One of the defining characteristics of these catalytic processes is the presence of crystalline structures with coordination networks. This feature—which distinguishes these catalysts from those with molecular network structures—is characterized by the presence of vacant sites or ligands, and it is this feature that gives its name to the generic catalytic process known as coordination polymerization. According to theories regarding the conformation of crystalline structures, these vacancies are necessary to achieve the crystal’s complete electroneutrality.
One of the typical components of the catalyst, titanium trichloride, exists in the α, β, γ, and δ forms, depending on its crystal structure.
The α form (and, to a lesser extent, the γ form) is the characteristic form and the most commonly found. It consists of layers of titanium atoms alternating with pairs of layers of chlorine atoms.
Schematic diagram of the arrangement of Ti and Cl atoms in α-TiCl₃
In this figure, Ti and Cl atoms are represented by small gray spheres and large yellow spheres, respectively. The Ti atoms are located in the octahedral interstices of the chlorine lattice, which, in turn, are packed into compact hexagonal structures. Every third Ti atom is missing from the crystal structure.
Active sites form at the Ti atoms (each of which has a vacant ligand associated with it) in this lattice.
"Activation" (the formation of an RCl₄Ti group, or, more generally, an RxClyTi group, where represents the vacancy, and x + y = 5) is achieved by replacing chlorine atoms in the original lattice with alkyl groups.
Based on an analysis of the crystal structure, it has been concluded that the truly active sites are those located at the edges of the crystal structure and not those located elsewhere in the crystal. This theory is further supported by studies of polymer growth from the crystal, in which it was observed that chains are generated only from the edges of the crystal.
The following figure schematically illustrates Arlmann-Cossee’s interpretation of the isotactic polymerization process of propylene to produce the isotactic form of PP, based on an RCl₄Ti active site. Parts (a) and (b) of the figure indicate the location in the crystal where a chlorine atom will first be replaced by an alkyl group and then by the incoming monomer molecules.
The monomer incorporated into the growing polymer chain (the rest of the polymer chain is not shown in the figure for clarity) is the large blue sphere seen coordinated with the vacancy in the active site located on the surface of the crystal lattice. The differences between the parts of the figure are related to the differences in position of two monomer molecules successively added to the active site.
As can be seen, the preference for monomer entry (ease of access to the active site) favors insertion according to the scheme in c, since that site is more accessible. However, because strict alternation is required, once the molecule has entered the most accessible position, it must migrate to the least accessible position to allow the other molecule to enter.
In detail: the strict alternation between the positions in (c) and (d) of the newly incorporated molecule generates the necessary stereospecificity: the mechanism proposed by these authors assumes that:
1) The monomer assumes the most favorable steric arrangement (and thus determines the position of the associated methyl group, also due to steric effects), and then
2) Move to the position indicated in (d) to allow a new one to be added according to the diagram in (c)
For this explanation to hold, the insertion rate must be lower than the migration rate, so that the molecules always insert in the same arrangement and their side groups always occupy the same relative position.
Thus, for the unsupported (C2H5)3Al-α-TiCl3 system, the predominant trend is the production of isotactic polypropylene at temperatures between 50 and 100 C. Analysis of the Arrhenius constants for the kinetics of the insertion and migration processes revealed that low temperatures favor the formation of syndiotactic structures, as they have a stronger effect on the migration rate (which decreases more rapidly than the insertion rate). For the same reason, higher temperatures would accelerate migration more rapidly and lead to an increase in atacticity, as they allow for more than one positional change via migration between two successive insertions (the strength of the coordination bond between the monomer and the catalyst is reduced by the increase in thermal energy, which confers greater individual activity to the monomer molecule).
The strong dependence of polymerization stereospecificity on the crystalline structure of the catalyst is demonstrated by the effect of changing the crystalline form of the catalyst under the same co-catalysis conditions.
* The γ form behaves almost like the α form, as it has similar crystal structures (the same “sandwich” arrangement of Cl-Ti-Cl layers), with the difference that the Cl atoms are arranged in a cubic lattice rather than a hexagonal one.
** The δ variety is usually obtained by grinding the α and γ varieties in a ball mill; therefore, it consists of a mixture of three layers with combined hexagonal and cubic chlorine crystal structures. The catalytic activity of the α and γ forms is generally lower than that of the δ form.
*** In contrast, the β form has a different crystal structure, with clusters of linear chains of titanium trichloride.
As a result of this arrangement, Ti atoms are characterized by the presence of two vacancies, which ensures a high degree of randomness in the sequential position of the insertions and, therefore, atacticity. For the reasons stated above, the β form has traditionally been considered of little commercial importance, since it was initially believed to produce only the amorphous variety of PP.
Diagram of the β structure
However, publications that appeared in the 1980s in the patent literature describe modified catalysts with high stereospecificity obtained through chemical treatments applied to the β variety. In other words, by appropriately manipulating the effects of different cocatalysts, it is possible to alter the behavior of the active site in a way that results in stereospecificity.