2 Early Soluble Unbridged Metallocene Catalysts
in Connection with Alkyl Aluminum Chlorides
Due to the complexity of the solid Ziegler–Natta catalysts, homogeneous systems
based on bis(cyclopentadienyl) group IV compounds, in particular on hydrocarbonsoluble titanocene complexes, gained increasing interest in the hope for more
mechanistic insight. They were discovered in the 1950s [4–7], shortly after the
appearance of Ziegler’s and Natta’s reports on solid state catalysts. Alkyl aluminum
compounds such as AlEt 2 Cl or AlEtCl 2 are required to activate these soluble catalysts.
The central problem of the insertion polymerization with Ziegler catalysts is
the lack of detailed knowledge concerning the nature, the lifetime, and the exact
concentration of the polymerization active species. These questions can be met – in
our opinion – only via a quantitative detection of all elementary reactions involved
in the Ziegler–Natta catalysis.
A useful concept pointing the way for the description of these metallocene
catalyst systems goes back to the kinetic studies of Reichert and Meyer [8–10].
In [8] in the summary Reichert wrote:
The rate of polymerization of ethylene by the soluble Ziegler–Natta catalysts Cp 2 TiEtCl/
AlEtCl 2 was investigated in toluene as solvent. From the dependence of the initial rate of
the polymerization follows, that the primary complex, formed by the catalyst components,
is not the active species of this system. The obtained dependences rather suggest, that the
active species is to be viewed as an equilibrium product and the location of this equilibrium
is determined by the ratio of the catalyst compounds in that manner, that only at high
Al/Ti-ratios the charged Ti(IV)-compound is equal to the initial concentration of the active
species.
In the summary of [10], he continues: “The active species of this catalyst system
is formed extremely rapid and seems to be a very dynamic equilibrium product.
Its concentration depends on the ratio Al/Ti. The propagation reaction therefore
corresponds to an intermitting process.”
Hence, during the growth of a polymer chain, each metal species with a
pendant chain appears to alternate between a “dormant” state and a state in which
it actively grows. This “intermittent-growth” model was further elaborated by Fink
and coworkers [11–19] in extensive kinetic and reactivity studies using precise
experimental tools designed especially for dealing with fast reactions (rapid in
situ start in the stirred back-mix reactor with highly sensitive and rapid measuring
of monomer consumption, plug-flow and stop-flow reactor, dynamic
13 C NMR
spectroscopy, use of
13 C-enriched ethylene as a molecular surveyor during the
polymerization). For the first time, these experiments demonstrated a sequence
of very fast elementary catalyst-forming reactions and also the dynamic alkyl
and chloride exchange reactions inherent in these homogeneous polymerization
systems.
It was Fink’s conclusion that, as a consequence of these successive equilibria
and their very different locations, the dependence of the initial polymerization rate
on the charged Al concentration (i.e., the Al isotherm) has to show a pronounced
sigmoid curve course in the initial part. Namely, in this case in the overall rate
6
G. Fink
in Connection with Alkyl Aluminum Chlorides
Due to the complexity of the solid Ziegler–Natta catalysts, homogeneous systems
based on bis(cyclopentadienyl) group IV compounds, in particular on hydrocarbonsoluble titanocene complexes, gained increasing interest in the hope for more
mechanistic insight. They were discovered in the 1950s [4–7], shortly after the
appearance of Ziegler’s and Natta’s reports on solid state catalysts. Alkyl aluminum
compounds such as AlEt 2 Cl or AlEtCl 2 are required to activate these soluble catalysts.
The central problem of the insertion polymerization with Ziegler catalysts is
the lack of detailed knowledge concerning the nature, the lifetime, and the exact
concentration of the polymerization active species. These questions can be met – in
our opinion – only via a quantitative detection of all elementary reactions involved
in the Ziegler–Natta catalysis.
A useful concept pointing the way for the description of these metallocene
catalyst systems goes back to the kinetic studies of Reichert and Meyer [8–10].
In [8] in the summary Reichert wrote:
The rate of polymerization of ethylene by the soluble Ziegler–Natta catalysts Cp 2 TiEtCl/
AlEtCl 2 was investigated in toluene as solvent. From the dependence of the initial rate of
the polymerization follows, that the primary complex, formed by the catalyst components,
is not the active species of this system. The obtained dependences rather suggest, that the
active species is to be viewed as an equilibrium product and the location of this equilibrium
is determined by the ratio of the catalyst compounds in that manner, that only at high
Al/Ti-ratios the charged Ti(IV)-compound is equal to the initial concentration of the active
species.
In the summary of [10], he continues: “The active species of this catalyst system
is formed extremely rapid and seems to be a very dynamic equilibrium product.
Its concentration depends on the ratio Al/Ti. The propagation reaction therefore
corresponds to an intermitting process.”
Hence, during the growth of a polymer chain, each metal species with a
pendant chain appears to alternate between a “dormant” state and a state in which
it actively grows. This “intermittent-growth” model was further elaborated by Fink
and coworkers [11–19] in extensive kinetic and reactivity studies using precise
experimental tools designed especially for dealing with fast reactions (rapid in
situ start in the stirred back-mix reactor with highly sensitive and rapid measuring
of monomer consumption, plug-flow and stop-flow reactor, dynamic
13 C NMR
spectroscopy, use of
13 C-enriched ethylene as a molecular surveyor during the
polymerization). For the first time, these experiments demonstrated a sequence
of very fast elementary catalyst-forming reactions and also the dynamic alkyl
and chloride exchange reactions inherent in these homogeneous polymerization
systems.
It was Fink’s conclusion that, as a consequence of these successive equilibria
and their very different locations, the dependence of the initial polymerization rate
on the charged Al concentration (i.e., the Al isotherm) has to show a pronounced
sigmoid curve course in the initial part. Namely, in this case in the overall rate
6
G. Fink