|[Previous Story] [Next Story]
PAYING ATTENTION TO ACTIVATORS
Metallocene catalyst researchers are studying cagey methylaluminoxane activators
ALEXANDER H. TULLO, C&EN NORTHEAST NEWS BUREAU
Metallocene catalysts are known for their precision. They have given polymer producers unprecedented control over the architecture of their products and the way they make them. But widespread industry adoption of metallocenes may rely on a substance few chemists know very much about: methylaluminoxane (MAO) activators.
Researchers now realize there may be better metallocene activators. They are turning their attention increasingly to MAO, with the hope that understanding how it works will enable them to come up with more efficient catalyst activators.
||COST CUTTER Akzo Nobel is developing a modified methylaluminoxane activator at its Deer Park, Texas, plant.
The chemical industry may not understand MAO, but it has been using it for a while. Before aluminoxanes were used to activate metallocenes, they were employed in butadiene rubber and propylene oxide polymerization, according to Brian L. Goodall, a researcher with MAO producer Albemarle.
In the 1980s, Goodall says, it was discovered that MAO increased the activity of metallocene catalysts by six orders of magnitude over the aluminum alkyl activators that were used previously. It was practical for the new industry to adopt MAO as the activator of choice.
Goodall notes a dichotomy. "You go through all sorts of lengths to make a beautiful, highly complex molecule that you call a single-site catalyst, and then you throw it into an activator that you don't understand," he says.
MAO HAS BEEN difficult to study. "People have tried every technique they can, but we still don't know much about it. We are dealing with a working hypothesis," Goodall says.
Andrew R. Barron, the Welch Chair of Chemistry and professor of material science at Rice University, has done much of the seminal work in elucidating structures for MAO. His work has led most chemists to think of MAO as a number of cage structures instead of linear polymer molecules as they had in the past.
But, Barron notes, it is MAO's tendency to form many cage structures that makes it difficult to study. "If you picked up a pair of nanotweezers and plucked out a molecule of MAO, you would get a molecule," he says. "If you did it again, you would get a different molecule."
Even though models of MAO are starting to emerge, no one has been able to crystallize it, which makes confirming its structure difficult. "Even if someone gets a structure, that's just one of the potential structures," Barron says. "We already have over 10 structures of tert-butylaluminoxane," he adds.
MAO's elusiveness wouldn't matter much if it were the perfect activator, but it isn't. It is made through the hydrolysis of trimethylaluminum, an expensive raw material. Moreover, great excesses of MAO have to be used relative to the amount of metallocene catalyst, increasing the cost of using it.
||EVOLUTION An early candidate to replace MAO, isobutylaluminoxane decamer is shown here with the isobutyl groups in gray and white and the core aluminum and oxygen in red and yellow, respectively.
COURTESY OF ALBEMARLE
In addition, MAO is unstable--it tends to precipitate in solution over time, according to Roy Simmons, Crompton Corp.'s manager of business development for aluminum alkyls. "It has a short shelf life because of its tendency to form gels," he says.
Metallocene catalysts, Albemarle's Goodall explains, cost several times as much as conventional Ziegler-Natta catalysts, and the MAO activator could comprise one-third to one-half of the total cost. He says the developers of metallocene technology are getting over previous hurdles such as processing difficulties and intellectual property disputes. "The final hurdle is that you have to be able to produce that new polymer for essentially the same cost as a conventional polymer," he adds.
In polymer markets with millions of tons of volumes, these small differences add up, says James C. Stevens, senior scientist at Dow Chemical's polyolefin business. "With some products, the choice of activator can make the difference between having a commercially viable process and having a catalyst package cost that is too high for the product or application to support," he adds.
Tobin J. Marks, a chemistry professor at Northwestern University, says industry has a lot to gain by better understanding MAO. "We don't understand the nature of what you get when you mix the catalyst and cocatalyst," he says. "Obviously, if you could understand what you get, you could make much better, more selective, and more active catalysts and we could replace MAO with something better."
From a chemistry perspective, MAO seems ideal. "The activator does many jobs, and there's really one out there that does everything at the moment, and that's MAO," says Albemarle's Goodall.
Goodall has identified four different tasks MAO performs in activating single-site catalysts. First, it acts as a scavenger for oxygen, moisture, and other impurities in the reactor. However, many other molecules can do the same job, he notes.
A metallocene catalyst typically consists of a complex of two cyclopentadienyl anion-based ligands with a metal cation. On its own, however, this metallocene is no good for polymerization--which is where MAO's next function comes in. MAO places methyl groups on the metal, giving the olefin raw materials a place to bond in the polymerization process.
Third, because the methylated metallocene is not electrophilic enough to attract olefins, Goodall says, MAO takes away another chlorine or methyl group bonded to the metal to give it a positive charge.
The final function of MAO, Goodall explains, lies in its large, three-dimensional structure, which diffuses the anionic charge that previously was held tightly by chlorine.
LAST FALL, Goodall's team at Albemarle won a $2 million grant from the Advanced Technology Program (ATP) of the National Institute of Standards & Technology to develop new activators for metallocene catalysts based on tackling all the different functions of MAO.
Goodall's strategy has been to find materials to handle all the jobs of MAO, but at a better cost.
Isobutylaluminoxane (IBAO) was an early candidate because of its lower cost and well-known structure, but Albemarle found that it wasn't a strong enough Lewis acid to pull an anion from a metallocene. So Albemarle turned to hydroxy IBAO, which has a Brønsted acidity that does the same job as MAO's Lewis acidity. In addition, hydroxy IBAO forms a cluster, which diffuses the anionic charge like MAO does.
Albemarle uses triisobutyl aluminum as a scavenger in the hydroxy IBAO system. The firm also methylates the metal before the catalyst is put into the reactor. But, Goodall notes, this can be a costly step that Albemarle is trying to improve upon.
Hydroxy IBAO, Goodall says, should be cheaper to produce than MAO. In addition, it doesn't have to be used in the large excesses that MAO does. But hydroxy IBAO also has its drawbacks, Goodall admits. "In the course of a few hours, it continues reacting with itself until there's no hydroxyl groups left and you're just left with IBAO, which isn't a good activator." Albemarle has been working on this problem, too, he says.
MAO producer Akzo Nobel is taking a different approach to the problem with what it calls modified MAO. The product offers better storage stability than conventional MAO, according to Hans Van Haarst, business manager for metal alkyls at Akzo Nobel.
In modified MAO, isobutyl and n-octyl groups are substituted for the methyl groups. "This prevents precipitation and unwanted reactions that occur in conventional MAO," he says.
Crompton says it will improve costs through increasing capacity. Crompton is considering a new plant in Germany and is planning its first unit in the U.S. as well. "Price is going to go down as we get more economies of scale," Simmons says.
Dow's effort to reduce activator costs has yielded "double activation." In double activation, both methyl groups of a group IV metallocene interact with two or more equivalents of a Lewis acid, according to Jack Kruper, a senior sientist at Dow. A strong Lewis acid, tris(perfluorophenylaluminum), is used to form the doubly activated species.
The chemistry is unusual, Kruper admits, but well studied and economically worthwhile. "This well-characterized species is tremendously more efficient in promoting ethylene/octene polymerization than a monoactivated species," he says. Dow has found the catalysts to be 30 times more active than monoactivated catalysts.
"Stronger and more efficient Lewis acid activators, such as those afforded by double activation, make for better atom economy and polymer properties as well," Dow's Stevens says.
Rice University's Barron notes that much of the activator work has predicated itself on finding strong Lewis acids, even though MAO itself isn't exactly a Lewis acid. Instead, he says, MAO has latent Lewis acidity--it becomes a Lewis acid when its cage structure opens in the presence of a metallocene.
Barron cautions that going for merely a strong Lewis acid creates activators that are sensitive to impurities. On the other hand, he adds, the strong Lewis acids are much more active than MAO.
Marks's group at Northwestern--which at various times has collaborated with Dow Chemical, Dow Corning, BP, and Albemarle, and has received funding from the Department of Energy--has developed different families of strong Lewis acid catalyst activators.
THE RESEARCH, Marks says, has led to more than just better activators. "The catalytically active cation-anion pair product you get when the activator activates is another knob that you can tune to change the properties of the catalysts," he adds. The activator, he explains, can affect the rate of polymerization, the polymer molecular weight, and the thermal stability of the catalyst system, as well as the stereochemistry and other properties of the final polymer.
The most commercially developed family is based on fluoroaryl borane, for which the Northwestern group has already filed patents and granted licenses.
In exploring activators, researchers have uncovered another layer of complexity in single-site catalysis. They are finding that they cannot look at any one ingredient in the reactor without looking at all the others, too. "For a while, people thought that if you understood everything about the metal complex, you understood everything about the whole system," Marks says. "It's turning out that the activators are a very important part also."
[Previous Story] [Next Story]
Chemical & Engineering News
Copyright © 2001 American Chemical Society