Ca-mediated styrene polymerization: Tacticity control by ligand design

Article abstract of DOI:10.1002/ejic.200700802

New heteroleptic benzylcalcium complexes with modified fluorenyl ligands were prepared which include a complex with a chelating dimethylamino substituent, [9-(2-Me₂N-ethyl)fluorenyl][2-Me₂N-α- Me₃Si-benzyl]Ca (4), and a series of complexes with bulky substituents like tBu-, Ph(Me)₂C-, Me(Ph)₂C- and Me(4-tBuC₆H₄)₂C- in the 2- and 7-positions of the fluorenyl ligand (complexes 5 through 9, respectively). Crystal structures of 4, [(2,7-tBu-9-Me₃Si-fluorenyl)(2-Me₂N-benzyl)Ca] ₂ (5) and [2,7-Ph(Me)₂C-9-Me₃Si-fluorenyl][2- Me₂N-α-Me₃Si-benzyl]Ca·THF (6) show that modification of the fluorenyl ligand hardly affects the coordination mode of the benzylic ligand. Also, in solution all compounds are of heteroleptic nature and it was shown that fluorenyl modifications hardly affect the barriers for inversion at the chiral benzylic carbon atom [18.7(2)-19.2(2) kcal/mol]. Complex 4 does not initiate styrene polymerization, which underscores the importance of a free coordination site at calcium. Complexes 5-9 initiate styrene polymerization, and the syndiotacticity of the obtained polymers increases with the bulkiness of the substituents on the fluorenyl ligands. Syndiotacticities up to r = 95 % (rr = 90 %) were obtained. Wiley-VCH Verlag GmbH & Co. KGaA, 2007.

Full text of DOI:10.1002/ejic.200700802

FULL PAPER
DOI: 10.1002/ejic.200700802
Ca-Mediated Styrene Polymerization: Tacticity Control by Ligand Design
Dirk F.-J. Piesik,[a] Karin Häbe,[b] and Sjoerd Harder*[a]
Keywords: Alkaline-earth metals / Calcium / Polymerization catalysis / Fluorene ligands / Polystyrene
New heteroleptic benzylcalcium complexes with modified
fluorenyl ligands were prepared which include a complex
with a chelating dimethylamino substituent, [9-(2-Me2N-
ethyl)fluorenyl][2-Me2N-α-Me3Si-benzyl]Ca (4), and a series
of complexes with bulky substituents like tBu-, Ph(Me)2C-,
Me(Ph)2C- and Me(4-tBuC6H4)2C- in the 2- and 7-positions
of the fluorenyl ligand (complexes 5 through 9, respectively).
Crystal structures of 4, [(2,7-tBu-9-Me3Si-fluorenyl)(2-Me2N-
benzyl)Ca]2 (5) and [2,7-Ph(Me)2C-9-Me3Si-fluorenyl][2-
Me2N-α-Me3Si-benzyl]Ca·THF (6) show that modification of
the fluorenyl ligand hardly affects the coordination mode of
the benzylic ligand. Also, in solution all compounds are of
heteroleptic nature and it was shown that fluorenyl modifica-
tions hardly affect the barriers for inversion at the chiral ben-
zylic carbon atom [18.7(2)–19.2(2) kcal/mol]. Complex 4 does
not initiate styrene polymerization, which underscores the
importance of a free coordination site at calcium. Complexes
5–9 initiate styrene polymerization, and the syndiotacticity of
the obtained polymers increases with the bulkiness of the
substituents on the fluorenyl ligands. Syndiotacticities up to
r = 95% (rr = 90%) were obtained.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
Introduction
Syndiotactic polystyrene is a relatively young polymeric
material[1,2] which receives much industrial as well as aca-
demic interest.[3–9] Its many advantageous properties,[10,11]
which include a high melting point (273 °C), fast crystalli-
zation rate, high modulus, low density, low dielectric con-
stant and excellent solvent resistance, have led to its success-
ful application as a specialty polymer.[12]
We recently introduced a new class of polymerization ini-
tiators based on calcium which produced syndiotactic poly-
styrene by a chain-end-controlled monomer insertion
(Scheme 1).[13–14] The initial catalyst systems consist of a
polymerization-active benzyl group and a passive spectator
ligand (generally a fluorenyl ligand). The monomeric (1)
Scheme 1.
be controlled by monomer concentration,[14] thus controls
the tacticity of the polymer (Scheme 1).
[14]
and more active dimeric (2)[15] forms are well-defined
A slight improvement in stereoregularity could also be
achieved by lowering the polymerization temperature.[14]
Despite continuous efforts however, no significant improve-
ment was obtained by variation of the spectator ligand.
Earlier attempts to increase the syndiospecificity in styrene
polymerization were based on increasing the steric bulk of
the spectator ligand; that is, our initial catalyst (1) was
modified by the introduction of a bulky hypersilyl substitu-
ent (3).[16] This would enforce a stronger interaction be-
tween the chiral polymer chain-end and the incoming
monomer, and thus improve stereocontrol. The bulkier cat-
alyst 3, however, gave polystyrene of very similar tacticity
but much shorter chain length. It was reasoned that the
use of bulky substituents did not only result in improved
communication between polymer and monomer, but also
led to more stereoerrors by retarding monomer insertion
and thus allowing inversion of the chiral chain-end.
complexes which have been structurally characterized by X-
ray diffraction.
These initiators, which can be regarded as a cross-breed
between organolithium initiators and cationic TiIII half-
sandwich catalysts,[2,6] combine the advantages of a living
polymerization with those of a stereoselective polymeriza-
tion. Syndiotactic polystyrene, however, was only obtained
for polymerization in neat styrene. This is due to the pre-
dominant ionicity and inherent weakness of the Ca–C
bond, which allows for inversion of the chiral chain-end.
The ratio between insertion and inversion rates, which can
[a] Anorganische Chemie, Universität Duisburg–Essen,
Universitätsstraße 5, 45117 Essen, Germany
Fax: +49-201-1832621
E-mail: sjoerd.harder@uni-essen.de
[b] Laboratorium für Anorganische Chemie, ETH Hönggerberg,
8093 Zürich, Switzerland
5652
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 5652–5661

Ca-Mediated Styrene Polymerization
ligand with a potential chelating dimethylamino substituent
was prepared. Deprotonation by (o-Me2N-α-Me3Si-
benzyl)2Ca·(THF)2 gave the heteroleptic benzylcalcium
complex 4.
Furthermore, the influence of steric bulk in remote posi-
tions was investigated by the introduction of large substitu-
ents in the 2- and 7-positions of the fluorenyl ligand. Sev-
eral 2,7-disubstituted fluorene ligands were prepared and
converted into heteroleptic benzylcalcium catalysts. In one
instance, this was done by reacting the substituted fluorene
with (o-Me2N-benzyl)2Ca, in which case a dimeric ben-
zylcalcium complex was typically formed (5).[15] In other
cases, reaction with (o-Me2N-α-Me3Si-benzyl)2Ca·(THF)2
gave monomeric benzylcalcium initiators (6–9). We were
able to obtain crystal structures of 4 as well as representa-
tive dimeric and monomeric benzylcalcium complexes (5
and 6).
The heteroleptic benzylcalcium complex 4 crystallizes
with two independent, but geometrically similar, molecules
In a further attempt to understand the factors that gov- in the asymmetric unit (Figure 1). The fluorenyl rings coor-
ern the stereoselectivity in calcium-mediated styrene poly- dinate in a slightly distorted η5-fashion in which the Ca–C
merizations, we prepared a series of new benzylcalcium cat- bond lengths vary between 2.630(3) and 2.782(3) Å. The
alysts with modified fluorenyl ligands.
average Ca–C(fluorenyl) bond of 2.710(3) Å is slightly
shorter than that observed in 1 [2.743(3) Å]. Likewise, the
C,N-bidentate benzyl ligand in 4 [Ca–C1 2.504(3) Å; Ca–
N2 2.463 Å] is coordinated in a similar fashion to that in 1
[Ca–C 2.506(2) Å; Ca–N 2.466(2) Å]. Whereas the coordi-
nation sphere of Ca in 1 is saturated by an additional THF
ligand, complex 4 crystallizes solvent-free with intramolecu-
lar coordination of the Me2N arm.
Results and Discussion
1. Syntheses and Structures of Modified Heteroleptic
Benzylcalcium Complexes
In order to investigate the importance of the metal’s co-
ordination sphere in the polymerization reaction, a fluorene
Eur. J. Inorg. Chem. 2007, 5652–5661
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
5653

D. F. -J. Piesik, K. Häbe, S. Harder
FULL PAPER
The heteroleptic benzylcalcium complex 6 (Figure 3)
crystallizes similarly to complex 1. The fluorenyl–Ca coor-
dination geometry is slightly distorted from η5-coordina-
tion. The Ca–C bond lengths range from 2.631(2) to
2.843(2) Å. The average Ca–C(fluorenyl) bond length of
2.734(2) Å compares well to that observed in 1 [2.743(3) Å].
Also, the C,N-bidentate benzyl ligand in 6 binds similarly;
the Ca–C [2.517(3) Å] and Ca–N [2.496(2) Å] bond lengths
for the C,N-bidentate benzyl ligand are only slightly longer
than those in 1 [Ca–C: 2.506(2) Å; Ca–N: 2.466(2) Å,
respectively]. In addition, the Ca–O(THF) bonds are very
similar [6: 2.331(2) Å; 1: 2.307(2) Å]. Therefore, the coordi-
nation geometry around Ca is hardly influenced by the
large Ph(Me)2C substituents in the 2- and-7 positions of the
fluorenyl ring.
Figure 1. Crystal structure of 4. Only one of the two crystallo-
graphically independent, but very similar, molecules is shown. Ex-
cept for the benzylic hydrogen, all hydrogen atoms have been omit-
ted for clarity. Average selected bond lengths [Å] and angles [°]: Ca–
C18 2.508(3), Ca–N1 2.487(2), Ca–N2 2.460(2); Ca–C(fluorenyl):
2.630(3), 2.658(3), 2.727(3), 2.739(3), 2.782(3); C18–Ca–N1
113.6(1), C18–Ca–N2 71.4(1), N1–Ca–N2 112.0(1).
The heteroleptic complex 5 (Figure 2) crystallizes as a
crystallographically C2-symmetric dimer with symmetri-
cally bridging benzyl ligands and terminal fluorenyl ligands.
Its crystal structure is very much comparable to that of di-
meric 2 in which the 2- and 7-positions of the fluorenyl ring
are not substituted. The fluorenyl ring in 5 coordinates to
Ca in η5-fashion with Ca–C bond lengths in the range of
2.698(2) to 2.800(2) Å. The average Ca–C bond length of
2.750(2) Å is equal to that in 2 [2.750(3) Å]. Also, both sym-
metrically bridging benzyl anions are bound very similarly;
the Ca–C17 [2.572(2) Å] and Ca–N1 [2.568(2) Å] bond
lengths in 5 are equal to those in 2 [2.561(3) Å and
2.569(3) Å, respectively] within standard deviation. The
bulky tBu substituents in the 2- and-7 positions therefore
do not affect the bonding of ligands to Ca.
Figure 3. Crystal structure of 6. Except for the benzylic hydrogen,
all hydrogen atoms have been omitted for clarity. Selected bond
lengths [Å] and angles [°]: Ca–C17 2.517(3), Ca–N1 2.496(2), Ca–
O1 2.331(2); Ca–C(fluorenyl): 2.631(2), 2.647(2), 2.768(2), 2.779(2),
2.843(2); C17–Ca–N1 70.95(8), C17–Ca–O1 99.63(8), N1–Ca–O1
113.37(6).
The complexes dissolve well in aromatic solvents (except
for dimer 5 which shows lower solubility), and their hetero-
leptic solid-state structures are retained in solution. The
monomeric complexes with the chiral o-Me2N-α-Me3Si-
benzyl ligand (4 and 6–9) display fluorenyl ligands with dia-
stereotopic sides at room temperature. Fast exchange by in-
version at the chiral benzylic carbon atom can be ac-
complished by heating the sample or by the addition of
THF; both processes have been discussed previously.[14] For
some of the complexes, an activation barrier for this inver-
sion process could be determined. These values for 4 [Tcoal
= 105 °C, ∆G‡ = 19.2(2) kcal/mol] and 6 [Tcoal = 85 °C, ∆G‡
= 18.7(2) kcal/mol] are, within standard deviation, equal to
that measured for 1 [Tcoal = 90 °C, ∆G‡ = 18.8(2) kcal/mol].
Therefore, like in the crystal structures, the bonding of the
benzyl ligands to calcium is neither affected by the replace-
ment of THF in 1 by a chelating arm, nor by the introduc-
tion of bulky substituents in the 2- and 7-positions of the
fluorenyl ring.
Figure 2. Crystal structure of 5. Except for the benzylic hydrogen
atoms, all hydrogen atoms have been omitted for clarity. Selected
bond lengths [Å] and angles [°]: Ca–C17 2.572(2), Ca–C17Ј
2.572(2), Ca–N1 2.568(2); Ca–C(fluorenyl): 2.698(2), 2.705(2),
2.749(2), 2.796(2), 2.800(2); Ca···CaЈ 3.5074(6), C17–Ca–C17Ј
93.95(6), C17–Ca–N1 68.08(5), C17Ј–Ca–N1 120.08(5).
5654
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 5652–5661

Ca-Mediated Styrene Polymerization
The benzylcalcium complex 5 is also a heteroleptic dimer
All of the heteroleptic complexes 5 through 9 were found
in benzene solution. Proof of its dimeric nature comes from to be active in styrene polymerization. Polymerizations were
1
the unusual H NMR chemical shifts for the aromatic ring generally run under two different sets of conditions: a) poly-
1
protons. The H NMR spectra of anionic benzyl species merization of a 10% solution of styrene in cyclohexane at
generally show a high-field triplet, characteristic for the 50 °C and b) polymerization of neat styrene at 20 °C. All
aromatic ring proton in the para-position with respect to polymers showed a unimodal MW distribution with
the benzylic carbon.[17,18] For 5, the most high-field aro- poly-dispersion-indexes (D) between 2 and 3 and a Gold
matic signal is a doublet at δ = 5.81 ppm, which can be distribution,[20,21] that is, tailing in the low molecular weight
assigned to a proton in the meta-position with respect to range (Table 1). The latter is typical of heteroleptic ben-
the benzylic carbon. This anomaly can be explained by the zylcalcium initiators and is, as we have shown before, re-
anisotropy effect caused by ring currents[19] in a nearby aryl lated to a slow initiation by the stabilized carbanion rather
ring: the meta-H is positioned in the shielding cone of the than to chain termination.[13–16] These rather broad molecu-
neighbouring ring. Similar observations for the analogous lar weight distributions, however, hinder the syntheses of
complex 2 have been discussed extensively.[15]
well-defined block-copolymers.
The tacticity of all polymers was analyzed by 13C NMR
spectroscopy, and tacticities were calculated using Bernouil-
lain statistics based on the new heptade assignment we re-
cently reported.[22] Figure 4a shows the 13C NMR signals
for the Cipso atom in the phenyl rings of the different poly-
styrenes obtained in solution at 50 °C. Under these condi-
tions, polymer produced by initiator 1 is essentially atactic
(the signals in the mr and rr regions are only slightly high
relative to those from an atactic polymer produced by an
alkyllithium initiator). Polymer obtained with initiator 5,
that is, an initiator with additional tBu groups in the 2-
and 7-positions of the fluorenyl ring, showed a noticeable
increase in syndiotactity (r = 87%). The 13C NMR signal
for the Cipso atom merely consists of four singlets which
represent the rrrrrr heptade and the three erroneous hep-
tades formed after a single stereochemically wrong insertion
(mrrrrr, rmrrrr and rrmrrr). Substitution of the tBu groups
for the more bulky Ph(Me)2C substituents (6) again leads
to improvement of the polymer’s syndiotacticity; the largest
signal is that of the rrrrrr heptade (r = 88%). Further in-
crease of steric bulk by the introduction of (Ph)2MeC sub-
stituents (7) gave a polymer of 92% syndiotacticity in diads.
Subsequent increase of steric bulk by substitution of the
Me3Si group for Et3Si (8) or by additional tBu substituents
on the Ph rings (9) gave no significant improvement of the
tacticity.
In conclusion, neither the introduction of an intramolec-
ularly coordinating amine ligand, nor substitution of the
fluorenyl ligand in the 2- and 7-positions, changes the struc-
tures of heteroleptic benzylcalcium complexes 1 and 2. In
fact, the structures and all bond lengths remain surprisingly
equal. The compositions of these complexes in solution are
basically similar to those in the solid state, and modification
of the fluorenyl ligands does not affect the dynamic behav-
iour of these complexes significantly.
2. Styrene Polymerization with Modified Heteroleptic
Benzylcalcium Complexes
Complex 4, the benzylcalcium complex with a fluorenyl
ligand containing a chelating amine arm, was found to be
completely inactive as an initiator in styrene polymeriza-
tion. Whereas dissociation of the THF ligand in initiator 1
generates a free site for monomer coordination, the strongly
bound intramolecular Me2N substituent in 4 effectively kills
the reactivity of this benzylcalcium complex towards sty-
rene. This underscores the importance of precoordination
of the styrene monomer in the insertion step. Effective ben-
zylcalcium catalysts should therefore have a relatively open
coordination sphere around the calcium metal. This obser-
vation is in line with the fact that large bulky groups in the
9-position, that is, close to the metal, substantially retard
monomer insertion.[16] Therefore, modification of the fluor-
enyl spectator ligand in the periphery instead of close to the
metal coordination site was pursued.
Spurred by the advantageous effect of steric bulk on the
syndioselectivity of styrene polymerization in solution, we
Table 1. Styrene polymerization with initiators 1, 5–9.
Initiator
T [°C]
[m][a] [molL–1]
t [min]
Mn [gmol–1]
D
rr [%]
r [%]
1[14]
5
50
50
50
50
50
50
20
20
20
20
20
20
–20
0.10
0.10
0.10
0.10
0.10
0.10
neat
neat
neat
neat
neat
neat
neat
45
45
60
60
50
45
60
40
1.115ϫ105
1.449ϫ105
2.014ϫ105
1.515ϫ105
2.377ϫ105
2.440ϫ105
0.960ϫ105
2.213ϫ105
1.235ϫ105
1.302ϫ105
3.022ϫ105
1.506ϫ105
1.870ϫ105
2.306
2.549
1.947
2.394
2.377
2.030
2.350
2.570
2.265
2.635
2.134
2.984
4.293
Ͻ50
76
77
85
85
85
79
83
85
86
88
88
91
Ͻ70
87
88
92
92
92
89
91
92
93
94
94
95
6
7
8
9
1[14]
5
6
45
60
95
7
8
9
50
9
3840[b]
[a] Concentration of monomer in cyclohexane. [b] 64 h.
Eur. J. Inorg. Chem. 2007, 5652–5661
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
5655

D. F. -J. Piesik, K. Häbe, S. Harder
FULL PAPER
Figure 4. 13C NMR signals of the phenyl Cipso atom in the polystyrene materials ([D2]tetrachloroethane, T = 100 °C). a) Polymers obtained
with the various initiators 1, 5–9; 10% styrene solutions in cyclohexane; 50 °C. b) Polymers obtained with the various initiators 1, 5–9;
100% styrene; 20 °C. The syndiotactic rrrrrr heptade is marked by ϫ, and the error heptades mrrrrr, rmrrrr and rrmrrr by o.
also performed polymerization studies in pure styrene (Fig-
ure 4b). Polymers obtained under these conditions generally
show much higher syndiotacticities than the polymers pro-
duced by polymerization in solution.[14–16] This is partly due
to a temperature effect (for polymerizations in solution, im-
proved syndiotacticity is also observed at lower tempera-
ture; Figure 5), but is also related to a higher ratio of the
insertion to inversion rates on account of the higher mono-
mer concentration (Scheme 1). The positive effect of in-
creasing steric bulk in the 2- and 7-positions of the fluor-
enyl ring on the syndiotacticity of the polymer is also no-
ticeable for polymerizations in neat styrene. It is, however,
less extreme than that observed for polymerizations in solu-
tion. The most syndiotactic polymer was obtained with the
most bulky calcium initiator 9 at –20 °C (r = 95%, rr =
90%); however, rather slow polymerization and a broad
molecular weight distribution were observed (Table 1).
The data for polymerization in solution in particular
Figure 5. 13C NMR signals of the phenyl Cipso atom in the poly-
show a clear increase in stereoselectivity upon increase in
styrene materials ([D2]tetrachloroethane, T = 100 °C) obtained
with initiator 5 under various conditions.
the steric bulk of the substituents in the periphery of the
ligand system. The reason for this trend is unclear. Figure 6
shows space-filling models of fluorenyl–Ca units containing
differently substituted ligand systems. It is evident from the
structural studies (vide supra) that substitution in the 2- the fluorenyl ligand restrict the dynamic behaviour of the
and 7-positions of the fluorenyl ligand has a negligible ef- polymer chain attached to calcium. Reducing the degrees
fect on the coordination sphere of the Ca2+ ion. It can only of freedom of the polymer chain and the coordinated sty-
be speculated that the sterically large groups at the sides of rene would lead to a higher selectivity in the insertion step.
5656
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 5652–5661

Ca-Mediated Styrene Polymerization
Experimental Section
General Comments: All experiments were carried out under argon
using predried solvents and Schlenk techniques. The following
starting materials were prepared according to literature: [(2-Me2N-
α-Me3Si-benzyl)]2Ca·(THF)2,[13] [9-(2-dimethylamino)ethyl]fluo-
rene,[23] 2,7-di-tert-butylfluorene[24] and 2,7-bis(1-methyl-1-phenyle-
thyl)fluorene.[25] Me3SiCl was freed from HCl by distillation from
dry N,N-diethylaniline. Other reagents were commercially available
and used as received. NMR spectra were recorded with Bruker
DPX300 and DRX500 machines.
Synthesis of 2,7-Bis(1,1-diphenylethyl)fluorene: 2,7-Dibenzoylfluo-
rene was prepared according to a modified literature procedure.[25]
A mixture of fluorene (10.0 g, 60.2 mmol), benzoic acid (16.0 g,
131 mmol) and polyphosphoric acid (200 g) was heated to 140 °C
in a conical flask for 50 h. The cooled mixture was treated carefully
with ice/water and, in later stages, with a hot aqueous potassium
hydroxide solution. The heated suspension was filtered, and the
residue dissolved in toluene. After aqueous workup and drying of
the organic layer, the solution was concentrated and 2,7-dibenzo-
ylfluorene precipitated at room temperature (10.7 g, 48%), m.p.
193 °C. 1H NMR (300 MHz, CDCl3, 25 °C): δ = 8.01 (s, 2 H, fluor-
enyl), 7.96–7.82 (m, 8 H, Ar), 7.62–7.58 (m, 2 H, Ar), 7.51–7.46
(m, 4 H, Ar), 4.04 (s, 2 H, CH2) ppm. 13C{1H} NMR (300 MHz,
CDCl3): δ = 196.6 (C=O), 144.5, 144.2, 137.9, 136.9, 132.3, 130.0,
129.7, 128.3, 126.9, 120.4, 36.9 (CH2) ppm. C27H18O2 (374.4):
calcd. C 86.61, H 4.85; found C 86.62, H 4.81.
Figure 6. Space-filling models of the fluorenyl–Ca unit with in-
creasing steric bulk of the fluorenyl ligand (top to bottom) as in
the initiators 1, 5 and 6. Left: view perpendicular to the fluorenyl
plane. Right: view along the fluorenyl plane.
Conclusions
Replacing the THF ligand in 1 with an intramolecular
chelating amine ligand does not affect the coordination of
the o-Me2N-α-Me3Si-benzyl ligand. Also, the introduction
of bulky substituents in the 2- and 7-positions of the fluor-
enyl ligand has no influence on ligand coordination to the
A twofold excess of phenylmagnesium bromide (14.5 g, 80.0 mmol)
was added to
a solution of 2,7-dibenzoylfluorene (7.80 g,
metal. In all cases, very similar fluorenyl–Ca and benzyl– 20.8 mmol) in toluene (100 mL) and was heated at reflux for 4 h.
After hydrolysis with 2  HCl and extraction with toluene, the or-
ganic layer was dried. 2,7-Bis(hydroxydiphenylmethyl)fluorene was
crystallized from a concentrated solution at –22 °C (7.35 g, 67%),
m.p. 253 °C. 1H NMR (300 MHz, CDCl3, 25 °C): δ = 7.63 (d, 3JH,H
= 8.0 Hz, 2 H, fluorenyl), 7.41 (s, 2 H, fluorenyl), 7.32–7.18 (m, 22
H, Ar), 3.74 (s, 2 H, CH2), 2.78 (s, 2 H, OH) ppm. 13C{1H} NMR
(300 MHz, CDCl3): δ = 147.4, 146.1, 143.8, 140.8, 128.3, 128.3,
127.6, 127.3, 124.9, 119.6, 82.6 (COH), 37.5 (CH2) ppm. C39H30O2
(530.7): calcd. C 88.27, H 5.70; found C 88.26, H 5.90.
Ca coordination geometries are observed.
The composition and structures of the benzylcalcium
complexes 4–9 in solution (benzene, toluene) are similar to
those observed in the solid state; that is, the Schlenk equi-
libria lie completely towards the heteroleptic side, and the
chiral benzylic carbon is configurationally stable on the
NMR time scale (the fluorenyl ligands show diastereotopic
sides). Also, the dynamic inversion of the chiral benzylic
carbon at higher temperatures is not affected by the ligand
modifications.
On the other hand, a significant effect of fluorenyl ligand
modifications was observed on the polymerization of sty-
rene. First of all, substitution of THF in 1 for an intramo-
lecularly coordinated Me2N substituent effectively kills the
To a suspension of 2,7-bis(hydroxydiphenylmethyl)fluorene (7.35 g,
13.8 mmol) in toluene (100 mL) and acetic acid (0.2 mL) was added
a
trimethylaluminum solution (2.0  in toluene, 35 mL,
70.0 mmol). After being heated at reflux for 4 h, the mixture was
poured into ice/water. The remaining product was dissolved in con-
centrated hydrochloric acid (20 mL), and the aqueous layer was
reactivity of this benzylcalcium complex towards styrene. extracted with dichloromethane. The organic solution was dried,
and the solvent was removed. 2,7-Bis(1,1-diphenylethyl)fluorene
Apparently, the strongly bound chelate arm prevents the
formation of a free coordination site, which effectively
blocks the polymerization reaction. Secondly, an increase in
the steric bulk of the substituents in the periphery of the
fluorenyl ligand resulted in greater syndioselectivity in sty-
rene polymerization.
These are important observations from which it can be
deduced that these benzylcalcium complexes likely function
as single-site catalysts in which precoordination of the
monomer and subsequent insertion are the key steps in sty-
rene polymerization. These new initiators allow syndioselec-
tive styrene polymerization not only in neat styrene, but
also in solution. The broad molecular weight distributions,
however, disable syntheses of well-defined block-copoly-
mers.
was obtained as a white solid product (5.75 g, 79%), m.p. 217 °C.
3
1H NMR (300 MHz, CDCl3, 25 °C): δ = 7.60 (d, JH,H = 8.1 Hz,
2 H, fluorenyl), 7.25–6.88 (m, 24 H, Ar), 3.71 (s, 2 H, CH2), 2.22
(s, 6 H, CH3) ppm. 13C{1H} NMR (300 MHz, CDCl3): δ = 149.8,
148.2, 143.7, 139.8, 129.3, 128.3, 128.1, 126.4, 125.8, 119.5, 53.1
[(CCH3)2], 37.6 (CH2), 31.2 (CH3) ppm. C41H34 (526.7): calcd. C
93.49, H 6.51; found C 93.16, H 6.45.
Synthesis of 2,7-Bis[1,1-bis(4Ј-tert-butylphenyl)ethyl]fluorene: The
first intermediate, 2,7-bis(4Ј-tBu-benzoyl)fluorene, was prepared
analogously to 2,7-dibenzoylfluorene (vide supra). After quenching
the reaction with water and aqueous potassium hydroxide, the resi-
due was extracted with dichloromethane, dried and isolated by re-
moval of the solvent. Crystallization from acetone at –27 °C gave
2,7-bis(4Ј-tBu-benzoyl)fluorene (26.8 g, 43%) as a white solid, m.p.
251 °C. 1H NMR (300 MHz, CDCl3, 25 °C): δ = 8.05 (s, 2 H, fluor-
Eur. J. Inorg. Chem. 2007, 5652–5661
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
5657

D. F. -J. Piesik, K. Häbe, S. Harder
FULL PAPER
3
enyl), 7.96–7.88 (m, 4 H, fluorenyl), 7.81 (d, JH,H = 8.1 Hz, 4 H,
(CH3), –2.81 (Me3Si) ppm. C34H38Si (474.8): calcd. C 86.02, H
3
Ar), 7.53 (d, JH,H = 8.3 Hz, 4 H, Ar), 4.05 (s, 2 H, CH2), 1.39 (s,
8.07; found C 85.96, H 7.94.
18 H, CH3) ppm. 13C{1H} NMR (300 MHz, CDCl3): δ = 197.0
(C=O), 156.9, 145.2, 144.9, 138.0, 135.8, 130.8, 130.3, 127.6, 126.0,
121.1, 37.7 (CH2), 35.9 (CCH3), 31.9 (CH3) ppm. C35H34O2
(486.6): calcd. C 86.38, H 7.04; found C 85.99, H 7.18.
Synthesis of 2,7-Bis(1,1-diphenylethyl)-9-Me3Si-fluorene: This com-
pound was synthesized according to the preparation of 2,7-
bis(tBu)-9-Me3Si-fluorene. A white solid precipitated from acetone
at –28 °C (88%), m.p. 143 °C. 1H NMR (300 MHz, CDCl3, 25 °C):
3
δ = 7.63 (d, JH,H = 7.4 Hz, 2 H, fluorenyl), 7.25–7.03 (m, 24 H,
A twofold excess of 4-tBu-phenylmagnesium bromide (19.9 g,
84.0 mmol) was added to a solution of 2,7-bis(4Ј-tBu-benzoyl)fluo-
rene (10.0 g, 20.5 mmol) in toluene (100 mL), and was heated at
reflux for 6.5 h. After the reaction mixture was hydrolyzed with 2 
HCl, extracted with diethyl ether and dried, the solvent was re-
moved. 2,7-Bis[hydroxybis(4Ј-tBu-phenyl)methyl]fluorene crys-
tallized from hexane at room temperature as a white solid (11.5 g,
Ar), 3.57 (s, 1 H, CH), 2.18 (s, 6 H, CH3), –0.40 (s, 9 H, Me3Si)
ppm. 13C{1H} NMR (300 MHz, CDCl3): δ = 149.6, 146.9, 145.6,
138.0, 128.8, 127.8, 125.9, 125.7, 124.9, 119.1, 52.7 (CCH3), 42.7
(CH), 30.7 (CH3), –2.84 (Me3Si) ppm. C44H42Si (598.9): calcd. C
88.24, H 7.07; found C 87.38, H 7.22.
Synthesis of 2,7-Bis[1,1-bis(4Ј-tBu-phenyl)ethyl]-9-Me3Si-fluorene:
This compound was synthesized according to the preparation of
2,7-bis(tBu)-9-Me3Si-fluorene. A white solid precipitated from tol-
uene at –28 °C (79%), m.p. 247 °C. 1H NMR (300 MHz, CDCl3,
1
74%), m.p. 249 °C. H NMR (300 MHz, CDCl3, 25 °C): δ = 7.56
3
(d, JH,H = 8.0 Hz, 2 H, fluorenyl), 7.43 (s, 2 H, fluorenyl), 7.26–
7.08 (m, 18 H, Ar), 3.73 (s, 2 H, CH2), 2.72 (br s, 2 H, OH), 1.23
(s, 36 H, CH3) ppm. 13C{1H} NMR (300 MHz, CDCl3): δ = 150.0,
145.8, 144.2, 143.3, 140.3, 127.6, 126.8, 124.8, 124.4, 119.0, 81.9
(COH), 37.7 (CH2), 34.5 [C(CH3)3], 31.3 [C(CH3)3] ppm. C55H62O2
(755.1): calcd. C 87.49, H 8.28; found C 87.07, H 7.97.
3
25 °C): δ = 7.62 (d, JH,H = 8.1 Hz, 2 H, fluorenyl), 7.21–6.90 (m,
20 H, Ar), 3.51 (s, 1 H, CH), 2.14 (s, 6 H, CCH3), 1.22 [s, 36 H,
C(CH3)3], –0.48 (s, 9 H, Me3Si) ppm. 13C{1H} NMR (300 MHz,
CDCl3): δ = 148.4, 147.4, 146.6, 145.4, 137.9, 128.3, 125.4, 125.2,
124.6, 118.9, 51.8 (CCH3), 42.5 (CH), 34.3 [C(CH3)3], 31.4 [C-
(CH3)3], 30.4 (CCH3), –2.75 (Me3Si) ppm. C60H74Si (823.3): calcd.
C 87.53, H 9.06; found C 87.28, H 9.02.
To a suspension of 2,7-bis[hydroxybis(4Ј-tBu-phenyl)methyl]fluo-
rene (4.80 g, 6.36 mmol) in toluene (100 mL) and acetic acid
(0.2 mL) was added a trimethylaluminum solution (2.0  in tolu-
ene, 12.5 mL, 25.0 mmol). After heating the mixture for 5 h, the
solution was poured into ice/water. After aqueous workup and dry-
ing of the organic layer, the solvent was removed. 2,7-Bis[1,1-bis(4Ј-
tBu-phenyl)ethyl]fluorene was obtained as a white solid after wash-
Synthesis of 2,7-Bis(1,1-diphenylethyl)-9-Et3Si-fluorene: n-Butyllith-
ium (2.45  in hexane, 1.88 mL, 4.61 mmol) was added to a pre-
cooled (–70 °C), stirred solution of 2,7-bis(1,1-diphenylethyl)fluo-
rene (2.20 g, 4.18 mmol) in THF (50 mL). After being slowly
heated to –10 °C and stirred over a period of 30 min at this tem-
perature, the solution was recooled (–70 °C), and chlorotriethyl-
silane (0.78 mL, 4.66 mmol) was added. The cooling bath was re-
moved, and the solution was stirred at room temperature for 20 h.
After aqueous workup, the solvent was removed. 2,7-Bis(1,1-di-
phenylethyl)-9-Et3Si-fluorene precipitated from acetone at –28 °C
as a pale yellow solid (2.26 g, 84%), m.p. 103 °C. 1H NMR
1
ing with hexane (3.92 g, 82%), m.p. 311 °C. H NMR (300 MHz,
3
CDCl3, 25 °C): δ = 7.54 (d, JH,H = 8.1 Hz, 2 H, fluorenyl), 7.23–
7.18 (m, 10 H, Ar), 7.01–6.95 (m, 10 H, Ar), 3.69 (s, 2 H, CH2),
2.13 (s, 6 H, CH3), 1.24 (s, 36 H, CH3) ppm. 13C NMR{1H}
(300 MHz, CDCl3): δ = 148.4, 148.1, 146.4, 143.1, 139.2, 128.3,
127.6, 125.2, 124.6, 118.8, 51.8 (CCH3), 37.1 (CH2), 34.3 [C-
(CH3)3], 31.4 [C(CH3)3], 30.6 (CCH3) ppm. C57H66 (751.1): calcd.
C 91.14, H 8.86; found C 90.61, H 8.78.
3
(300 MHz, CDCl3, 25 °C): δ = 7.65 (d, JH,H = 8.1 Hz, 2 H, fluor-
enyl), 7.21–7.00 (m, 22 H, Ar), 6.93 (s, 2 H, fluorenyl), 3.69 (s, 1
H, CH), 2.17 (s, 6 H, CH3), 0.51 [t, 9 H, Si(CH2CH3)3], 0.08 [q, 6
H, Si(CH2CH3)3] ppm. 13C{1H} NMR (300 MHz, C6D6): δ =
150.6, 150.3, 147.8, 146.6, 139.2, 126.7, 126.7, 126.5, 126.1, 120.1,
53.5 (CCH3), 39.8 (CH), 31.3 (CH3), 8.00 [Si(CH2CH3)3], 3.00
[Si(CH2CH3)3]. C47H48Si (641.0): calcd. C 88.07, H 7.55; found C
87.88, H 7.34.
Synthesis of 2,7-Di-tert-butyl-9-(trimethylsilyl)fluorene: n-Butyllith-
ium (2.5  in hexane, 12.0 mL, 30.0 mmol) was added to a pre-
cooled (–70 °C), stirred solution of 2,7-di-tert-butylfluorene[24]
(7.70 g, 27.7 mmol) in THF (50 mL). The solution was slowly
warmed to –10 °C and stirred for 30 min at this temperature, then
cooled again to –70 °C, and chlorotrimethylsilane (4.1 mL,
32.2 mmol) was added. The cooling bath was removed, and the
solution was stirred at room temperature for 24 h. After aqueous
workup, the solvent was removed under vacuum, and colourless
needles were recovered (8.65 g, 89%), m.p. 145 °C. 1H NMR
Synthesis of 4: A solution of [9-(2-dimethylamino)ethyl]fluorene[23]
(734 mg,
(THF)2
3.09 mmol)
and
[(2-Me2N-α-Me3Si-benzyl)]2Ca·
[13]
(1.87 g, 3.13 mmol) in toluene (20 mL) was heated to
85 °C over a period of 5 h. All volatiles were removed, and the
product was dried under vacuum to yield complex 4 with one mole-
cule of THF. Drying of this product under vacuum over a period
of 30 min (90 °C, 1 Torr) and cooling a toluene solution thereof to
–30 °C gave yellow-orange plates of 4 (1.09 g, 73%) which are only
soluble in hot benzene. NMR spectroscopic data are given for 4
with one equivalent of THF; m.p. 132 °C. 1H NMR (300 MHz,
3
(300 MHz, CDCl3, 25 °C): δ = 7.74 (d, JH,H = 8.0 Hz, 2 H, fluor-
3
enyl), 7.53 (s, 2 H, fluorenyl), 7.38 (d, JH,H = 8.0 Hz, 2 H, fluor-
enyl), 3.83 (s, 1 H, CH), 1.42 (s, 18 H, CH3), –0.04 (s, 9 H, Me3Si)
ppm. 13C{1H} NMR (300 MHz, CDCl3): δ = 148.6, 145.6, 137.9,
122.2, 121.0, 119.0, 42.5 (CH2), 34.8 [C(CH3)3], 31.7 [C(CH3)3],
–2.60 (Me3Si) ppm. C24H34Si (350.6): calcd. C 82.22, H 9.77; found
C 81.82, H 9.94.
3
C6D6, 25 °C): δ = 8.33 (d, JH,H = 8.1 Hz, 1 H, fluorenyl), 8.23 (d,
Synthesis of 2,7-Bis(1-methyl-1-phenylethyl)-9-(trimethylsilyl)fluo-
rene: This compound was synthesized according to the preparation
of 2,7-bis(tBu)-9-Me3Si-fluorene. Colourless needles crystallized
3JH,H = 8.0 Hz, 1 H, fluorenyl), 7.69 (d, 3JH,H = 8.3 Hz, 1 H, fluor-
3
3
enyl), 7.56 (t, JH,H = 8.1 Hz, 1 H, fluorenyl), 7.45 (d, JH,H
=
8.4 Hz, 1 H, fluorenyl), 7.29–7.22 (m, 2 H, fluorenyl), 7.10–6.99
1
3
from acetone at –28 °C (72%), m.p. 122 °C. H NMR (300 MHz,
(m, 2 H, Ar and fluorenyl), 6.80 (t, JH,H = 7.1 Hz, 1 H, Ar), 6.35
3
3
3
CDCl3, 25 °C): δ = 7.71 (d, JH,H = 7.9 Hz, 2 H, fluorenyl), 7.32–
(d, JH,H = 7.0 Hz, 1 H, Ar), 6.25 (t, JH,H = 7.1 Hz, 1 H, Ar),
7.15 (m, 14 H, Ar), 3.71 (s, 1 H, CH), 1.77 (s, 6 H, CH3), 1.75 (s, 3.13–3.09 (m, 2 H, CH2), 2.89 (br., 4 H, THF), 2.84–2.76 (m, 1 H,
6 H, CH3), –0.23 (s, 9 H, Me3Si) ppm. 13C{1H} NMR (300 MHz,
CDCl3): δ = 151.3, 148.4, 145.6, 137.8, 127.9, 126.8, 125.5, 123.7,
CH2), 2.30–2.19 (m, 1 H, CH2), 2.15–1.72 (br., 6 H, NCH3), 1.65
(s, 6 H, NCH3), 1.22 (br., 4 H, THF), 0.41 (s, 9 H, Me3Si), 0.14 (s,
122.9, 119.0, 43.1 [(CCH3)2], 42.5 [(CCH3)2], 31.1 (CH3), 30.8 1 H, CH) ppm. 13C{1H} NMR (300 MHz, C6D6, 25 °C): δ = 147.8,
5658
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 5652–5661

Ca-Mediated Styrene Polymerization
3
3
135.6, 132.3, 131.4, 126.5 (benzyl), 124.5, 124.1, 123.5, 123.3, 122.1,
benzyl), 6.30 (d, JH,H = 7.1 Hz, 1 H, benzyl), 6.17 (t, JH,H =
122.0 (benzyl), 119.6, 118.6, 118.4, 116.9, 116.8, 114.7 (benzyl), 7.2 Hz, 1 H, benzyl), 2.61 (br., 4 H, THF), 2.26 (s, 3 H, CH3), 2.23
112.5 (benzyl), 91.7 (C9), 68.1 (THF), 63.2, 61.3, 44.8, 44.7, 25.3 (s, 3 H, CH3), 1.87 (s, 3 H, NCH3), 1.81 (s, 3 H, NCH3), 0.97 (br.,
(THF), 22.8, 22.3, 2.8 (Me3Si) ppm.
4 H, THF), 0.45 (s, 1 H, CH), 0.34 (s, 9 H, Me3Si-fluorenyl), 0.24
(s, 9 H, Me3Si-benzyl) ppm. 13C{1H} NMR (500 MHz, C6D6,
25 °C): δ = 150.6, 150.2, 150.2, 150.1, 147.3, 144.4, 143.7, 141.4,
141.2, 129.4, 129.4, 129.3, 129.2, 129.2, 128.3, 128.1, 128.1, 128.0,
126.5, 126.2, 126.2, 126.1, 126.0, 124.5, 121.7 (benzyl), 121.5, 121.5,
121.3, 121.2 (benzyl), 119.3, 118.7 (benzyl), 113.0 (benzyl), 88.9 (C-
9), 68.6 (THF), 53.4 (CCH3), 53.2 (CCH3), 44.8 (CH), 44.6
(NCH3), 41.5 (NCH3), 30.9 (CH3), 30.3 (CH3), 25.0 (THF), 2.56
(Me3Si-fluorenyl), 2.19 (Me3Si-benzyl) ppm.
Synthesis of 5: solution of 2,7-bis(tBu)-9-Me3Si-fluorene
A
(475 mg, 1.35 mmol) and [(2-Me2N-benzyl)]2Ca[15] (469 mg,
1.52 mmol) in benzene (10 mL) was heated to 60 °C over a period
of 2.5 h. Slow cooling of the reaction mixture to room temperature
gave large yellow crystals of 5 (537 mg, 76%), m.p. 151 °C (dec.).
1H NMR (500 MHz, C6D6/[D8]THF, 25 °C): δ = 7.85 (s, 2 H, fluor-
3
3
enyl), 7.74 (d, JH,H = 8.2 Hz, 2 H, fluorenyl), 6.85 (d, JH,H
=
8.2 Hz, 2 H, fluorenyl), 6.34 (t, 3JH,H = 7.8 Hz, 1 H, benzyl), 6.31–
3
6.21 (m, 2 H, benzyl), 5.80 (t, JH,H = 6.4 Hz, 1 H, benzyl), 2.04 Synthesis of 8: As complex 8 is highly soluble even in cold hexane,
(s, 6 H, NCH3), 1.52 (br s, 2 H, CH2), 1.33 (s, 18 H, CH3), 0.44 (s, it was synthesized in situ by reacting 2,7-bis(1,1-diphenylethyl)-
1
9 H, Me3Si) ppm. H NMR (500 MHz, C6D6, 25 °C): δ = 8.52 (d,
3JH,H = 8.6 Hz, 1 H, fluorenyl), 8.47 (d, 3JH,H = 8.6 Hz, 1 H, fluor-
enyl), 8.25 (s, 1 H, fluorenyl), 8.05 (s, 1 H, fluorenyl), 7.47 (d, 3JH,H
9-Et3Si-fluorene
with
[(2-Me2N-α-Me3Si-benzyl)]2Ca·(THF)2
(1.05 equiv.) in C6D6. Stirring the orange solution for 50 h at 70 °C
led to quantitative formation of the heteroleptic complex 8 which,
because of its extreme solubility, could not be freed from 2-Me2N-
α-Me3Si-toluene by washing procedures. It was therefore used in
solution for further polymerization experiments. The product has
3
= 8.5 Hz, 1 H, fluorenyl), 7.39 (d, JH,H = 8.2 Hz, 1 H, fluorenyl),
6.67 (t, JH,H = 7.3 Hz, 1 H, benzyl), 6.10 (t, JH,H = 6.9 Hz, 1 H,
benzyl), 5.85 (d, JH,H = 7.8 Hz, 1 H, benzyl), 5.81 (d, JH,H
7.2 Hz, 1 H, benzyl), 1.78 (s, 3 H, NCH3), 1.51 (s, 9 H, CH3), 1.42
3
3
3
3
=
been fully characterized by 2D NMR methods. 1H NMR
3
(s, 9 H, CH3), 1.24 (s, 3 H, NCH3), 0.73 (s, 9 H, Me3Si), –0.55 (d, (500 MHz, C6D6, 25 °C): δ = 7.93 (d, JH,H = 8.5 Hz, 1 H, fluor-
2JH,H = 11.8 Hz, 1 H, CH2), –0.96 (d, JH,H = 11.7 Hz, 1 H, CH2) enyl), 7.87 (d, JH,H = 8.5 Hz, 1 H, fluorenyl), 7.58 (s, 1 H, fluor-
2
3
ppm. 13C{1H} NMR (500 MHz, C6D6/[D8]THF, 25 °C): δ = 148.0, enyl), 7.43 (s, 1 H, fluorenyl), 7.22–6.83 (m, 23 H, Ar), 6.58 (t,
143.5, 143.3, 136.9, 125.4 (benzyl), 123.0, 122.3 (benzyl), 120.1, 3JH,H = 7.5 Hz, 1 H, benzyl), 6.23 (d, 3JH,H = 7.0 Hz, 1 H, benzyl),
3
117.9 (benzyl), 116.1, 112.8, 110.0 (benzyl), 85.2 (C-9), 43.3
6.10 (t, JH,H = 7.5 Hz, 1 H, benzyl), 2.52 (br., 4 H, THF), 2.21 (s,
[N(CH3)2], 40.9 (CH2), 34.8 [C(CH3)3], 32.1 [C(CH3)3], 2.39 3 H, CH3), 2.16 (s, 3 H, CH3), 1.80 [s, 6 H, N(CH3)2], 0.83–0.74
(Me3Si) ppm.
[m, 15 H, Si(CH2CH3)3], 0.29 (s, 1 H, CH), 0.27 (s, 9 H, Me3Si)
ppm. 13C{1H} NMR (500 MHz, C6D6, 25 °C): δ = 150.6, 150.3,
150.2, 150.1, 147.3, 144.5, 143.6, 142.0, 141.8, 130.1, 129.4, 129.4,
129.3, 129.0, 128.2, 128.1, 128.1, 128.0, 126.4 (benzyl), 126.2, 126.1,
125.4, 124.4 (benzyl), 123.9, 122.1, 121.6, 121.5, 121.4, 121.4, 121.2,
119.4, 119.1 (benzyl), 118.5, 113.0 (benzyl), 86.5 (C-9), 68.6 (THF),
53.4 (CCH3), 53.2 (CCH3), 45.0 (CH), 44.4 (NCH3), 41.7 (NCH3),
30.8 (CH3), 30.2 (CH3), 25.0 (THF), 8.50 [Si(CH2CH3)3], 6.40
[Si(CH2CH3)3], 2.58 (Me3Si-benzyl) ppm.
Synthesis of 6: A mixture of 2,7-bis(1-methyl-1-phenylethyl)-9-
Me3Si-fluorene (315 mg, 664 µmol) and [(2-Me2N-α-Me3Si-ben-
zyl)]2Ca·(THF)2
[13]
(417 mg, 698 µmol) in THF (4.0 mL) was
stirred for 2.5 h at 60 °C. The solvent was removed under vacuum,
and the remaining orange product was dissolved in warm hexane
(4 mL). Cooling the solution to room temperature yielded yellow
crystals (410 mg, 80%), m.p. 132 °C (dec.). 1H NMR (500 MHz,
C6D6, 25 °C): δ = 8.15 (s, 1 H, fluorenyl), 8.00 (s, 1 H, fluorenyl),
3
3
7.89 (d, JH,H = 8.5 Hz, 1 H, fluorenyl), 7.79 (d, JH,H = 8.5 Hz, 1
Synthesis of 9: As complex 9 is highly soluble even in cold hexane,
it was synthesized in situ with a procedure similar to that for 8.
The orange solution of 9 prepared in situ was used for further
3
3
H, fluorenyl), 7.41 (d, JH,H = 7.5 Hz, 2 H, Ar), 7.34 (d, JH,H
7.5 Hz, 2 H, Ar), 7.22–6.91 (m, 8 H, Ar), 6.86 (d, JH,H = 8.5 Hz,
1 H, benzyl), 6.67 (t, JH,H = 7.1 Hz, 1 H, benzyl), 6.32 (d, JH,H
= 7.6 Hz, 1 H, benzyl), 6.19 (t, JH,H = 7.1 Hz, 1 H, benzyl), 2.63
=
3
3
3
1
polymerization experiments. H NMR (500 MHz, C6D6, 25 °C): δ
3
3
= 7.97 (d, JH,H = 8.5 Hz, 1 H, fluorenyl), 7.91 (d, 3JH,H = 9.0 Hz,
(br., 4 H, THF), 1.96 (s, 3 H, CH3), 1.87 (s, 3 H, CH3), 1.85 (s, 3
H, NCH3), 1.84 (s, 3 H, NCH3), 1.79 (s, 6 H, CH3), 0.95 (br., 4 H, 7.22–6.82 (m, 19 H, Ar), 6.57 (t, JH,H = 7.0 Hz, 1 H, benzyl), 6.27
THF), 0.64 (s, 9 H, Me3Si-fluorenyl), 0.51 (s, 1 H, CH), 0.45 (s, 9
H, Me3Si-benzyl) ppm. 13C{1H} NMR (500 MHz, C6D6, 25 °C): δ
= 151.4, 151.2, 147.2, 146.1, 145.6, 141.6, 141.4, 136.1, 128.3, 127.3,
1 H, fluorenyl), 7.43 (s, 1 H, fluorenyl), 7.36 (s, 1 H, fluorenyl),
3
3
3
(d, JH,H = 6.5 Hz; 1 H, benzyl), 6.10 (t, JH,H = 6.5 Hz, 1 H,
benzyl), 2.70 (br., 4 H, THF), 2.26 (s, 3 H, CH3), 2.24 (s, 3 H,
CH3), 1.90 (s, 6 H, NCH3), 1.81 (s, 6 H, NCH3), 1.19 [s, 9 H,
127.2, 125.9, 125.9, 125.5 (benzyl), 123.9 (benzyl), 121.4, 121.3, C(CH3)3], 1.16 [s, 9 H, C(CH3)3], 1.15 [s, 9 H, C(CH3)3], 1.13 [s, 9
121.2, 121.0, 119.2 (benzyl), 118.2, 118.2, 118.0, 117.6, 113.1 (ben-
zyl), 87.9 (C-9), 68.5 (THF), 44.8 (NCH3), 44.4 (CH), 43.6 [C-
(CH3)2], 43.4 [C(CH3)2], 41.5 (NCH3), 31.4 (CH3), 31.3 (CH3), 30.9
H, C(CH3)3], 1.00 (br., 4 H, THF), 0.40 (s, 1 H, CH), 0.28 (s, 9 H,
Me3Si-benzyl), 0.12 (s, 9 H, Me3Si-fluorenyl) ppm. 13C{1H} NMR
(500 MHz, C6D6, 25 °C): δ = 148.6, 148.6, 147.7, 147.4, 147.3,
(CH3), 30.7 (CH3), 24.9 (THF), 2.71 (Me3Si-fluorenyl), 2.48 147.2, 147.0, 144.9, 144.6, 141.3, 141.1, 129.2, 129.1, 129.1, 129.0,
(Me3Si-benzyl) ppm.
126.5 (benzyl), 125.0, 125.0, 124.5 (benzyl), 121.7, 121.5, 121.2,
121.1, 119.3, 119.3 (benzyl), 118.8, 113.0 (benzyl), 88.7 (C-9), 68.6
(THF), 52.6 (CCH3), 52.5 (CCH3), 44.8 (CH), 44.4 (NCH3), 41.5
(NCH3), 34.5 [C(CH3)3], 34.4 [C(CH3)3], 31.5 [C(CH3)3], 30.5
(CCH3), 30.1 (CCH3), 25.1 (THF), 2.70 (Me3Si-fluorenyl), 2.22
(Me3Si-benzyl) ppm.
Synthesis of 7: A solution of 2,7-bis(1,1-diphenylethyl)-9-Me3Si-flu-
orene (242 mg, 404 µmol) and [(2-Me2N-α-Me3Si-benzyl)]2Ca·
[13]
(THF)2
(245 mg, 410 µmol) in THF (3.0 mL) was stirred for
2.5 h at 60 °C. After removing the solvent under vacuum, the re-
maining product was washed with hexane (2 mL) and dried in
vacuo. Repeating this procedure resulted in a yellow product
Styrene Polymerization: Polymerizations of styrene were performed
(307 mg, 83%), m.p. 202 °C (dec.). 1H NMR (500 MHz, C6D6, in a thermostatted 100 mL stainless steel Büchi reactor at normal
3
3
25 °C): δ = 8.02 (d, JH,H = 8.6 Hz, 1 H, fluorenyl), 7.95 (d, JH,H pressure either in cyclohexane (1  styrene solution, 1 m catalyst,
= 8.6 Hz, 1 H, fluorenyl), 7.61 (s, 1 H, fluorenyl), 7.52 (s, 1 H,
fluorenyl), 7.33–6.96 (m, 23 H, Ar), 6.69 (t, JH,H = 7.3 Hz, 1 H,
50 °C) or in pure styrene (1 m catalyst, 20 °C). In a typical poly-
merization experiment, the reactor was loaded with dry cyclohex-
3
Eur. J. Inorg. Chem. 2007, 5652–5661
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
5659

D. F. -J. Piesik, K. Häbe, S. Harder
FULL PAPER
Table 2. Crystal data for the compounds 4, 5 and 6.
Compound
4
5
6
Formula
MW
Size [mm]
Crystal system
C58H76Ca2N4Si2·C7H8
1057.70
0.7ϫ0.7ϫ0.2
monoclinic
Cc
C66H90Ca2N2Si2
1047.74
0.5ϫ0.4ϫ0.3
monoclinic
C2/c
C50H65CaNOSi2
792.29
0.5ϫ0.4ϫ0.3
triclinic
¯
P1
Space group
a [Å]
b [Å]
c [Å]
α
10.6545(8)
43.504(4)
12.9522(9)
90
96.288(6)
90
5967.4(8)
4
1.177
17.271(1)
15.677(1)
23.585(1)
90
110.099(1)
90
5996.7(7)
4
1.161
10.9274(4)
16.0377(5)
16.3604(6)
110.334(2)
94.677(2)
97.161(2)
2643.4(2)
4
β
γ
V [Å3]
Z
ρ [gcm–3]
1.103
0.195
µ(Mo-Kα) [mm–1]
0.273
0.270
T [°C]
θ (max.)
–120
25.1
–90
29.3
–90
27.5
Reflections total, unique
Rint
8146, 5571
0.017
127637, 8205
0.068
159543, 12019
0.077
Observed reflections [IϾ2σ(I)]
Parameter
5061
966
6476
331
9479
500
R1
wR2
0.028
0.072
0.051
0.139
0.065
0.155
GOF
1.04
–0.30/0.26
1.10
–0.60/0.74
1.08
–0.46/0.36
Max./min. residual electron density [eÅ–3]
ane (90 mL, dried with CaH2 and distilled from nBuLi) and dry
styrene (11.5 mL, ca. 100 mmol, freshly distilled from CaH2 and
stored over alox pearls). A solution of the initiator (0.1 mmol) in
benzene (1.0 mL) was added through a port. The usual appearance
of a red colour and a slight increase in temperature (1–3 °C) indi-
cated that the polymerization started immediately. After a polyme-
rization time of 45–60 min, the mixture was quenched with oxygen-
free methanol. Evaporation of all solvents yielded the polymer in
quantitative yields. Polymerization in neat styrene should only be
carried out at temperatures Յ20 °C, should be monitored continu-
ously and should be quenched when temperatures exceed 25 °C.
The polymers were analyzed by GPC and high temperature 1H and
13C NMR (solvent: [D2]tetrachloroethane). The tacticity of the
polymer was checked by analyzing the 13C NMR signal for the
Cipso in the phenyl ring.[22]
Acknowledgments
We thank Dieter Bläser and Prof. Dr. Roland Boese (University of
Duisburg-Essen) for measurements of the X-ray diffraction data
for complexes 5 and 6 and Dr. Konrad Knoll (BASF AG, Ludwig-
shafen) for GPC analyses.
[1] N. Ishihara, M. Kuramoto, Japan Patent 62 187 708, 1985.
[2] N. Ishihara, M. Kuramoto, M. Uoi, Macromolecules 1986, 21,
2464.
[3] N. Tomotsu, N. Ishihara, T. H. Newman, M. T. Malanga, J.
Mol. Catal. A 1998, 128, 167.
[4] C. Pellecchia, D. Pappalardo, L. Oliva, A. Zambelli, J. Am.
Chem. Soc. 1995, 117, 6593.
[5] R. Po, N. Cardi, Prog. Polym. Sci. 1996, 21, 47.
[6] M. K. Mahanthappa, R. M. Waymouth, J. Am. Chem. Soc.
2001, 123, 12093.
[7] J. Schellenberg, N. Tomotsu, Prog. Polym. Sci. 2002, 27, 1925.
[8] E. Kirillov, C. W. Lehmann, A. Razavi, J.-F. Carpentier, J. Am.
Chem. Soc. 2004, 126, 12240.
[9] J. Hitzbleck, K. Beckerle, J. Okuda, T. Halbach, R. Mühlhaupt,
Macromol. Symp. 2006, 236, 23.
[10] J. R. Wünsch in Kunststoff Handbuch 4: Polystyrol (Eds.: H.
Gausepohl, R. Gellert), Hanser, München, 1996, 82.
[11] M. Malanga, Adv. Mater. 2000, 12, 1869.
[12] J. Schellenberg, H.-J. Leder, Adv. Polym. Technol. 2006, 25, 141.
[13] S. Harder, F. Feil, A. Weeber, Organometallics 2001, 20, 1044.
[14] F. Feil, S. Harder, Angew. Chem. 2001, 113, 4391; Angew.
Chem. Int. Ed. Engl. 2001, 40, 4261.
Crystal Structures: Crystal diffraction data were measured on En-
raf–Nonius CAD4 and Siemens SMART CCD diffractometers. All
crystal structures were solved with SHELXS-97[26] and refined with
SHELXL-97.[27] PLATON[28] was used for geometry calculations
and graphics. Crystal data are given in Table 2.) The space group
for 4 was checked with the ADSYM procedure incorporated in
the program PLATON. No additional symmetry was found, and
therefore the true space group is Cc. The Flack parameter was re-
fined to 0.00(2). Crystals of 6 contain one equivalent of hexane in
the asymmetric unit. This molecule was extensively disordered and
treated with the SQUEEZE procedure[29] incorporated in PLA-
TON.
[15] S. Harder, F. Feil, Organometallics 2002, 21, 2268.
[16] F. Feil, S. Harder, Eur. J. Inorg. Chem. 2003, 3401.
[17] D. Hoffmann, W. Bauer, F. Hampel, N. J. R. van Ei-
kema Hommes, P. v. R. Schleyer, P. Otto, U. Pieper, D. Stalke,
D. S. Wright, R. Snaith, J. Am. Chem. Soc. 1994, 116, 528.
[18] F. Feil, S. Harder, Organometallics 2000, 19, 5010.
CCDC-646400 (compound 4), -646401 (compound 5), and
-646402 (compound 6) contain the supplementary crystallographic
data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
5660
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 5652–5661

Ca-Mediated Styrene Polymerization
[19] a) C. W. Haigh, R. B. Mallion, Progr. Nucl. Magn. Reson.
Spectrosc. 1980, 13, 303. Although this is the classical explana-
tion, it should be noted that recent calculations introduce a
different explanation: b) C. S. Wannere, P. v. R. Schleyer, Org.
Lett. 2003, 5, 605.
[25] H. G. Alt, R. Zenk, J. Organomet. Chem. 1996, 522, 39.
[26] G. M. Sheldrick, SHELXS-97, Program for Crystal Structure
Solution 1997, University of Göttingen, Germany.
[27] G. M. Sheldrick, SHELXL-97, Programs for the Determination
of Crystal Structures, University of Göttingen, Germany, 1997.
[28] A. L. Spek, PLATON, A Multipurpose Crystallographic Tool
2000, Utrecht University, Utrecht, The Netherlands.
[29] P. A. van der Sluis, A. L. Spek, Acta Crystallogr. Sect. A 1990,
46, 194.
[20] L. Gold, J. Chem. Phys. 1958, 28, 91.
[21] K. Knoll in Kunststoff Handbuch 4: Polystyrol (Eds.: H. Gause-
pohl, R. Gellert), Hanser, München, 1996, 67.
[22] F. Feil, S. Harder, Macromolecules 2003, 36, 3446.
[23] R. D. Culp, A. H. Cowley, Organometallics 1996, 15, 5380.
[24] S. Kajigaeshi, T. Kadowaki, A. Nishida, S. Fujisaki, M. Nogu-
chi, Synthesis 1984, 4, 335.
Received: July 27, 2007
Published Online: November 2, 2007
Eur. J. Inorg. Chem. 2007, 5652–5661
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
5661