Zeolite crystals form three-dimensional honeycomb-like structures containing ordered arrays of tiny pores. Many zeolites occur naturally as minerals; others are synthesized commercially for specific uses, such as catalysis. "A nice example of their catalytic activities is in the 'cracking' of heavy oil into gasoline," says Ryong Ryoo, a chemist at the KAIST Institute for the NanoCentury in Daejeon, South Korea.
That particular reaction is carried out by a zeolite called MFI (or ZSM-5), one of the most important catalysts in the petrochemicals industry. Ryoo and his colleagues have developed a new synthesis that greatly increases MFI's performance (see page 246) — and the same approach could be applied to the synthesis of other zeolites to improve their functions.
The 'honeycomb' framework of most zeolites contains silicon, aluminium and oxygen. Cations, water and other molecules sit within the pores, where catalytic conversion of substrates of appropriate size and shape occurs. The assembly of zeolites is typically guided by organic cations that function as structure-directing agents. "Usually people use quaternary ammonium ions," says Ryoo. "These interact with silicate minerals in aqueous solution to help them undergo polymerizations."
One problem with traditional zeolite synthesis is that the size of the pores that form — less than 1 nanometre in diameter — makes it difficult for substrates to diffuse along the crystal structure of the zeolite and gain access to all of the sites of catalysis. The challenge, Ryoo says, has been to "increase diffusion without changing the micropore diameter of the zeolite".
One way to accomplish this is to reduce the thickness of the zeolite crystal, thereby decreasing the lengths of the diffusion paths. "Many people had already tried this approach, but, as far as I know, no one had succeeded in reaching a single-unit-cell thickness," says Ryoo. He thought that he could succeed by changing the make-up of the structure-directing agent.
In attempts to direct the synthesis of MFI, researchers had typically used a surfactant containing a quaternary ammonium group (tetrapropylammonium) at one end. "We needed a more powerful structure-directing agent," says Ryoo, who predicted, on the basis of earlier work, that such an agent would contain two quaternary ammonium groups.
Ryoo first instructed Minkee Choi, one of the graduate students in his lab, to put a di-quaternary ammonium group between two long hydrocarbon chains. But this compound didn't work in zeolite synthesis, so Ryoo suggested shortening one of the chains. "I guessed that the probability of success would be no more than 10% — not a small probability — and that is what I told my students," he says.
He asked another graduate student, Kyungsu Na, to help with the synthesis. "Surprisingly, Kyungsu's first try was quite successful," says Ryoo. The resultant zeolite consisted of 2-nanometre-thick sheets, he recalls.
Having successfully synthesized a new MFI zeolite structure, Ryoo's group spent about two years looking for interesting properties. One of their more exciting discoveries was that when the zeolite is used to transform methanol into gasoline, the catalytic longevity is greatly increased. "I expected a long catalytic lifetime, but I was surprised by the result. It was much longer than I expected," he says.