
C(7)ꢀC(8)
C(8)ꢀC(9)
C(9)ꢀC(1)
C(1)ꢀC(2)
C(2)ꢀC(3)
C(3)ꢀC(4)
C(4)ꢀC(5)
C(5)ꢀC(6)
O(1)ꢀC(17)
1.436(6)
1.418(8)
1.390(7)
1.389(7)
1.432(7)
1.398(7)
1.370(8)
1.430(10)
1.322(9)
1.395(7)
1.425(7)
The ZrꢀCl distances of 2.4183(14) are within the
range observed previously for other bis(indenyl)-
zirconium dichlorides (2.407(1)–2.465(2) A) [11–
14,22–25]. However, the ClꢀZrꢀCl* angle of 101.43(8)°
is notably larger than the common value for the non-
bridged 1- or 2-substituted bis(indenyl) zirconium
,
,
dichloride complexes (91–98.98 A) and is close to the
ansa-metallocenes [11–14,22–24] (Table 3). Resconi et
al. reported that this value was a little sensitive to the
steric hindrance of the ligands [9c]. In our case, the
bigger value is most likely to be caused by the bulky
substituents at C(9) and C(9*) of indenyl fragments.
The bite angle of Cp(cnt)ꢀZrꢀCp%(cnt) in rac-1a is
131.97°, which is in the range of normal nonbridged
bis(1-subsituted)indenyl zirconocene complexes, and is
Bond angles
ClꢀZrꢀCl*
C(7)ꢀZrꢀC(7*)
C(9)ꢀZrꢀC(9*)
Cp(cnt)ꢀZrꢀCp%(cnt)
101.43(8)
78.5(2)
175.88(18)
131.97
C(1)ꢀZrꢀC(1*)
ClꢀZrꢀC(7)
ClꢀZrꢀC(9)
163.97(19)
126.66(16)
85.38(11)
much larger than the ansa-metallocene complexes. The
two bulky substituents seem to have less influence on
this factor, and also have less influence on the distance
of ZrꢀCp(cnt).
Erker et al. [11,12] have proposed three possible
solid-state conformers for nonbridged rac-bis(1-R-in-
denyl) zirconocene complexes (Scheme 4). From the top
view of the crystal structure of rac-1a (Fig. 3) it is
3 After being stored for a certain time (several months as solid
state), the proton signal of the methoxyl groups in rac-2a also had a
significant shift from l 3.06 to 3.88. However the new complex
1
cannot be characterized well by H-NMR, since the transformation is
not yet completed. Simple heating of the complex in solution cannot
promote this procedure.

H. Ma et al. / Journal of Organometallic Chemistry 650 (2002) 114–122
119
Table 3
Relevant bond lengths (A) and angles (°) of different nonbridge and ansa-zirconocene complexes
,
Zirconocene
Lengths
Angles
ZrꢀCl
ZrꢀCp(cnt)
ClꢀZrꢀCl
Cp(cnt)ꢀZrꢀCp%(cnt)
rac-1a
2.4183(14)
2.449(5)/2.434(4)
2.429(7)
2.274
2.224
2.231
–
2.218/2.242
2.240/2.237
2.400/2.233
2.314/2.234
2.232/2.240
2.532
101.43(8)
96.45(16)
98.98(2)
91.6(1)
94.6(1)
94.2(1)
131.97
132.72
128.66
129.6
129.0
130.1
130.9
134.7
131.2
132.5
117.62
118.3
117.4
rac-[(1-Et-Ind)2]ZrCl2 [13] (R,R)
rac-[(1-Ph-Ind)2]ZrCl2 [13]
[(1-Nem-Ind)2]ZrCl2 [11b] a
[(1-(p-F-Bz)-Ind)2]ZrCl2 [14]
[(2-(3,5-Me2-Ph-Ind)2)ZrCl2 [10a]
[(2-(3,5-(CF3)2-Ph-Ind)2)ZrCl2 [10a]
[(2-Men-Ind)2]ZrCl2 [22] b
2.424(1)
2.417(3)/2.429(3)
2.425(2)/2.430(2)
2.415(1)/2.407(1)
2.4238(9)/2.4337(9)
2.4289(10)/2.4429(9)
2.430(1)/2.462(1)
2.4122(12)
94.5(1)
93.98(3)
91.91(3)
95.84(5)
98.82(2)
98.69(9)
100.69(2)
[(2-Morp-Ind)2]ZrCl2 [23] a
rac-[CH2(3-t-Bu-1-Ind)2]ZrCl2 [9c]
rac-Me2C(3-t-Bu-1-Ind)2]ZrCl2 [9a]
rac-Me2C(3-Me3Si-1-Ind)2]ZrCl2 [9a]
2.262
–
–
2.405(2)
2.4205(4)
a Nem=neomenthyl; Men=menthyl; Morp=morpholino.
b Two independent structures in the same crystal.
Scheme 4.
1
other parts of the molecule. From the H-NMR spec-
troscopic data, the same conclusion could be made. The
proton signals of THF in rac-1a are consistent with a
free molecule, and so are those in the other two com-
plexes, meso-1b and rac-2a.
found that the two ꢀC(CH3)2ꢀo-C6H4ꢀOCH3 sub-
stituents are syn orientated over the open side of the
metallocene, and the two six-member rings of the in-
denyl parts are positioned laterally. This conformer is
consistent with the idealized structure of rac-B [11–13].
Furthermore, the aromatic parts of the two substituents
are orientated far away from the metallocene center
(Fig. 1), which is thought to reasonably reduce the
steric hindrance caused by syn orientation of the two
substituents.
From the crystallographic data of rac-1a, we found
that the strong steric hindrance caused by the two
bulky substituents of ꢀC(CH3)2ꢀo-C6H4ꢀOCH3 and
also by the two indenyl parts led to some structure
features, which are more similar to those of ansa-bis(in-
denyl) zirconium dichloride complexes. So it is sug-
gested that properly bulky substituents in the
five-member rings of nonbridged metallocene might
have similar influence on the metallocene structure as
the bridge part does on the ansa-metallocene complex,
which produces similar structure parameters.
The solvate THF molecule in rac-1a exists freely; no
efficient interaction could be found between it and
Fig. 3. The top view of rac-1a (hydrogen atoms are omitted for
clarity, with different numbering system as Fig. 2).

120
H. Ma et al. / Journal of Organometallic Chemistry 650 (2002) 114–122
3. Conclusions
mental analyses were performed with an EA-1106 mi-
croanalyzer. IR spectra were recorded in KBr using a
Nicolet 5SXC or Nicolet Magna-IR 550 instrument.
Mass spectra were carried out with a HP-5989A spec-
trometer at 70 eV. 1H-NMR, 1H–1H TOCSY and
NOESY were recorded on a Bruker 400MHz or Ad-
vance 500MHz instruments using CDCl3 as solvent.
o-Bromoanisole [25], 6,6%-dimethyl-benzofulvene, and
6,6%-methyl,ethyl-benzofulvene [17] were prepared ac-
cording to the literature procedures.
Bulky-group-substituted indene ligands of 3-[o-
CH3OꢀC6H4ꢀC(R1)(R2)]ꢀC9H6 (R1=R2=CH3, 1;
R1=CH3, R2=C2H5, 2) were synthesized and their
lithiated salts reacted with ZrCl4(THF)2 to give corre-
spondingly nonbridged bis(1-substituted)indenyl zirco-
nium dichloride complexes. The rac- and meso-isomers
1a/1b or 2a/2b could be separated and purified by
fractional recrystallization from THF and petroleum
ether. Although the chiral sp3 carbon in the side chain
of ligand 2 was introduced, only one set of rac/meso-
zirconocene complexes could be detected and separated.
The chirality in the a-carbon of the side chain con-
trolled the planar chirality of the resulting complex,
which led to the reduction of isomers.
The absolute structure of 1a was determined success-
fully by X-ray diffraction to be the racemic isomers; the
two ꢀC(CH3)2ꢀo-C6H4ꢀOCH3 substituents in it are syn
orientated over the open side of the bent structure, and
the two six-member rings of the indenyl parts are
positioned laterally. Furthermore, the aromatic parts of
the two substituents are orientated far away from the
metallocene center, which reasonably reduces the steric
hindrance caused by syn orientation of the two sub-
stituents. 1b is the mesomeric isomer. According to the
spectroscopic characters of these zirconocene com-
plexes, the racemic nature of 2a and mesomeric nature
of 2b were assigned.
4.1. Synthesis of 3-[o-CH3OꢀC6H4ꢀC(CH3)2]ꢀC9H6 (1)
To the solution of o-bromoanisole (0.143 mol, 26.7
g) in 100 ml n-hexane, was added dropwise the solution
of n-BuLi (0.143 mol, 75 ml) in hexane. Plenty of white
precipitates were produced. The obtained white precipi-
tates of o-CH3OC6H4Li were filtrated and dissolved in
Et2O.
A solution of 6,6%-dimethyl-benzofulvene (0.123 mol,
19.2 g) in 50 ml Et2O was then added dropwise at room
temperature (r.t.) to the above lithium solution. The
mixture was stirred overnight and then hydrolyzed.
After work-up, 1 was obtained as pale orange crystals
in 71% (23 g), m.p.: 89–91 °C. 1H-NMR (l, ppm,
CDCl3): 7.39 (m, 2H), 7.19 (ddd, 1H, 3J=8.1 Hz,
4
3
3J=7.4 Hz, J=1.68 Hz), 7.03 (td, 1H, J=7.4 Hz,
4
3J=7.4 Hz, J=1.0 Hz), 6.95 (m, 2H), 6.77 (dd, 1H,
4
3
3J=8.1 Hz, J=1.1 Hz), 6.70 (dd, 1H, J=7.7 Hz,
The four zirconocene complexes are highly symmetri-
4J=1.0 Hz), 6.34 (t, 1H, 3J=2.2 Hz), 3.46 (s, 3H), 3.35
3
cal in solution as characterized by room temperature
(d, 2H, J=2.2 Hz), 1.72 (s, 6H). MS (m/z): 264 (13,
1H-NMR, but H–1H NOESY of meso-1b shows that
M ), 149 (100, M −C9H7 ). IR (cm−1, KBr): 3070m,
3000m, 2985m, 2970m, 1600m, 1580m, 1485s, 1460s,
1435s, 1396m, 1375m, 1355m, 1290m, 1245s, 1150m,
1085m, 1050m, 1025s, 770s, 750s, 725s. Anal. Calc. for
C19H20O: C, 86.32; H, 7.63. Found: C, 86.37; H, 7.50%.
1
+
+
’
some of the NOE interactions arise from the two
separated indenyl parts of the same molecular, which
can only be explained well by taking into account the
torsion isomers in solution.
The racemic zirconocene complexes 1a and 2a are
not stable enough structurally. The methoxyl groups of
the side chain tended to coordinate to the metal center
in solid state or in solution, which caused a significant
shift of the proton signals of methoxyl groups from
3.15 to 3.89, or from 3.06 to 3.88, respectively. The
coordinated product of rac-1a was well characterized
4.2. Synthesis of
3-[o-CH3OꢀC6H4ꢀC(CH3)(C2H5)]ꢀC9H6 (2)
Ligand 2 was synthesized similarly to 1 except that
6,6%-methyl,ethyl-benzofulvene was used. Pale orange
crystals were obtained in 75%, m.p.: 58–60 °C. 1H-
NMR (l, ppm, CDCl3): 7.39 (m, 2H), 7.19 (ddd, 1H,
1
by H-NMR.
3
4
3J=8.1 Hz, J=7.4 Hz, J=1.68 Hz), 7.03 (td, 1H,
4
3J=7.4 Hz, J=1.0 Hz), 6.94 (m, 2H), 6.76 (dd, 1H,
4. Experimental
3J=8.1 Hz, 4J=1.2 Hz), 6.65 (d, 1H, 3J=7.7 Hz),
6.33 (t, 1H, 3J=2.2 Hz), 3.39 (s, 3H), 3.36 (d, 2H,
3
All reactions were carried out under an inert atmo-
sphere of dry Ar using standard Schlenk techniques.
Reaction vessels were flame dried under vacuum then
filled with Ar several times. Tetrahydrofuran, Et2O,
petroleum ether and n-hexane were distilled under Ar
from purple or blue sodium/benzophenone ketyl ac-
cording to the solvents prior to use. Dichloromethane
was distilled from anhydrous P2O5 prior to use. Ele-
3J=2.2 Hz), 2.42 (m, 1H, J=7.4 Hz), 2.11 (m, 1H,
3J=7.4 Hz), 1.63 (s, 3H), 0.70 (t, 3H, J=7.4 Hz). MS
3
+
+
’
(m/z): 278 (25, M ), 163 (100, M −C9H7 ). IR
(cm−1, KBr): 3070m, 2980m, 2955m, 1599m, 1580m,
1485s, 1465s, 1445s, 1390m, 1380m, 1370m, 1305m,
1285s, 1245s, 1180m, 1150m, 1090m, 1025s, 980m,
920w, 770s, 760s, 725s. Anal. Calc. for C20H22O: C,
86.29; H, 7.96. Found: 86.07; H, 7.88%.

H. Ma et al. / Journal of Organometallic Chemistry 650 (2002) 114–122
121
4.3. Synthesis of rac-[(p5-C9H6-1-C(CH3)2ꢀo-C6H4ꢀ
OCH3)2ZrCl2]·THF (rac-1a) and meso-[(p5-C9H6-
1-C(CH3)2ꢀo-C6H4ꢀOCH3)2ZrCl2]·THF (meso-1b)
5.08 mmol), and n-BuLi (2.3 ml, 5.08 mmol) were used.
After fractional crystallization from THF and
petroleum ether, analytical products rac-2a and meso-
2b were obtained. Rac-2a: orange needle crystal, yield
10%, m.p.: 206–210 °C. 1H-NMR (l, ppm, CDCl3):
To the solution of ZrCl4(THF)2 (1.619 g, 4.29 mmol)
in 50 ml THF, was added at r.t. the solution of lithium
salt prepared from ligand 1 (2.264 g, 8.58 mmol) in 80
ml THF and n-BuLi (7.6 ml, 8.58 mmol) in n-hexane.
The reaction mixture was stirred for 24 h and then the
solvent was removed in vacuo to give an orange solid.
It was extracted with 100 ml CH2Cl2 and filtrated. The
solvent of the filtrate was removed again under vacuum
to give 3.1 g of rac-1a/meso-1b mixture. Fractional
recrystallization of this solid from THF and petroleum
ether afforded pure rac-1a and meso-1b sequentially.
rac-1a: orange prismatic crystal, 9.1% (300 mg), m.p.:
3
4
7.63 (m, 2H-4), 7.51 (dd, 2H-8, J9,8=7.8 Hz, J10,8
=
1.6 Hz), 7.11–7.08 (m, 2H-7, 2H-10), 7.01 (m, 2H-5,
2H-6), 6.94 (td, 2H-9, 3J10,9=7.8 Hz, 3J8,9=7.8 Hz,
3
4J9,11=1.1 Hz), 6.63 (d, 2H-3, J2,3=3.3 Hz), 6.54 (dd,
3
4
2H-11, J10,11=8.1 Hz, J9,11=1.1 Hz), 6.14 (d, 2H-2,
3J3,2=3.3 Hz), 3.75 (t, 2H-19, J20,19=6.6 Hz), 3.06 (s,
3
6H-14), 2.54 (m, 2H-15, 3J13,15=7.4 Hz), 2.20 (m,
3
2H-15, J13,15=7.4 Hz), 1.98 (s, 6H-12), 1.85 (p, 2H-20,
3J19,20=6.6 Hz), 0.55 (t, 6H-13, J15,13=7.4 Hz). MS
3
(m/z): 664 (56, M+−1/2THFꢀCH3Cl), 585 (100, M+
−1/2THFꢀ2CH3ClꢀEt ). IR (cm−1, KBr): 3070w,
’
1
215–216 °C. H-NMR (l, ppm, CDCl3): 7.66 (m, 2H-
2980s, 2960s, 2880m, 2835m, 1599m, 1580m, 1485s,
1460s, 1440s, 1380m, 1350w, 1330w, 1280m, 1248s,
1200w, 1160w, 1150m, 1095m, 1060m, 1040m, 1030s,
800s, 755s. Anal. Calc. for C40H42Cl2O2Zr·1/2THF
(753.01): C, 67.00; H, 6.16. Found: C, 67.04; H, 6.44%.
meso-2b: yellow crystal, yield 24%, m.p.: 206–207 °C.
1H-NMR (l, ppm, CDCl3): 7.52 (d, 2H-4, 2H-8), 7.14
3
4
4), 7.51 (dd, 2H-8, J9,8=7.8 Hz, J10,8=1.6 Hz), 7.21
3
3
(m, 2H-7), 7.10 (ddd, 2H-10, J11,10=8.1 Hz, J9,10
=
4
7.8 Hz, J8,10=1.6 Hz), 7.06 (m, 2H-5, 2H-6), 6.93 (td,
2H-9, 3J8,9=7.8 Hz, 3J10,9=7.8 Hz, 4J11,9=1.1 Hz),
3
4
6.60 (dd, 2H-11, J10,11=8.1 Hz, J9,11=1.1 Hz), 6.54
3
5
(dd, 2H-3, J2,3=3.3 Hz, J7,3=0.8 Hz), 6.15 (d, 2H-2,
3J3,2=3.3 Hz), 3.75 (t, 4H-19, J20,19=6.5 Hz), 3.15 (s,
(ddd, 2H-10, J11,10=8.1 Hz, J9,10=7.8 Hz, J8,10=
3
3
3
4
6H-14), 2.06 (s, 6H-12), 1.90 (s, 6H-13), 1.85 (p, 4H-20,
1.55 Hz), 7.00–6.96 (m, 2H-7, 2H-5), 6.93–6.90 (m,
2H-6, 2H-9), 6.88 (d, 2H-2, J3,2=3.3 Hz), 6.55 (dd,
3
3J19,20=6.5 Hz). MS (m/z): 651 (50.10, M+−THF–
+
3
4
’
’
Cl ), 357 (100, M −THF–L ꢀCH3ClꢀCH4). IR (KBr,
cm−1): 3070w, 2980s, 2930s, 2880m, 2840m, 1599m,
1580m, 1485s, 1460s, 1440s, 1380m, 1365m, 1290m,
1248s, 1150m, 1090m, 1060m, 1030s, 805m, 755s. Anal.
Calc. for C42H46Cl2O3Zr (760.96): C, 66.29; H, 6.09.
Found: C, 66.25; H, 6.25%.
2H-11, J10,11=8.1 Hz, J9,11=1.1 Hz), 6.07 (d, 2H-3,
3J2,3=3.3 Hz), 3.01 (s, 6H-14), 2.86 (m, 2H-15,
3J13,15=7.4 Hz), 2.37 (m, 2H-15, J13,15=7.4 Hz), 1.90
3
3
(s, 6H-12), 0.71 (t, 6H-13, J15,13=7.4 Hz). MS (m/z):
+
+
’
664 (7, M −CH3Cl), 585 (57, M −2CH3ClꢀEt ). IR
(KBr, cm−1): 3100w, 3070w, 3030w, 2970s, 2935s,
2880m, 2830m, 1599m, 1580m, 1485s, 1460s, 1445s,
1380m, 1350w, 1330w, 1290m, 1248s, 1180w, 1150m,
1095m, 1040m, 1030s, 1000m, 870w, 790s, 755s. Anal.
Calc. for C40H42Cl2O2Zr (716.91): C, 67.02; H, 5.91.
Found: C, 66.93; H, 6.20%.
meso-1b: yellow needle crystal, 12% (400 mg), m.p.:
185–187 °C. 1H-NMR (l, ppm, CDCl3): 7.56 (dd,
2H-8, 3J9,8=7.8 Hz, 4J10,8=1.5 Hz), 7.48 (d, 2H-4,
3J5,4=8.4 Hz), 7.16 (ddd, 2H-10, 3J11,10=8.1 Hz,
4
3
3J9,10=7.8 Hz, J8,10=1.5 Hz), 7.08 (d, 2H-7, J6,7
=
8.4 Hz), 7.03–6.94 (m, 2H-5, 2H-6, 2H-9), 6.78 (d,
3
3
2H-2, J3,2=3.3 Hz), 6.63 (dd, 2H-11, J10,11=8.1 Hz,
4.5. X-ray crystallography
4J9,11=1.0 Hz), 6.00 (d, 2H-3, J3,1=3.3 Hz), 3.75 (t,
3
4H-19, 3J20,19=6.4 Hz), 3.13 (s, 6H-14), 1.99 (s, 6H-12),
The crystal data of rac-1a were collected on a Rigaku
AFC-7R single crystal diffractometer at 293 K using
3
1.98 (s, 6H-13), 1.85 (p, 4H-20, J10,20=6.4 Hz). MS
(m/z): 636 (16, M+−THFꢀCH3Cl), 586 (100, M+−
THFꢀ2CH3Cl). IR (KBr, cm−1): 3070w, 2980s, 2930s,
2880m, 2840m, 1599m, 1580m, 1485s, 1460s, 1440s,
1380m, 1365m, 1290m, 1248s, 1150m, 1090m, 1050m,
1030s, 810m, 790s, 755s. Anal. Calc. for C42H46Cl2O3Zr
(760.96): C, 66.29; H, 6.09. Found: C, 65.58; H, 6.13%.
Mo–Ka radiation (u=0.71069 A, graphite monochro-
,
matized, scan type ꢀ/2q). Intensities were corrected for
Lorentz-polarization effect; an empirical psi-scans cor-
rection was applied (0.7521–1.0000). Unit cell parame-
ters were initially obtained from 25 reflections
(14.50BqB22.43). The structure was solved primarily
by the direct methods using SHELXS-97 system and
refined by full-matrix least-squares on F2 using all of
the reflections with SHELXL-97 program. Hydrogen
atoms were solved geometrically and refined with mixed
methods. Crystal data for rac-1a: C38H38Cl2O2Zr·
C4H8O, M=760.95, prismatic orange crystal; crystal
dimensions, 0.20×0.22×0.25 mm; orthorhombic, a=
4.4. Syntheses of rac-[(p5-C9H6-1-C(CH3)(C2H5)ꢀ
o-C6H4ꢀOCH3)2ZrCl2]·1/2THF (rac-2a) and meso-
[(p5-C9H6-1-C(CH3)(C2H5)ꢀo-C6H4ꢀOCH3)2ZrCl2]
(meso-2b)
The synthetic method was similar. Except that
,
ZrCl4(THF)2 (0.9577 g, 2.54 mmol), ligand 2 (1.4108 g,
12.910(3), b=12.622(4), c=23.040(4) A, V=

122
H. Ma et al. / Journal of Organometallic Chemistry 650 (2002) 114–122
3
(b) J.A. Even, R.L. Lones, A. Razavi, J.D. Ferrara, J. Am.
Chem. Soc. 110 (1988) 6255.
[9] (a) L. Resconi, F. Piemontesi, I. Camurati, O. Sudmeijer, I.E.
Nifant’ev, P.V. Ivchenko, L.G. Kuz’mina, J. Am. Chem. Soc.
120 (1998) 2308;
,
3754.4(16) A , space group Pbcn (c60), Z=4, Dcalc
=
1.346 g cm−3, F(000)=1584.00, v(Mo–Ka)=0.209
cm−1, T=293 K; of 3268 reflections measured, all are
independent. Number of reflections and parameters in
refinement are 3268, and 219. Goodness-of-fit, and the
weighted R factor based on F2 are 1.035 and 0.1834;
final R inicate based on F for all data is 0.0859.
(b) M. Toto, L. Cavallo, P. Corradini, G. Moscardi, L. Resconi,
G. Guerra, Macromolecules 31 (1998) 3431;
(c) L. Resconi, D. Balboni, G. Baruzzi, C. Fiori, S. Guidotti, P.
Mercandelli, A. Sironi, Organometallics 19 (2000) 420.
[10] (a) G.W. Coates, R.M. Waymouth, Science 267 (1995) 217;
(b) E. Hauptman, R.M. Waymouth, J.W. Ziller, J. Am. Chem.
Soc. 117 (1995) 11586;
5. Supplementary material
(c) R. Kravchenko, A. Masood, R.M. Waymouth, Organometal-
lics 16 (1997) 5909;
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Centre, CCDC No. 164332 for complex rac-1a.
Copies of this information may be obtained free of
charge from The Director, CCDC, 12 Union Road,
Cambridge CB2 1EZ, UK (Fax: +44-1223-336033;
e-mail: deposit@ccdc.cam.ac.uk or www: http://
www.ccdc.cam.ac.uk). 1H–1H TOCSY and NOESY
spectra of meso-1b are available.
(d) M.D. Bruce, G.W. Coates, E. Hauptman, R.M. Waymouth,
J.W. Ziller, J. Am. Chem. Soc. 119 (1997) 11174;
(e) C.D. Tagge, R.L. Kravchenko, T.K. Lal, R.M. Waymouth,
Organometallics 119 (1999) 380.
[11] (a) G. Erker, B. Temme, J. Am. Chem. Soc. 114 (1992) 4004;
(b) G. Erker, M. Aulbach, M. Knickmeier, D. Wingbermu¨hle, C.
Kruger, M. Nolte, S. Werner, J. Am. Chem. Soc. 115 (1993)
4590.
[12] C. Kru¨ger, F. Lutz, M. Notle, G. Erker, M. Aulbach, J.
Organomet. Chem. 452 (1993) 79.
[13] N.E. Grimmer, N.J. Coville, C.B. de Koning, J.M. Smith, L.M.
Cook, J. Organomet. Chem. 616 (2000) 112.
[14] G. Jany, M. Gustafsson, T. Repo, E. Aitola, J.A. Dobado, M.
Klinga, M. Leskela¨, J. Organomet. Chem. 553 (1998) 173.
[15] G.Y. Lee, M. Xue, M.S. Kang, O.C. Kwon, J.-S. Yoon, Y.-S.
Lee, H.S. Kim, H. Lee, I.-M. Lee, J. Organomet. Chem. 558
(1998) 11.
[16] M. Knickmeier, G. Erker, T. Fox, J. Am. Chem. Soc. 118 (1996)
9623.
[17] (a) K.J. Stone, R.D. Little, J. Org. Chem. 49 (1984) 1849;
(b) H.G. Alt, M. Jung, G. Kehr, J. Organomet. Chem. 562
(1998) 153.
Acknowledgements
We gratefully acknowledge the financial support
from Special Funds for Major State Basic Research
Projects (G1999064801) and National Natural Science
Foundation of China (20072004).
[18] H. Ma, J. Huang, Y. Qian, Inorg. Chem. Comm. 4 (2001) 515.
[19] H. Ma, J. Huang, Y. Qian, unpublished results.
[20] B.J. Grimmond, J.Y. Corey, N.P. Rath, Organometallics 18
(1999) 404.
References
[1] A.L. McKnight, R.M. Waymouth, Chem. Rev. 98 (1998) 2587.
[2] S.D. Ittel, L.K. Johnson, M. Brookhart, Chem. Rev. 100 (2000)
1169.
[3] H.G. Alt, A. Ko¨ppl, Chem. Rev. 100 (2000) 1205.
[4] G.W. Coates, Chem. Rev. 100 (2000) 1223.
[5] L. Resconi, L. Cavallo, A. Fait, F. Piemontesi, Chem. Rev. 100
(2000) 1253.
[21] (a) A.A.H. van der Zeijden, C. Mattheis, Organometallics 16
(1997) 2651;
(b) A.A.H. van der Zeijden, C. Mattheis, J. Organomet. Chem.
584 (1999) 274;
(c) A.H. van der Zeijden, C. Mattheis, R. Fro¨hlich, Chem.
Ber./Recl. 130 (1997) 1231.
[22] R.L. Halterman, D.R. Fahey, E.F. Bailly, D.W. Docker, O.
Stenzel, J.L. Shipman, M.A. Khan, S. Dechert, H. Shumann,
Organometallics 19 (2000) 5464.
[6] (a) H. Schnutenhaus, H.H. Brintzinger, Angew. Chem. Int. Ed.
Engl. 18 (1979) 777;
(b) F.R.W.P. Wild, L. Zsolnai, G. Huttner, H.H. Brintzinger, J.
Organomet. Chem. 232 (1982) 233.
[23] S. Knu¨ppel, J.-L. Faure´, G. Erker, G. Kehr, M. Nissinen, R.
Fro¨hlich, Organometallics 19 (2000) 1262.
[7] (a) H. Sinn, W. Kaminsky, Adv. Organomet. Chem. 18 (1980)
99;
(b) W. Kaminsky, K. Ku¨lper, H.H. Brintzinger, F.R.W.P. Wild,
Angew. Chem. 97 (1985) 507.
[8] (a) J.A. Even, J. Am. Chem. Soc. 106 (1984) 6355;
[24] R. Leino, H.J.G. Luttikhedde, A. Lehtonen, R. Sillanpa¨a¨, A.
Penninkangas, J. Strande´n, J. Mattinen, J.H. Na¨sman, J.
Organomet. Chem. 558 (1998) 171.
[25] B.S. Furniss, et al., Vogel’s Textbook of Practical Organic
Chemistry, 4th ed., Longman, 1987.