Dehydrocoupling polymerization of arylsilanes with chloro(aryloxy)bis(cyclopentadienyl)zirconium complex catalysts

Article abstract of DOI:10.1016/S0022-328X(99)00495-7

Complexes generated by treating chloro(aryloxy)bis(cyclopentadienyl)zirconium [3; aryloxy=2,6-di-tert-butyl-4-methylphenoxy (3a), 2,6-diisopropylphenoxy (3b), phenoxy (3c), 2,3,4,5,6-pentafluorophenoxy (3d), and 2,6-dimethoxyphenoxy (3e)] with one equivalent of n-BuLi catalyze dehydrocoupling polymerization of arylsilanes 1 [aryl=phenyl (1a), p-methoxyphenyl (1b), p-dimethylaminophenyl (1c), p-methylthiophenyl (1d)] at room temperature to afford corresponding polysilanes. Their catalytic performance was compared with the conventional system generated from dichlorobis(cyclopentadienyl)zirconium (4) and two equivalents of n-BuLi. The reaction of 1a using 3a/n-BuLi, 3b/n-BuLi or 3c/n-BuLi gave high-molecular-weight polysilanes (M_w=13.3×10³ and M_n=5.8×10³ with 3a; M_w=10.4×10³ and M_n=4.2×10³ with 3b; M_w=8.8×10³ and M_n=5.3×10³ with 3c). Formation of the polysilane was fast in comparison with 4, in particular when the catalyst was ligated by a sterically demanding (3a) or electron-withdrawing (3d) aryloxy group. The time course of the gel permeation chromatography profiles suggested that complex 3a showed faster chain growth than 3d, 3e and 4. The extent of the formation of undesired cyclics was relatively small (<10%) in the sterically demanding 3a/n-BuLi- or 3b/n-BuLi-catalyzed reaction.

Full text of DOI:10.1016/S0022-328X(99)00495-7

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 595 (2000) 1–11
Dehydrocoupling polymerization of arylsilanes with
chloro(aryloxy)bis(cyclopentadienyl)zirconium complex catalysts
Yasushi Obora, Masato Tanaka *
National Institute of Materials and Chemical Research, Tsukuba, Ibaraki 305-8565, Japan
Received 28 June 1999; accepted 13 August 1999
Abstract
Complexes generated by treating chloro(aryloxy)bis(cyclopentadienyl)zirconium [3; aryloxy=2,6-di-tert-butyl-4-methylphenoxy
(3a), 2,6-diisopropylphenoxy (3b), phenoxy (3c), 2,3,4,5,6-pentafluorophenoxy (3d), and 2,6-dimethoxyphenoxy (3e)] with one
equivalent of n-BuLi catalyze dehydrocoupling polymerization of arylsilanes 1 [aryl=phenyl (1a), p-methoxyphenyl (1b),
p-dimethylaminophenyl (1c), p-methylthiophenyl (1d)] at room temperature to afford corresponding polysilanes. Their catalytic
performance was compared with the conventional system generated from dichlorobis(cyclopentadienyl)zirconium (4) and two
equivalents of n-BuLi. The reaction of 1a using 3a/n-BuLi, 3b/n-BuLi or 3c/n-BuLi gave high-molecular-weight polysilanes
(Mw=13.3×103 and Mn=5.8×103 with 3a; Mw=10.4×103 and Mn=4.2×103 with 3b; Mw=8.8×103 and Mn=5.3×103
with 3c). Formation of the polysilane was fast in comparison with 4, in particular when the catalyst was ligated by a sterically
demanding (3a) or electron-withdrawing (3d) aryloxy group. The time course of the gel permeation chromatography profiles
suggested that complex 3a showed faster chain growth than 3d, 3e and 4. The extent of the formation of undesired cyclics was
relatively small (B10%) in the sterically demanding 3a/n-BuLi- or 3b/n-BuLi-catalyzed reaction. © 2000 Elsevier Science S.A. All
rights reserved.
Keywords: Dehydrocoupling polymerization; Arylsilanes; Aryloxy–zirconocene complex
formed in large quantities. Exceptional cases that af-
ford respectable molecular weights and suppress the
formation of cyclics include the use of ‘cation-like’ early
transition metal complexes reported by Harrod’s group
[9e,i], ethylenebis(indenyl)zirconium complex (5) [10]
reported by Waymouth and co-workers [9z], and
mixed-ring CpCp*Zr-based catalysts reported by
Tilley’s group [9j]. In this paper, we wish to report the
catalytic performance of a family of chloro(aryl-
oxy)bis(cyclopentadienyl)zirconium complexes (3) used
in conjunction with one equivalent of n-BuLi (Eq. (1)).
1. Introduction
Synthesis of polysilanes is experiencing vigorous
growth because these polymers have significant poten-
tial for electronic, optoelectronic, and ceramic applica-
tions [1]. The intriguing electronic properties of
polysilanes arise from the delocalized s-electrons along
the backbone [2], and are influenced by the molecular
weight [3], polymer conformation [4], and substituents
attached to the backbone [5]. The Wurtz-type coupling
reaction of halosilanes with alkali metals is the only
commercial process and most widely used in the labora-
tory to synthesize high molecular weight polysilanes [6].
On the other hand, since the discovery of Harrod and
co-workers [7], the Group 4 metallocene-catalyzed de-
hydrocoupling of primary silanes has been intensively
studied as an alternate synthetic method [8,9]. How-
ever, the molecular weights of the resulting polymers
are usually low and undesired cyclic products are also
* Corresponding author. Tel.: +81-298-54-4430; fax: +81-298-54-
4474.
(1)
0022-328X/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(99)00495-7

2
Y. Obora, M. Tanaka / Journal of Organometallic Chemistry 595 (2000) 1–11
Table 1
Performance of zirconocene complexes in dehydrocoupling of 1a a
Entry
Zr
Time (h)
Raw polymer
Fractionation polymer Mw/103 b (Mw/Mn) b
Mw/103 b (Mw/Mn) b
Cyclics c (%)
1
2
3
4
5
6
3a
3a
3b
3c
3d
3e
4
4
4
4
120
168
120
120
96
120
120
2
9.2 (5.0)
10.1 (5.5)
8.4 (4.9)
6.9 (4.8)
6.6 (4.9)
2.5 (4.2)
2.3 (3.8)
1.6 (2.8)
2.4 (3.8)
4.3 (4.0)
8 (6)
7 (10)
8 (9)
11.9 (2.1)
13.3 (2.3)
10.4 (2.5)
8.8 (1.7)
9.7 (1.7)
3.4 (2.0)
3.2 (2.0)
nd d
17 (21)
19 (23)
nd d (13)
36 (40)
nd d
7 e
8 e,f
9 e,f
10 e,f
24
120
nd d (13)
12 (14)
3.7 (2.5)
5.9 (2.2)
a Reactions were run under conditions I; zirconocene (0.04 mmol), n-BuLi (0.04 mmol), 1a (0.8 ml, 6.4 mmol), and toluene (0.2 ml), room
temperature (23°C). Conversion of 1a was\99%.
b Determined by GPC using polystyrene standards.
c Determined by GPC. The numbers in parentheses are determined by 1H-NMR.
d Not determined.
e The quantity of n-BuLi was two equivalents relative to the zirconocene complex.
f B(C6F5)3 (0.04 mmol) was used as cocatalyst.
Some of these zirconium complexes have been tested
as catalysts (precursors) for polyolefin synthesis [11].
If the aryloxy ligand is retained throughout the catal-
ysis, the catalyst is envisioned to have only one coor-
dination site for polymerization, which allows better
control of dehydrocoupling polymerization of arylsi-
lanes by tuning the steric and electronic nature of the
aryloxy ligand. Indeed, some of the family, although
very much dependent of the structure, could show
respectable performance, e.g. enhanced molecular
weight of the resulting polysilanes, suppressed forma-
tion of cyclics, higher reaction rate, and applicability
to functionalized arylsilane monomers. Corriu and co-
workers [9h] reported the performance of diaryloxyti-
tanocene and zirconocene complexes without the use
of any external reductants such as n-BuLi. We believe
the active species generated in their case is different
from ours (Chart 1).
2. Results and discussion
The reactions catalyzed by 3 were carried out un-
der conditions I in most cases, i.e. at room tempera-
ture (r.t.) (23°C) in toluene (0.2 ml) using a 160:1:1
molar ratio of a silane monomer (6.4 mmol), 3 (0.04
mmol), and n-BuLi (0.04 mmol) in a closed Schlenk
flask. However, in control experiments in which the
conventional
dichlorobis(cyclopentadienyl)zirconium
(4) complex was used, we followed Corey’s standard
procedure [9x,e%,h%] in which the 4/n-BuLi molar ratio
was set at 1:2 by doubling the quantity of n-BuLi.
Although the monomer was consumed in the first few
hours, the reaction was continued for several days to
enhance the molecular weight. The mixtures resulting
from the polymerization were analyzed by gel perme-
ation chromatography (GPC) with calibration using
polystyrene standards. As has been usually encoun-
tered in dehydrocoupling polymerization of primary
silanes with conventional catalyst systems, the molec-
ular-weight distributions obtained with the new cata-
lyst systems were also bimodal in most cases,
comprising linear polysilane of a higher molecular
weight and lower molecular weight (Mw= ꢀ500)
cyclics1. Fractionation by preparative GPC with oxy-
gen-free toluene eluent led to isolation of the linear
polysilane as an off-white or colorless sticky liquid,
1 Regardless of the resolution of linear and cyclic polysilanes in
GPC, the Mw and Mn for the raw polymer in Table 1 are given for
the entire polymer covering the whole bimodal distribution.
(Chart 1)

Y. Obora, M. Tanaka / Journal of Organometallic Chemistry 595 (2000) 1–11
3
which was soluble in most organic solvents such as
benzene and THF and was further analyzed by GPC and
by 29Si-NMR spectroscopy (vide infra). Reprecipitation
from toluene–petroleum ether according to Harrod’s
procedure, after filtration of the reaction mixture (entry
1, Table 1) through a Florisil column to remove the
catalyst, also led to isolation of the higher-molecular-
weight polysilane (in 56% yield), which showed nearly the
same spectroscopic data as the higher-molecular-weight
fraction obtained by preparative GPC. As the results
summarized in Table 1 show, some of the aryloxy–zir-
conocene complexes 3 display superiority to the conven-
tional complex 4 in three aspects in dehydrocoupling of
phenylsilane 1a and prove to be listed among the best
systems known for the dehydrocoupling.
polymer. The data are also listed in Table 1. In most cases
where the degree of polymerization is relatively high (Mw
of the raw polymer ]ca. 3000), the GPC profile shows
distinct separation between the linear polysilanes and the
cyclics, so that the extent of the formation of the cyclics
in the raw polymer can be readily determined by GPC.
When the profile is not sufficiently resolved because of
a low degree of polymerization or because of the cyclics
being formed only in a small quantity, the extent of the
cyclics formation can be determined by integration of the
signals arising from SiꢀH units in linear (4.3–4.9 ppm)
1
and cyclic (4.9–5.3 ppm) structures in H-NMR spec-
troscopy. Although the values obtained by GPC and
1H-NMR spectroscopy appear slightly different from
each other, the difference is not serious.
A second feature in the aryloxy–zirconocene complex-
catalyzed reactions, as entries 1–3 in Table 1 indicate, lies
in the tunability of the aryloxy ligand structure for
minimized formation of the cyclics. Thus, the formation
of the cyclics in the reactions using 3a or 3b was 510%
of the entire raw product, which was smaller than the
36% observed with 4, and was in the same range as or
slightly better than the value obtained using Harrod’s
‘cation-like’ system used under similar conditions (entry
10). The efficacy in this respect decreases in the order
3a]3b\3e\3c]3d. The trends appear to primarily
arise from the steric demand of the ortho substituent, and
agree with Tilley’s argument on the cyclics formation
mechanism, which involves highly congested four-center
transition state [8b].
2.1. Molecular weight
First, some of the new zirconocene systems can en-
hance the ultimate molecular weight of the polysilane as
compared with 4. For example, the Mw value of the
higher-molecular-weight fraction obtained with the
chloro(2,6 - di - tert - butyl - 4 - methylphenoxy)bis(cyclo -
pentadienyl)zirconium (3a)/n-BuLi system was 11.9×
103 (entry 1), while the value in the control experiment
using the conventional 4/2n-BuLi system was only 3.2×
103 (entry 7). Other new systems substituted by 2,6-diiso-
propylphenoxy (3b, entry 3), phenoxy (3c, entry 4), and
2,3,4,5,6-pentafluorophenoxy (3d, entry 5) groups per-
formed similarly, although 2,6-dimethoxyphenoxy com-
plex (3e, entry 6) did not show the efficacy, presumably
due to the electronic nature of the methoxy group.
Fine tuning of the steric and electronic effects of the
ligand is crucial for higher molecular weights. Tilley
could achieve high molecular weights, e.g. Mw=13.9×
103 and Mn=5.7×103 (after fractionation), by using
CpCp*Zr complexes [9j]. Harrod reported that the
molecular weight could be even higher, e.g. Mw=13.8×
103, Mn=7.3×103 (after fractionation), when the poly-
merization was catalyzed by ‘cation-like’ catalysts such
as the one generated from CpCp*ZrCl2, 2n-BuLi, and
B(C6F5)3 [9i]. The highest molecular weights that the new
aryloxy complexes could achieve were found with 3a;
Mw=13.3×103, Mn=5.8×103 (entry 2). These values
are not only higher than those for dichloro complex 4,
but are similar or close to the highest values reported by
Tilley and Harrod. Thus, our strategy to introduce an
aryloxy ligand into the zirconium center proves to be a
useful tool to enhance the ultimate molecular weight,
although the performance depends on the structure of the
aryloxy ligand.
2.3. Time course of the dehydrocoupling
Thirdly, monitoring the time course of the reaction by
1H-NMR spectroscopy revealed the superiority of some
of the new catalyst systems in the reaction rate. The
reaction to examine this aspect was run under conditions
2.2. Formation of cyclics
In experiments to compare the catalyst performance,
we determined the percentage of cyclics in the raw
Fig. 1. Progress of the dehydrocoupling of 1a using zirconocene
complex catalysts.

4
Y. Obora, M. Tanaka / Journal of Organometallic Chemistry 595 (2000) 1–11
II, i.e. at r.t. in toluene (0.5 ml) and mesitylene (50 ml,
as internal standard for 1H-NMR spectroscopy) using a
320:1:1 molar ratio of 1a (6.4 mmol), 3 (0.02 mmol),
and n-BuLi (0.02 mmol) under constant flow of nitro-
gen gas. Since the signals assignable to the ꢀ(SiHPh)nꢀ
structures (4.3–5.3 ppm) are distinctly separated from
the signals due to PhSiH3 (4.26 ppm) and mesitylene
(2.15 ppm), integration readily affords the quantities of
the monomer and the product. Fig. 1 shows, for in-
stance, that the use of perfluorophenoxy complex 3d
rapidly promoted the dehydrocoupling as compared
with 4. Complexes 3a and 3c behaved similarly but to a
modest extent. However, the efficacy of dimethoxyphe-
noxy complex 3e was only marginal. The comparison of
the performance among complexes 3 concludes that the
dehydrocoupling rate is enhanced by the introduction
of an electron-withdrawing (such as F) or sterically
demanding (such as tert-butyl and isopropyl at the
ortho position) group into the aryloxy ligand.
The time course of the GPC profiles (Fig. 2) indicates
that the chain growth was efficiently promoted by the
use of 3a ligated by the sterically demanding aryloxy
ligand, while it was slow in the 3e- or 4-catalyzed
reaction. Complex 3d displayed intermediate perfor-
mance in this respect.
2.4. Microstructure of the resulting polysilane
Waymouth and co-workers reservedly proposed a
tentative assignment of the microstructure of the
Fig. 3. 29Si-NMR (DEPT) spectra of poly(phenylsilane)s (higher
molecular weight fractions) prepared by the use of zirconocene com-
plex catalysts 3a (A), 3b (B), 3c (C), 3d (D), 3e (E), and 4 (F). The
asterisk denotes SiH2 signals.
poly(phenylsilane) [9z], which was solely based on the
similarity in 29Si-NMR spectroscopy to trans-
hexaphenylcyclohexasilane [12]. However, the chemical
shifts of cyclic silicon compounds can differ from those
of linear analogs [13]. Because stereochemistry-defined
linear oligosilanes are not available, we have no war-
rant for any assignment of the microstructure based on
the spectroscopic comparison. Indeed, recent publica-
tions show the complexity of the microstructure analy-
sis based on the 29Si-NMR spectroscopy [6c,9a]. In this
context we simply illustrate in Fig. 3 the spectra of the
polymers formed by using the new and conventional
systems (for comparison) at r.t. and do not discuss the
stereoregularity in detail.
Fig. 2. Time course of GPC profiles in the zirconocene complex-cata-
lyzed dehydrocoupling of phenylsilane: 3a (a–d), 3d (e–h), 3e (i–l), 4
(m–p). Reaction time: (a, e, i, m) 5 min; (b, f, j, n) 20 min; (c, g, k,
o) 2 h (d, h, l, p) 5 h.

Y. Obora, M. Tanaka / Journal of Organometallic Chemistry 595 (2000) 1–11
5
Table 2
Effects of reaction temperature and concentrations of 1a and 3a in 3a/n-BuLi-catalyzed dehydrocoupling of 1a a
Entry
Temperature (°C)
Solvent
Time (h)
Raw polymer
Mw/103 b (Mw/Mn) b
Cyclics c (%)
11
50
50
50
23
23
23
0
0
0
0
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
None
2
5
24
2
5
24
5
24
72
120
2
5
24
24
4.5 (4.0)
4.9 (4.9)
5.2 (4.5)
2.8 (2.8)
5.1 (3.9)
8.6 (4.8)
0.7 (3.1)
0.9 (3.0)
1.3 (4.2)
1.5 (4.8)
3.4 (3.6)
5.3 (4.1)
8.1 (4.5)
1.8 (3.9)
17
16
12
13
14
15
16
17
18
19
20
21
22
23
24 e
19
nd d
9
9
nd d
nd d
nd d
nd d
nd d
10
23
23
23
23
None
None
Toluene
7
51
a Reactions were run under conditions I [3a (0.04 mmol), n-BuLi (0.04 mmol), 1a (0.8 ml, 6.4 mmol), and toluene (0.2 ml for entries 11–20, 3.0
ml for entry 24), room temperature (23°C)] except entries 21–23, where the reaction was run neat without toluene solvent. Conversion of 1a was
\99% unless otherwise noted.
b Determined by GPC using polystyrene standards.
c Determined by GPC.
d Not determined.
e Conversion of 1a was 85%.
aryloxy ligand-free catalyst carries the catalysis (vide
infra). On the other hand, the reaction run at 0°C gave
only low molecular weight (B2.0×103) oligosilanes
(entries 17–20). Since the conversion of the monomer
was 100% in each entry, the low molecular weight is
presumably because the monomer–chain and chain–
chain couplings required for the chain growth are very
slow at this temperature.
The chain length of the linear polysilane and the
extent of cyclics formation in the reactions using neat
phenylsilane (entries 21–23) were very similar to those
in corresponding reactions run in the presence of a
small quantity of toluene (entries 14–16), which does
not agree with Harrod’s experience [9i]. When a larger
quantity of toluene was used as solvent (entry 24) the
cyclics were extensively formed and the molecular
weight was much lower than in entry 16, as anticipated
from the Tilley’s observation and argument based on
the unimolecular cyclization mechanism [8b].
2.5. Effect of reaction conditions
In addition to the structure of the catalyst, various
other factors such as the presence or absence of the
solvent, temperature, concentration of the monomer,
etc. are reported to influence the dehydrocoupling pro-
cess [8b,9e]. For instance, reactions in neat phenylsilane
at lower temperatures are envisioned, based on the
consideration by Tilley, to be favorable for suppression
of the cyclics formation [8b]. Waymouth found that
polymerization at higher concentrations of Cp2ZrMe2
or Cp2ZrHCl caused a change in the profiles of 29Si-
NMR spectra of the resulting polymers [9z]. Harrod
reported that the presence of even a small quantity of
an aromatic solvent significantly decrease the activity
because of competition between the SiꢀH and aromatic
ring for the coordination site [9e]. In this context, we
briefly examined the effects of these factors under con-
ditions I using 3a as catalyst precursor to further clarify
the nature of the aryloxy systems (Table 2).
The reaction run at 50°C decreased the ultimate
molecular weight and increased the formation of the
cyclics as compared with the reaction at r.t. (entry 13
versus entry 16). The 29Si-NMR spectrum of the poly-
mer from entry 13 is illustrated in Fig. 4. The origin of
the increased formation of the cyclics at 50°C is am-
biguous; it can be due either to more extensive involve-
ment of the higher energy cyclization pathway at the
higher temperature as argued by Tilley [8b] and/or to a
possible change of the catalyst structure, such that an
Fig. 4. 29Si-NMR (DEPT) spectrum of poly(phenylsilane) obtained
by the use of 3a/BuLi catalyst at 50°C.

6
Y. Obora, M. Tanaka / Journal of Organometallic Chemistry 595 (2000) 1–11
Table 3
Dehydrocoupling of 1a with preheated 3a/n-BuLi catalyst a
Entry
Temperature
(°C)
Solvent
Time (h)
Raw polymer
Fractionation polymer Mw/103 b (Mw/Mn) b
Mw/103 b (Mw/Mn) b
Cyclics c (%)
25
26
27
23
23
23
Toluene
Toluene
Toluene
2
5
24
2.5 (3.3)
4.2 (3.9)
4.8 (4.1)
nd d
nd d
nd d
nd d
9 (11)
6.0 (2.1)
a Reactions were run under conditions I; 3a (0.04 mmol), n-BuLi (0.04 mmol), 1a (0.8 ml, 6.4 mmol), and toluene (0.2 ml), room temperature
(23°C). Conversion of 1a was \99%. The catalyst was preheated with 0.05 ml of phenylsilane at 50°C for 30 min before injection of further 1a
(6.4 mmol); see the text.
b Determined by GPC using polystyrene standards.
c Determined by GPC. The numbers in parentheses are determined by 1H-NMR.
d Not determined.
2.6. Attempted study on the structure of the acti6e
species: comparison with Corriu’s system
mixture and the reaction was continued at the same
temperature over 24 h. The molecular weight of the
polymer after 24 h (entry 27) was lower than that from
entry 16, indicative of the active species having at least
partly experienced a structural change upon the pre-
heating. Neither (2,6-diisopropylphenoxy)phenylsilane
nor diisopropylphenol was found by GC–MS to be
1H-NMR and GC–MS studies on the reaction of an
equimolar mixture of 3b, n-BuLi and 1a revealed that
neither (2,6-diisopropylphenoxy)phenylsilane nor 2,6-
diisopropylphenol was formed at r.t. (23°C), suggesting
that the ZrꢀOAr bond remained intact. When the NMR
tube was heated at 50°C for 30 min, several very weak
signals emerged in the SiꢀH region (4.3–5.3 ppm).
However, none of these agreed in chemical shift with
(2,6-diisopropylphenoxy)phenylsilane. Besides these sig-
nals the NMR spectrum displayed very weak new sig-
nals also in the Cp region (5.8–6.1 ppm), suggesting a
structural change having taken place to some extent
upon heating, although it was unable to identify what it
was. GC–MS analysis did not detect the formation of
(2,6-diisopropylphenoxy)phenylsilane nor diisopropy-
lphenol in the reaction mixture, but diphenyldisilane
was found to be formed. Accordingly, the weak signals
that appeared at 4.3–5.3 ppm are presumably mainly
due to diphenyldisilane and other oligosilanes. Corriu
and co-workers [9h] examined the catalytic performance
of diaryloxytitanium and zirconium complexes without
addition of any external reductants like n-BuLi. In
order to trigger the catalysis at r.t., they normally
preheated the reaction mixture for a while to 50°C to
activate the catalyst species, such that the starting
aryloxy–metal precursors were converted to HꢀM spe-
cies with incidental formation of aryloxysilanes by the
reaction with the hydrosilane monomer. Accordingly,
their resulting polymers showed similar characteristics,
independent on the structure of the aryloxy ligand in
the precursor complex. We also examined the catalytic
performance of 3a in a procedure similar to Corriu’s
(Table 3). Thus, a mixture prepared as in entries 14–16
(Table 2), i.e. 3a (0.04 mmol), n-BuLi (0.04 mmol), and
toluene (0.2 ml), was preheated for 30 min at 50°C with
a small amount (0.4 mmol) of 1a. After cooling the
mixture to r.t., 1a (6.4 mmol) was further added to the
1
formed in this mixture. H-NMR spectroscopy of the
mixture at 50°C was not informative of the structure of
active catalyst species (vide supra) either. However,
based on the polymerization characteristics depending
on the structure of the catalyst, we can safely presume
that, at least at r.t., the aryloxy ligand is bound to
zirconium throughout the catalysis.
2.7. Applicability to functionalized monomers
Another feature of the new catalyst systems is their
applicability to arylsilanes having ‘coordinative’ func-
tional groups (Table 4). There have been only a limited
number of attempts to polymerize substituted phenylsi-
lanes reported. Waymouth and co-workers reported
reactivities of arylsilanes having non-coordinative sub-
stituents, such as meta and para substituted tolylsilanes
and (trifluoromethyl)phenylsilanes, relative to that of
unsubstituted phenylsilane, using a conventional zir-
conocene-based catalyst [9m]. We have learned that the
conventional zirconocene catalysts do not polymerize
arylsilanes such as p-methoxyphenylsilane (1b) and p-
dimethylaminophenylsilane (1c) that possess coordina-
tive functional groups, presumably due to the catalyst
deactivation by coordination of methoxy and dimethyl-
amino groups to the metal center. For example, under
conditions I, the reaction of 1b using the 5/2n-BuLi
catalyst system gave only the corresponding disilane
and trisilane [14]. An attempted polymerization of 1c
did not proceed at all in the presence of the same
catalyst [14]. In contrast, the 3a/n-BuLi catalyst system
did polymerize 1b, 1c, and p-methylthiophenylsilane

Y. Obora, M. Tanaka / Journal of Organometallic Chemistry 595 (2000) 1–11
7
Table 4
Dehydrocoupling of functionalized silanes with 3a/n-BuLi catalyst a
c
c
Entry Silane
Time (h)
Conv. (%) b Raw polymer Mw (103) (Mw/Mn) c
Fractionation polymer Mw (103) (Mw/Mn) c
2
28
29
30
1a
1b
1c
1d
168
168
72
\99
68
91
10.1 (5.5) d
2.1 (3.9) d
1.0 (2.0)
13.3 (2.5)
2.8 (3.5)
1.0 (1.7)
9.8 (2.2)
96
\99
8.6 (3.7) d
a Reactions were run under conditions I; 3a (0.04 mmol), n-BuLi (0.04 mmol), silane (6.4 mmol), and toluene (0.2 ml), room temperature (23°C).
b Determined by 1H-NMR.
c Determined by GPC using polystyrene standards.
d The raw polymers contain 7–10% of low molecular weight fraction (Mwꢀ500).
(1d) to give poly(arylsilane)s (2b–d), although the
molecular weight was not high in all cases. 29Si-NMR
profiles of the resulting polymers are illustrated in Fig.
5 including the one for polyphenylsilane obtained using
the same catalyst. The present catalyst system is useful
for the synthesis of functionalized oligo- and/or
poly(arylsilane)s.
Unlike the substituted phenylsilanes, 2-thienylsilane,
a heteroaromatic silane having a coordinative sulfur
functionality, did not polymerize; the reaction of 2-
thienylsilane using the 3a/n-BuLi system under the
conditions I was slow and formed only the correspond-
ing disilane and trisilane (144 h, conversion 59%: disi-
lane 52%, trisilane 7%). As experienced with other
catalyst systems, polymerization of alkylsilanes was less
efficient than that of phenylsilane; for example, the
reaction of n-HexSiH3 by the same procedure as entry
2, Table 1, gave only oligomers (Mw=1.0×103).
2.8. Conclusions
The ligand structure is a crucial factor that influences
the catalysis. The results herein described indicate that
the performance (molecular weight, extent of the for-
mation of the undesired cyclics, reaction rate) of aryl-
oxy–zirconocene-based catalysts can be substantially
modified by introducing sterically demanding and elec-
tron-withdrawing aryloxy ligands. Furthermore, the
present catalyst system is useful for preparation of
various functionalized poly(arylsilane)s.
3. Experimental
3.1. General procedures and materials
All manipulations of air-sensitive materials were per-
formed with rigorous exclusion of oxygen and moisture
in Schlenk-type glassware on a dual-manifold Schlenk
line interfaced to a high-vacuum system, or in an
argon-filled glove box with a recirculator (B1 ppm
O2). NMR spectra were measured in C6D6 on a Bruker
Fig. 5. 29Si-NMR (DEPT) spectra of poly(arylsilane)s 2a (A), 2b (B),
2d (C), and 2c (D) (higher-molecular-weight fractions) prepared by
the use of 3a/BuLi catalyst. The asterisk denotes SiH2 signals.
1
ARX-300 instrument (300 MHz for H; 75.5 MHz for

8
Y. Obora, M. Tanaka / Journal of Organometallic Chemistry 595 (2000) 1–11
13C; 59.6 MHz for 29Si). An INEPT technique was used
for 29Si-NMR measurements (~=27 ms, Z=12 ms).
The chemical shifts were referenced to Me4Si (0 ppm)
solution at r.t. over a period of 2 h. After addition of
the Grignard reagent, the solution was refluxed for 24
h. The reaction mixture was cooled and filtered to
remove the salt to give a yellow solution.
1
for H-, 13C- and 29Si-NMR. IR spectra were obtained
using a Jasco FT/IR-5000 spectrometer. GC–MS
analyses were performed with Shimadzu QP-1000 and/
or QP-5000 spectrometers (EI, 70 eV). GC–(HR)MS
analyses (EI, 70 eV) were performed on a Jeol DX-303
spectrometer. Molecular weights of polymers were mea-
sured with a GPC system equipped with a Shimadzu
LC-6A high-pressure pump, Shodex KF-80M (pore
size: 6 mm), KF-802 (pore size: 6 mm), and KF-801
(pore size: 6 mm) columns, and Shimadzu RID-2A,
Shodex SE-61 (RI) and/or Shimadzu SPD-M1A (UV)
detectors using THF eluent and polystyrene standards.
Preparative-scale GPC was carried out with a Japan
Analytical Industry Co. Ltd. LC-908-G30 system
equipped with Jaigel-1H (i.d. 20×600 mm, pore size:
10 mm) and Jaigel-2H (i.d. 20×600 mm, pore size: 10
mm) columns. Elemental analyses were performed by
the Analysis Center, National Institute of Materials
and Chemical Research. C6H5SiH3 and n-C6H13SiH3
were obtained from the Chisso Corporation and dis-
tilled from LiAlH4 before use. Other liquid materials
3.2.3. Reduction with LiAlH4
To a mixture of 3.50 g (92.3 mmol) of LiAlH4 and
250 ml of Et2O placed in a 500 ml round-bottom flask
equipped with a reflux condenser and pressure-equaliz-
ing dropping funnel was slowly added the foregoing
yellow solution containing p-MeSC6H4SiCl3 over a pe-
riod of 3 h. Gentle reflux was maintained during the
addition. After completion of the addition, the mixture
was further refluxed for 48 h, cooled and was quenched
by a slow addition of water. The ether layer was
separated, washed with water, dried over magnesium
sulfate, and was filtered. The filtrate was concentrated
and the residue was fractionally distilled to give 8.77 g
(46% overall yield based on the starting bromide) of 1d.
1
B.p. 45°C/0.4 mmHg. H-NMR: l 1.93 (s, 3H), 4.22 (s,
3H), 6.90 (d, 2H), 7.23 (d, 2H). 13C-NMR: l 14.57,
125.87, 126.79, 129.02, 136.41. 29Si-NMR (DEPT): l
−59.92. IR (neat): 3066, 2988, 2922, 2158, 1582, 1541,
1485, 1437, 1383, 1321, 1255, 1131, 1083, 917, 870, 806,
741, 727, 690, 642, 518, 474 cm−1. MS (relative inten-
sity): m/z 155 [M++1, 12], 154 [M+, 74], 152 (23), 139
(38), 138 (88), 137 (28), 124 (12), 109 (11), 107 (24), 106
(14), 105 (29), 103 (14), 91 (31), 80 (7), 79 (87), 78 (47),
77 (100), 76 (72), 69 (8), 66 (10), 65 (16), 63 (8), 61 (28),
60 (13), 55 (13), 54 (10), 53 (59), 52 (10), 51 (31), 50
(18), 45 (19). HRMS: Calc. for C7H10SSi [M+]
154.0272. Found 154.0256. Anal. Calc. for C7H10SSi: C,
54.49; H, 6.53. Found: C, 54.62; H, 6.03%.
,
and solvents were dried over appropriate agents (4 A
molecular sieves, sodium, CaH2 and LiAlH4) and dis-
tilled in vacuo or under nitrogen. ZrCl4 (Aldrich), 5
(Aldrich), and Cp2ZrCl2 (Strem) were commercial prod-
ucts and used as received. Compounds 3a [15], 3b [11],
Cp2Zr(Me)Cl [16], B(C6F5)3 [17] and 2-thienylsilane [18]
were synthesized according to the literature methods.
p-CH3C6H4SiH3 was obtained by reduction of p-
CH3C6H4SiCl3 (Chisso Corp.) with lithium aluminum
hydride according to the procedure given for 1d (vide
infra).
3.3. p-Methoxyphenylsilane (1b)
3.2. p-Methylthiophenylsilane (1d)
B.p. 75°C/10 mmHg (lit. [19] 180°C/740 mmHg).
1H-NMR: l 3.33 (s, 3H), 4.28 (s, 3H), 6.71 (d, 2H),
7.33 (d, 2H). 29Si-NMR: l −60.26. IR (neat): 3008,
2960, 2838, 2532, 2156, 1897, 1599, 1566, 1504, 1464,
1441, 1400, 1311, 1282, 1251, 1183, 1118, 1035, 922,
874, 826, 795, 754, 731, 646, 627, 509, 487 cm−1. MS
(relative intensity): m/z 138 [M+, 29], 137 (15), 122, (7),
121 (13), 108 (8), 107 (17), 91 (20), 79 (13), 78 (21), 77
(29), 67 (10), 66 (15), 65 (19), 61 (13), 60 (26), 59 (100),
55 (9), 53 (24), 51 (12), 50 (9), 43 (10). HRMS: Calc. for
C7H10OSi [M+] 138.0500. Found 138.0515. Anal. Calc.
for C7H10OSi: C, 60.82; H, 7.29. Found: C, 60.64; H,
8.05%.
3.2.1. Generation of p-methylthiophenylmagnesium
bromide
A 300 ml round-bottom flask equipped with a reflux
condenser and a pressure-equalizing dropping funnel
was charged with 2.99 g (123 mmol) of Mg and 100 ml
of Et2O. The magnesium turnings were activated by the
addition of a few drops of 1,2-dibromoethane and a
solution of p-bromothioanisole (25 g, 123 mmol) dis-
solved in 50 ml of Et2O was slowly added at a rate to
maintain a gentle reflux. After the addition the solution
was further stirred at r.t. for 12 h.
3.2.2. Generation of (p-methylthiophenyl)trichlorosilane
In a 500 ml round-bottom flask equipped with a
reflux condenser and pressure-equalizing dropping fun-
nel were placed 200 ml of Et2O and 50.0 g of SiCl4 (198
mmol). The foregoing solution of the Grignard reagent
was slowly added through the funnel to the SiCl4–Et2O
3.4. p-Dimethylaminophenylsilane (1c)
1
M.p. 43–45°C (lit. [20] 49–50°C). H-NMR: l 2.42
(s, 3H), 4.46 (s, 3H), 6.51 (d, 2H), 7.46 (d, 2H).
29Si-NMR (DEPT): l −60.86. IR (C6H6): 3092, 3074,
3038, 2328, 2152, 1961, 1816, 1601, 1516, 1481, 1357,

Y. Obora, M. Tanaka / Journal of Organometallic Chemistry 595 (2000) 1–11
9
1178, 1118, 1036, 922, 851, 810, 675 cm−1. MS (relative
intensity): m/z 152 [M++1, 11], 151 [M+, 69], 150
(52), 148 (17), 134 (7), 121 (74), 120 (100), 119 (12), 107
(16), 106 (7), 105 (20), 104 (10), 103 (13), 93 (7), 91 (9),
79 (11), 78 (8), 76 (24), 74 (21), 73 (21), 72 (10), 67 (9),
66 (11), 65 (7), 56 (8), 55 (10), 54 (8), 53 (34), 44 (17),
43 (8), 42 (21). HRMS: Calc. for C8H13NSi [M+]
151.0817. Found 151.0801. Anal. Calc. for C8H13NSi:
C, 63.52; H, 8.66. Found: C, 63.18; H, 8.71%.
MS (relative intensity): m/z 408 [M+, 13], 393 (25), 381
(36), 355 (12), 347 (13), 345 (21), 343 (51), 331 (38), 330
(18), 328 (18), 319 (18), 305 (18), 293 (75), 292 (19), 290
(19), 281 (70), 269 (45), 267 (20), 255 (30), 243 (83), 231
(100), 229 (20), 227 (34), 225 (36), 219 (90), 217 (19),
212 (20), 205 (37). HRMS: Calc. for C18H19ClO3Zr
408.0069. Found 408.0072. Anal. Calc. for
C18H19ClO3Zr: C, 52.73; H, 4.67. Found: C, 52.52; H,
4.71%.
3.5. Chloro(phenoxy)bis(cyclopentadienyl)zirconium (3c)
3.8. Dehydrogenati6e coupling of silanes with
chloro(aryloxy)zirconocene complex catalysts
To a mixture of Cp2ZrCl(Me) (300 mg, 1.10 mmol)
and toluene (20 ml) placed in a 30 ml Schlenk flask was
slowly added a solution of phenol (103 mg, 1.10 mmol)
in toluene (10 ml) over a period of 20 min. The mixture
was further stirred for 20 h at r.t., evaporated in vacuo,
and the residue was extracted with 50 ml of hexane.
After evaporation of the hexane solution to ca. 10 ml,
the residual solution was cooled to −78°C to give 142
A typical reaction procedure (entry 1): A 30 ml J.
Young-valve-equipped Schlenk flask was loaded in a
glove box with 3a (20 mg, 0.04 mmol) and toluene (0.2
ml). One equivalent of n-BuLi (28 ml, 1.6 M solution in
hexane) was added to the flask at −78°C and the
temperature was raised to 0°C. After the mixture was
stirred for 30 min at 0°C, phenylsilane (0.8 ml, 6.4
mmol) was added, and the mixture was set at r.t.
(23°C). Intense evolution of H2 gas was immediately
observed. After the reaction over 120 h, a small amount
of THF (0.2 ml) was added to deactivate the catalyst
and an aliquot of the mixture was analyzed by GPC
with oxygen-free toluene eluent. The remainder of the
solution was evaporated in vacuo (23°C/0.1 mmHg) to
give a viscous polymeric material (660 mg inclusive of
the catalyst residue, ca. 94% based on phenylsilane
charged), which was fractionated by preparative GPC
with oxygen-free toluene eluent to give a higher molec-
ular weight fraction of poly(phenylsilane) (380 mg,
56%).
1
mg (37%) of a white solid. M.p. 217–219°C dec. H-
NMR: l 5.92 (s, 10H), 6.67–7.20 (m, 5H). 13C-NMR: l
114.33, 115.60, 118.04, 120.57, 129.79. MS (relative
intensity): m/z 348 [M+, 14], 331 (20), 283 (13), 269
(18), 255 (23), 243 (40), 231 (48), 219 (45), 205 (20), 193
(33), 181 (100), 169 (100), 162 (35), 124 (13), 100 (88).
HRMS: Calc. for C16H15ClOZr 347.9858. Found
347.9874. Anal. Calc. for C16H15ClOZr: C, 54.91; H,
4.32. Found: C, 55.06; H, 4.27%.
3.6. Chloro(2,3,4,5,6-pentafluorophenoxy)bis-
(cyclopentadienyl)zirconium (3d)
A procedure similar to that for 3c, using 2,3,4,5,6-
pentafluorophenol (202 mg, 1.10 mmol), gave 77 mg
(16% yield) of the title compound as a white solid. M.p.
150–152°C. 1H-NMR: l 5.89 (s, 10H). MS (relative
intensity): m/z 438 [M+, 62], 404 (16), 337 (15), 294
(38), 292 (57), 291 (17), 290 (58), 274 (17), 261 (21), 259
(71), 258 (18), 257 (100), 256 (40), 213 (17), 229 (52),
228 (16), 227 (78), 226 (18), 225 (86), 213 (30), 211 (54),
210 (19), 209 (69). HRMS: Calc. for C16H10ClF5OZr
437.9387. Found 437.9434. Anal. Calc. for
C16H10ClF5OZr: C, 43.68; H, 2.29. Found: C, 43.27; H,
2.17%.
In a repeated experiment, the product was fraction-
ated according to Harrod’s procedure as follows. First,
the reaction mixture diluted with toluene was filtered
through a Florisil column to remove the catalyst. Evap-
oration of the filtrate gave 620 mg (91%) of a raw
polymer, GPC analysis of which showed Mw=9.9×
103 and Mn=2.3×103. Then the raw polymer was
dissolved in 2 ml of toluene and the mixture was
poured into 4 ml of petroleum ether to precipitate a
linear fraction (360 mg, 53%) of Mw=13.9×103 and
Mn=4.3×103. The structure was confirmed by spec-
tral comparison with the literature data [9j%].
3.7. Chloro(2,6-dimethoxyphenoxy)bis(cyclopenta-
dienyl)zirconium (3e)
3.9. Time course monitoring of the dehydrogenati6e
coupling of phenylsilane (Fig. 1)
A procedure similar to that for 3c, starting with 200
mg of Cp2Zr(Me)Cl (0.74 mmol) and 115 mg of 2,6-
dimethoxyphenol (0.74 mmol), gave 48 mg (16%) of the
title compound as a yellow waxy solid. 1H-NMR: l
3.45 (s, 6H), 6.11 (s, 10H), 6.44–6.76 (m, 3H). 13C-
NMR: l 56.14, 106.61, 114.20, 114.80, 115.83, 118.98.
3.9.1. A typical reaction procedure
A 30 ml J. Young-valve-equipped Schlenk flask was
loaded in a glove box with 3a (10 mg, 0.02 mmol),
toluene-d8 (0.5 ml) and mesitylene (50 ml, an internal
standard for 1H-NMR spectroscopy). One equivalent of

10
Y. Obora, M. Tanaka / Journal of Organometallic Chemistry 595 (2000) 1–11
n-BuLi (14 ml, 1.6 M solution in hexane) was added to
the flask at −78°C and the temperature was raised to
0°C. After the mixture was stirred for 30 min at 0°C,
the system was opened to the constant flow of nitrogen
gas, the purity of which was\99.9995%. Then the flask
was warmed to r.t. by external heating with a water
bath, and stirred for 5 min at r.t. Phenylsilane (0.8 ml,
6.4 mmol) was injected to the flask. A 20 ml portion of
the mixture was taken by a micro syringe and was
immediately added to ꢀ5 ml of THF placed in an
NMR tube to deactivate the catalyst. An NMR sample
was prepared by adding C6D6 (0.5 ml) to this mixture.
The formation of poly(phenylsilane) and consumption
of the monomer were easily monitored by 1H-NMR
spectroscopy with probes of the SiH signals at 4.3–5.3
ppm for poly(phenylsilane) and 4.26 ppm for PhSiH3,
relative to the signal due to mesitylene at 2.15 ppm.
A solution for GPC analysis was similarly prepared.
Calc. for C7H8SSi: C, 55.21; H, 5.29. Found: C, 54.80;
H, 5.16%.
Acknowledgements
We are grateful to the Japan Science and Technology
Corporation (JST) for partial financial support through
the CREST (Core Research for Evolutional Science
and Technology) program.
References
[1] (a) M. Zeldin, K.J. Wynne, H.R. Allcock, Inorganic and
Organometallic Polymers, ACS Symposium Series 360, Ameri-
can Chemical Society, Washington, DC, 1988. (b) J.M. Zeigler,
F.W.G. Fearon, Silicon-Based Polymer Science: A Comprehen-
sive Resources, Advances in Chemistry Series 224, American
Chemical Society, Washington, DC, 1990.
[2] (a) J. Maxka, F. Mitter, D. Powell, R. West, Organometallics 10
(1991) 660. (b) A.R. Wolff, I. Nozue, J. Maxka, R. West, J.
Polym. Sci. Part A: Polym. Chem. 26 (1988) 701. (c) J. Michl,
J.W. Dowing, T. Karatsu, A.J. McKinley, G. Poggi, G.M.
Wallraff, R. Sooriyakumaran, R.D. Miller, Pure Appl. Chem. 60
(1988) 939.
[3] (a) Y.P. Sun, G.M. Wallraff, R.D. Miller, J. Michl, J. Pho-
tochem. Photobiol. A: Chem. 62 (1992) 333. (b) Y.P. Sun, J.
Michl, J. Am. Chem. Soc. 114 (1992) 8186. (c) Y.P. Sun, Y.
Hamada, L.M. Huang, J. Maxka, J.S. Hsiao, R. West, J. Michl,
J. Am. Chem. Soc. 114 (1992) 6301.
[4] (a) P. Trefonas III, J.R. Damewood Jr., R. West, R.D. Miller,
Organometallics 4 (1985) 1318. (b) Y.-P. Sun, R.D. Miller, R.
Sooriyakumaran, J. Michl, J. Inorg. Organomet. Polym. 1 (1991)
3. (c) H.S. Plitt, J. Michl, Chem. Phys. Lett. 198 (1992) 430. (d)
R. Imhof, D. Antic, D.E. David, J. Michl, J. Phys. Chem. A 101
(1997) 4579. (e) E.K. Karikari, A.J. Greso, B.L. Famer, R.D.
Miller, J.F. Rabolt, Macromolecules 26 (1993) 3937. (f) S.
Mazie`res, M.K. Raymond, G. Raabe, A. Prodi, J. Michl, J. Am.
Chem. Soc. 119 (1997) 6682. (g) R. Imhof, H. Teramae, J. Michl,
Chem. Phys. Lett. 270 (1997) 500.
3.10. Poly(p-methoxyphenylsilane) (2b)
Sticky semi-solid, yield 32% (after fractionation),
Mw=2806, Mn=793, which corresponds to a 5.8-mer.
1H-NMR: l 3.25 (br, s, 3H), 4.35–5.05 (br, s, SiꢀH),
5.05–5.40 (br, s, SiꢀH in small cyclics), 6.40–7.70 (m,
4H). 13C-NMR: l 54.41, 114.14, 122.18, 138.62, 160.77.
IR (C6H6): 3008, 2934, 2838, 2690, 2526, 2460, 2384,
2292, 2098, 1978, 1893, 1804, 1773, 1593, 1564, 1499,
1462, 1398, 1311, 1247, 1181, 1100, 1031, 959, 915, 743,
696, 600, 518, 493, 466 cm−1. UV (GPC): umax 314 nm.
Anal. Calc. for (C7H8OSi)5.8H2, a 5.8-mer: C, 61.56; H,
6.16. Found: C, 61.54; H, 5.72%.
3.11. Oligo(p-dimethylaminophenylsilane) (2c)
Sticky oil, yield 39% (after fractionation), Mw=971,
1
Mn=564, which corresponds to a 3.7-mer. H-NMR: l
[5] (a) R.D. Miller, R. Sooriyakumaran, Macromolecules 21 (1988)
3120. (b) K. Pannell, J.M. Rozell, J.M. Zeigler, Macromolecules
21 (1988) 276.
[6] (a) R.D. Miller, J. Michl, Chem. Rev. 89 (1989) 1359. (b) R.D.
Miller, D. Thompson, R. Sooriyakumaran, G.N. Fickes, J.
Polym. Sci. Part A: Polym. Chem. 29 (1991) 813. (c) R.G. Jones,
R.E. Benfield, P.J. Evans, S.J. Holder, J.A.M. Locke, J.
Organomet. Chem. 521 (1996) 171, and references cited therein.
[7] C. Aitken, J.F. Harrod, E. Samuel, J. Organomet. Chem. 269
(1985) C11.
2.45 (br, s, 6H), 4.80–5.42 (br, s, SiꢀH), 6.32–6.60 (m,
2H), 7.39–7.95 (m, 2H). 13C-NMR: l 39.74, 112.74,
116.94, 137.96, 151.08. IR (C6H6): 3088, 2888, 2814,
2148, 1885, 1601, 1543, 1516, 1481, 1446, 1359, 1325,
1284, 1230, 1207, 1118, 1064, 1006, 922, 808, 731, 677,
648, 629, 512, 474 cm−1. UV (GPC): umax 326 nm.
Anal. Calc. for (C8H11NSi)3.7H2, a 3.7-mer: C, 64.14; H,
7.76. Found: C, 63.91; H, 7.34%.
[8] For reviews see: (a) J.F. Harrod, in: K.J. Smith, E.C. Sanford
(Eds.), Progress in Catalysis, Elsevier, Amsterdam, 1992, p. 147.
(b) T.D. Tilley, Acc. Chem. Res. 26 (1993) 22, and references
cited therein. (c) Y. Mu, J.F. Harrod, in: J.F. Harrod, R.M.
Laine (Eds.), Inorganic and Organometallic Polymers and
Oligomers, Kluwer, Dordrecht, the Netherlands, 1991, p. 23. (d)
J.Y. Corey, J. Braddock-Wilking, Chem. Rev. 99 (1999) 175.
[9] (a) B.J. Grimmond, J.Y. Corey, Organometallics 18 (1999) 2223.
(b) J.L. Huhmann, J.Y. Corey, N.P. Rath, J. Organomet. Chem.
533 (1997) 61. (c) G.L. Casty, C.G. Lugmair, N.S. Radu, T.D.
Tilley, J.F. Walzer, D. Zargarian, Organometallics 16 (1997) 8.
(d) N. Choi, S. Onozawa, T. Sakakura, M. Tanaka,
Organometallics 16 (1997) 2765. (e) V.K. Dioumaev, J.F. Har-
3.12. Higher-molecular-weight fraction of
poly(p-methylthiophenylsilane) (2d)
Sticky solid, yield 62% (after fractionation). 1H-
NMR: l 1.96 (br, s, 3H), 4.38–4.81 (br, s, SiꢀH),
4.90–5.15 (br, s, SiꢀH in small cyclics), 6.65–7.65 (m,
4H). IR (C6H6): 3064, 3028, 3004, 2922, 2364, 2344,
2106, 1901, 1578, 1539, 1495, 1483, 1437, 1383, 1321,
1253, 1197, 1081, 1015, 957, 913, 803, 748, 731, 694,
650, 605, 520, 489, 464. UV (GPC): umax 314 nm. Anal.

Y. Obora, M. Tanaka / Journal of Organometallic Chemistry 595 (2000) 1–11
11
rod, J. Organomet. Chem. 521 (1996) 133. (f) V.K. Dioumaev,
J.F. Harrod, Organometallics 15 (1996) 3859. (g) R.M. Shaltout,
J.Y. Corey, Tetrahedron 51 (1995) 4309. (h) S. Bourg, R.J.P.
Corriu, M. Enders, J.J.E. Moreau, Organometallics 14 (1995)
564. (i) V.K. Dioumaev, J.F. Harrod, Organometallics 13 (1994)
1548. (j) T. Imori, T.D. Tilley, Polyhedron 13 (1994) 2231. (k) H.
Li, F. Gauvin, J.F. Harrod, Organometallics 12 (1993) 575. (l)
J.Y. Corey, J.L. Huhmann, X.-H. Zhu, Organometallics 12
(1993) 1121. (m) J.P. Banovetz, H. Suzuki, R.M. Waymouth,
Organometallics 12 (1993) 4700. (n) E. Hengge, M. Weinberger,
J. Organomet. Chem. 443 (1993) 167. (o) T. Kobayashi, T.
Sakakura, T. Hayashi, M. Yumura, M. Tanaka, Chem. Lett. 7
(1992) 1157. (p) J.Y. Corey, X.-H. Zhu, J. Organomet. Chem.
439 (1992) 1. (q) E. Hengge, M. Weinberger, J. Organomet.
Chem. 441 (1992) 397. (r) J.F. Harrod, Y. Mu, E. Samuel, Can.
J. Chem. 70 (1992) 2980. (s) H.-G. Woo, R.H. Heyn, T.D. Tilley,
J. Am. Chem. Soc. 114 (1992) 5698. (t) H.-G. Woo, J.F. Walzer,
T.D. Tilley, J. Am. Chem. Soc. 114 (1992) 7047. (u) J.F. Harrod,
Y. Mu, E. Samuel, Polyhedron 10 (1991) 1239. (v) J.E. Durham,
J.Y. Corey, W.J. Welsh, Macromolecules 24 (1991) 4823. (w)
H.-G. Woo, J.F. Walzer, T.D. Tilley, Macromolecules 24 (1991)
6863. (x) J.Y. Corey, X.-H. Zhu, T.C. Bedard, L.D. Lange,
Organometallics 10 (1991) 924. (y) C.M. Forsyth, S.P. Nolan,
T.J. Marks, Organometallics 10 (1991) 2543. (z) J.P. Banovetz,
K.M. Stein, R.M. Waymouth, Organometallics 10 (1991) 3430.
(a%) T. Sakakura, H.J. Lautenschlager, M. Nakajima, M.
Tanaka, Chem. Lett. 6 (1991) 913. (b%) F. Gauvin, J.F. Harrod,
Can. J. Chem. 18 (1990) 638. (c%) J.F. Harrod, T. Ziegler, V.
Tschinke, Organometallics 9 (1990) 897. (d%) C. Aitken, J.P.
Barry, F. Gauvin, J.F. Harrod, A. Malek, D. Rousseau,
Organometallics 8 (1989) 1732. (e%) L.S. Chang, J.Y. Corey,
Organometallics 8 (1989) 1885. (f%) W.H. Cambell, T.K. Hilty, L.
Yurga, Organometallics 8 (1989) 2615. (g%) T. Nakano, H. Naka-
mura, Y. Nagai, Chem. Lett. 1 (1989) 83. (h%) J.Y. Corey, L.S.
Chang, E.R. Corey, Organometallics 6 (1987) 1595. (i%) J.F.
Harrod, S.S. Yun, Organometallics 6 (1987) 1381. (j%) C. Aitken,
J.F. Harrod, U.S. Gill, Can. J. Chem. 65 (1987) 1804.
[10] (a) J.A. Ewen, J. Am. Chem. Soc. 106 (1984) 6355. (b) W.
Kaminsky, K. Kulper, H. Brintzinger, F.R.W.P. Wild, Angew.
Chem. Int. Ed. Engl. 24 (1985) 507.
[11] T. Repo, G. Jany, M. Salo, M. Palomo, M. Leskela, J.
Organomet. Chem. 541 (1997) 363.
[12] (a) E. Hengge, J. Organomet. Chem. Libr. 9 (1980) 261. (b) E.
Hengge, D. Kovar, H.P. So¨lladl, Monatsch. Chem. 110 (1979)
805.
[13] D.A. Stanislawski, R. West, J. Organomet. Chem. 204 (1981)
295.
[14] Y. Obora, M. Tanaka, manuscript in preparation.
[15] A. Antin˜olo, G.S. Bristow, G.K. Campbell, A.W. Duff, P.B.
Hitchcock, R.A. Kamarudin, M.F. Lappert, R.J. Norton, N.
Sarjudeen, D.J.W. Winterborn, J.L. Atwood, W.E. Hunter, H.
Zheng, Polyhedron 8 (1989) 1601.
[16] (a) P.C. Wailes, H. Weigold, J. Organomet. Chem. 24 (1970)
405. (b) P.C. Wailes, H. Weigold, A.P. Bell, J. Organomet.
Chem. 33 (1971) 181.
[17] A.G. Massey, A.J. Park, J. Organomet. Chem. 2 (1964) 245.
[18] S. Rozite, I. Mazeika, A. Gaukman, N. Erchak, E. Lukevics, J.
Organomet. Chem. 348 (1988) 169.
[19] S.D. Rosenberg, J.J. Walburn, H.E. Ramsden, J. Org. Chem. 21
(1957) 1606.
[20] Y. Nagai, M. Ohtsuki, T. Nakano, H. Watanabe, J. Organomet.
Chem. 35 (1972) 81.
.
.