Organochromium catalysts with chromium oxide

Although there are many differences between chromium oxide catalysts and the organochromium catalysts, when they are bonded to the support, organochromium catalysts usually display a similar, but exaggerated, MW response in the polymer produced relative to what is observed with chromium oxide catalysts. For example, the MW of polymer produced with each type of catalyst usually decreased as the support calcination temperature was raised. Similarly, when both chromium oxide and the organochromium compounds were deposited onto aluminophosphate supports, they always yielded lower-MW polymer as the amount of phosphate in the support was raised. 

The tetravalent chromium alkyl compounds were found to give catalysts that are somewhat more active than the catalyst made from the divalent chromium counterpart, under commercial reaction conditions (90-110 °C, 0.5-1.5 mol ethylene L ). Indeed, they were among the most active organochromium catalysts tested in our laboratory. Their overall 1-h yield was usually also superior to that observed with some of the best chromium oxide on silica-titania catalysts. Even when compared with chromium oxide systems used with a cocatalyst, the catalysts made with tetravalent chromium alkyls were equal or better in activity. Unfortunately, for commercial applications, these catalysts also tend to make some oligomers and wax as well. 

In addition to incorporated oligomers, which produce even-numbered branches, methyl branching is also detected in small amounts in the polymers made with many organochromium (but not chromium oxide) catalysts. Chromocene is especially known for this behavior [303,654,679,680]. It is usually thought to result from (3-hydride elimination to the chromium, followed by reinsertion of the same chain or (perhaps a comonomer) in the backwards 2,1 position. The number of methyl branches formed is usually not large enough to have a significant effect on the resin density. 

HDPE resias are produced ia industry with several classes of catalysts, ie, catalysts based on chromium oxides (Phillips), catalysts utilising organochromium compounds, catalysts based on titanium or vanadium compounds (Ziegler), and metallocene catalysts (33—35). A large number of additional catalysts have been developed by utilising transition metals such as scandium, cobalt, nickel, niobium, molybdenum, tungsten, palladium, rhodium, mthenium, lanthanides, and actinides (33—35) none of these, however, are commercially significant. 

Many other organochromium compounds have since been synthesized and found to be active, including those with chromium exhibiting every valence up to Cr (IV). Chromocene is a well-studied example of an active divalent compound (52-55). Pentadiene-Cr(II) (56) is another, along with allyl-Cr(II) (52, 57). Allyl-Cr(III) is also active (52, 57-61). -Stabilized alkyls of Cr(II) and Cr(IV) such as trimethylsilylmethyl-Cr(IV), which also polymerizes ethylene when supported on an oxide carrier, have been synthesized and tested in this laboratory (57,62). All these organochromium catalysts are comparable in activity to the Cr(VI)/silica standard. 

The active site concentration on the organochromium catalysts may be higher than that of the oxide catalysts. The activity usually assumes a more linear increase with chromium loading than on the oxide catalysts, at least up to 2% Cr. Yermakov and Zakharov, studying allyl-Cr(III)/silica catalysts, stopped the polymerization with radioactive methanol, and found that the kill mechanism is different from that on the oxide catalysts (59). The proton of the methanol, and not the alkoxide, became attached to the polymer. This suggests a polarity opposite to that of the oxide catalysts, with the site being more positive than the chain. 

In the literature, most of the early discussion of the "active" valence is in reference to silica-supported chromium oxide catalysts. However, many organochromium compounds of widely differing valence are also known to be active upon contact with a support and subsequent exposure to ethylene. For example, as early as 1961, Walker et al. showed that diare-nechromium(O) compounds polymerize ethylene when deposited onto silica or another support [280,281]. The Cr(0) is probably oxidized by silanol groups to Cr(I), consistent with the inference that it too can be an active precursor. 

A number of organochromium compounds form highly active polymerization catalysts when deposited onto an already calcined oxide carrier. Usually the carrier plays an essential role, because without it such compounds rarely exhibit any activity. In many respects, the organochromium catalysts are quite different from their chromium oxide counterparts. Examples of organochromium compounds which form active catalysts include those with chromium in any of the formal chromium valences from Cr(0) to Cr(IV). 

SCHEME 39 The similarity of polymers made with organochromium and chromium oxide catalysts suggests some common active species, perhaps like those proposed here. (R is a generic alkyl group, not necessarily the same in each species.)... 

Chromium oxide on aluminophosphate produces polymers having a broader MW distribution than its Cr/silica counterparts, which is evidence of greater heterogeneity of Cr species on the catalyst surface. Organochromium compounds on aluminophosphate also produce polymers having broad MW distributions, and with these catalysts these same trends become unusually clear. Perhaps because the chromium tends to bind through only one link to the surface instead of two, it is often possible to obtain more detailed information about the catalyst from the resultant polymer. 

In this way the subtleties of the normal commercial polyethylene grades are reproduced faithfully, but without the expense of purchasing, purifying, and storing a-olefin comonomer. Continuous addition of the organochromium compound directly into the reactor, or with the chromium oxide catalyst to a pre-contacting vessel which flows to the reactor, provides precise and instantaneous control of the resin density (i.e. level of branching). 

The valence of the starting organochromium compound has been varied from Cr(0) to Cr(IV), but seems to make little difference. All species are quite active, and all initiate polymerization rapidly in comparison to the oxide catalysts. There is no induction time, since the chromium is already reduced, and no gradual rise in rate. Polymerization usually starts immediately on contact with ethylene and either holds steady or slowly declines during a 1 hr run. 

The Phillips catalyst is not alkylated when it goes into the reactor, and metal alkyl cocatalysts are not normally used. Thus, in contrast to Ziegler, Ballard, or metallocene catalysts, the Phillips catalyst has no Cr-alkyl bond into which ethylene may be inserted. Instead, the chromium somehow reacts with ethylene to generate such a bond. This characteristic is not unique, as many catalyst types also display this ability.8 This issue has been the source of much interest and speculation for half a century. On some catalysts, CO reduction is known to cleanly produce Cr(II). Reaction with ethylene could involve a formal oxidation [52,94,141,250-252,269,322-325,339-345] and many pathways involving Cr(IV) have been proposed, sometimes based on organochromium analogs, such as shown in Scheme 8 [94,250-252,315-319,321-325,342,346-349]. 

The trimethylsilylmethyl derivative of chromium(II) can even be added to Cr(VI) oxide catalysts, to produce hybrid or "two valent" catalysts that are considerably more active than either component alone [660,682]. Presumably, in addition to reacting with silanol groups, the organochromium component also reacts with and reduces the Cr(VI) oxide to an unknown, but highly active, state. In this way, inactive Cr(VI) oxide sites can react with the organochromium compound to produce a new active site. 

A catalyst based on chromocene, Bis(cyclopentadienyl)chromium, was developed in the 1960s at Union Carbide by G. L. Karapinka and cowork-ers and was the first commercial chromium-based catalyst that was prepared with an organochromium compound containing Cr-carbon bonds in the starting material. In addition, the starting material based on Cr(II), did not need to be oxidized to a Cr(VI) species to obtain a high activity ethylene polymerization catalyst. 

---