If you have an ACS member number, please enter it here so we can link this account to your membership. (optional)

ACS values your privacy. By submitting your information, you are gaining access to C&EN and subscribing to our weekly newsletter. We use the information you provide to make your reading experience better, and we will never sell your data to third party members.



Sowing The Seeds Of Oriented Films

Technique allows systematic manipulation of pore orientation in molecular sieve thin films

by Michael Freemantle
February 27, 2006 | A version of this story appeared in Volume 84, Issue 9

Look At This!
Credit: Courtesy Of Michael Tsapatsis
Tsapatsis (from left), postdoc Shubhajit Ghosh, and Ph.D. student Jungkyu Choi examine the porous structure of MFI zeolite.
Credit: Courtesy Of Michael Tsapatsis
Tsapatsis (from left), postdoc Shubhajit Ghosh, and Ph.D. student Jungkyu Choi examine the porous structure of MFI zeolite.

Micrometer-thin molecular sieve films are exciting considerable research interest as potential materials for chemosensors, catalysts, microelectronic devices, and separation membranes. Before such applications can be developed, however, a number of scientific and technical hurdles need to be overcome.

"Zeolite films in particular hold promise for practical applications, but their fabrication imposes challenges, including control of thickness, grain size, and pore orientation," says Michael Tsapatsis, professor of chemical engineering and materials science at the University of Minnesota, Twin Cities.

Zeolites are naturally occurring or synthetic aluminosilicate materials with porous framework structures that can host water and a variety of molecules and cations. They are used as hosts for catalysts in industrial processes and also for water purification and softening, laundry detergents, soil treatment, and the preparation of medical-grade oxygen.

Over the past decade, Tsapatsis and coworkers have been investigating the development and use of molecular sieve materials, including synthetic zeolites with 10-membered aluminosilicate rings, variously known as MFI, ZSM-5, or silicalite-1. Pellets of microcrystalline MFI powder are widely used for catalytic processes in the chemical and petrochemical industries. MFI films are also being developed as membranes for gas separation (C&EN, Oct. 3, 2005, page 49).

"We are particularly interested in preparing MFI films that can be used as membranes because the size of MFI pores is close to that of many industrially important molecules," Tsapatsis says.

Zeolite membranes are invariably prepared as composite materials, mainly with porous supports of stainless steel or alumina. A key problem is that the composite is not reproducible, observes Joaquin Coronas, an expert on MFI gas-separation membranes who is associate professor of chemical engineering at the University of Zaragoza, Spain. "For instance, one can find MFI-type zeolite membranes with not only different values of permeability and selectivity but also with different qualitative behaviors. Some silicalite membranes can separate butane isomers at high temperatures, while others only do so at low temperatures. A few very high quality membranes can separate xylene isomers, but some of these cannot separate butane isomers."

The discrepancies in performance arise principally from variations in the porosity and chemical composition of the support, Coronas says. "Quite often, the zeolite is synthesized inside the support pores, and so it is not possible to know the thickness of the membrane," he explains. "In some cases, the zeolite precursor gel attacks the support, and undesirable reactants are incorporated into the zeolite phase."

The Tsapatsis group has been working on methods for systematically controlling the pore orientation of continuous MFI zeolite films. "The performance of a zeolite membrane depends on its microstructure," Tsapatsis says. "The pore structure of MFI is anisotropic. We have shown that the membrane performance depends strongly on the preferred crystallographic orientation, so it is important to control the way the crystals, and therefore the pores, are oriented in the membrane."

Seeded Growth
The tetrapropylammonium (TPA) cation is used to prepare zeolite seeds with straight pores perpendicular to the surface of an alumina porous support. TPA trimer is used to grow the b-oriented film (bottom left) from the seeds. The a-oriented film (bottom right), with zigzag pores perpendicular to the surface, employs TPA trimer for seeding and TPA monomer for growth.
Credit: Credit: Courtesy Of Michael Tsapatsis

The directions of channels or pores in MFI zeolite crystals relate to three mutually perpendicular crystallographic axes: a, b, and c. Pores in the a-direction are arranged in a zigzag pattern and interconnected with straight pores in the b-direction. Pores in the c-direction follow a "tortuous path," Tsapatsis observes.

Films that consist of crystals with their b-crystal axis perpendicular to the support surface are said to be b-oriented. "Since MFI-type zeolite has an anisotropic structure with straight pores along the b-axis, b-oriented membranes should provide the fastest permeation pathways," Coronas explains. "They should be faster than a-oriented ones which, in turn, should be faster than c-oriented membranes."

MFI films that are b-oriented perform best in the separation of xylene isomers and in other aromatic separations, according to Tsapatsis. "In the b-oriented MFI structure, straight pores with a diameter of about 0.55 nm run along the b-crystallographic axis," he says. "That means that the molecules to be separated by the b-oriented membrane travel through the straight pores. Films with the c-orientation perform better for separating butane isomers and other linear/branched hydrocarbon separations."

The orientation of MFI pores depends on how the films are prepared. In the so-called in situ method of synthesis, zeolite crystals nucleate on the support surface that is directly in contact with an alkaline solution containing the zeolite precursors. However, the success of in situ methods in yielding uniformly oriented MFI films is limited, according to the Minnesota team.

An alternative synthesis method, known as the secondary (or seeded) growth technique, decouples zeolite nucleation from zeolite growth by depositing a seed layer of zeolite crystals on the support surface. The technique enables one of the crystallographic directions to be oriented perpendicular to the support.

"We were the first academic group that, simultaneously with the ExxonMobil zeolite membrane group, reported in 1996 on the synthesis of zeolite membranes using secondary growth," Tsapatsis says. One type of film they synthesized was c-oriented MFI films. "We used randomly oriented seed layers that were deposited by dip-coating the porous alumina substrate in a colloidal suspension of 100-nm MFI crystals."

The technique relies on the use of structure-directing agents in the growth solution. The agents become incorporated at the pore intersections of the crystals in the films during growth. The structure-directing properties of the agents are thought to arise from their geometric shape, size, and charge distribution. The Tsapatsis group showed that secondary growth from a seed layer in a growth solution made from tetraethyl orthosilicate (TEOS), water, and tetrapropylammonium (TPA) hydroxide as a structure-directing agent leads to the development of films as thin as 1 ??m that can be used as membranes.

"We call these the first generation of oriented films," Tsapatsis says. "Their orientation is not uniform throughout the film thickness, but it gradually develops as the film becomes thicker."

In 2003, Tsapatsis and coworkers reported that seeds that are b-oriented rather than randomly oriented can be used to prepare b-oriented zeolite membranes on porous alumina supports (Science 2003, 300, 456). The seeds are formed by using TPA as a structure-directing agent and a trimer of TPA as the agent for secondary growth from the seed layer.

The group also demonstrated that the alumina-supported b-oriented membranes can be used to separate xylene isomers with high flux-that is, low resistance to gas or vapor flow-and high selectivity.

Last year, the Minnesota group reported that the same technique can be extended to grow b-oriented MFI films on rough, porous stainless steel supports precoated with a layer of mesoporous silica (Ind. Eng. Chem. Res. 2005, 44, 9086). That work was carried out in collaboration with chemistry professor Wilhelm Schwieger's group at the University of Erlangen-Nuremberg, in Germany.

"The ability to synthesize thin, oriented zeolite films is a key step in moving this technology forward," comments chemical engineer Richard D. Noble of the University of Colorado. "There are several steps described in this paper that need to be considered, such as the initial state of the support layer and the orientation of the seed layer as well as its attachment to the surface. This paper also raises an important point of the mass-transfer resistance of the support."

According to Tsapatsis, to make a practical membrane, it is necessary to grow the selective molecular sieve layer on a high-flux porous support. "Commercially available porous stainless steel membranes are such supports," he says. "Stainless steel as a support material for zeolite membranes possesses the important advantage of ductility, and it is compatible with the most commonly used plant equipment parts.

"Sealing at high temperatures is less problematic than with ceramics, and its better heat conductivity can be advantageous for membrane reactor applications involving heat transfer to or from the reaction region," he adds. "However, to date, highly p-xylene-selective, stainless steel supported MFI membranes have not been reported. The b-oriented MFI films we prepared are the best available for xylene isomer separations, and we currently deposit them on stainless steel tubular supports that are appropriate for industrial use."

In a recent paper, the Minnesota group and Zhiping Lai, an assistant professor of chemical engineering at Nanyang Technological University, in Singapore, showed that uniformly a-oriented MFI zeolite films can be prepared by seeded growth (Angew. Chem. Int. Ed. 2006, 45, 1154). To grow the films, the group uses an alkaline TEOS solution and a trimer of TPA as the structure-directing agent for seed formation. The support is a layer of mesoporous silica on the polished side of a porous alumina disk. Films are grown from the seed layer with the TPA monomer as the structure-directing agent.

"Until now, no reports existed on the synthesis of uniformly a-oriented MFI films," Tsapatsis says. "We cannot predict how these films are going to perform as membranes. However, now they are available for the first time, and based on our previous findings, we expect that their behavior will be different from existing membranes."

The Minnesota researchers plan to systematically test and optimize the performance of the a-oriented films and then determine potential applications for the membranes. They intend, for example, to investigate their use for the separation of xylene isomers, saturated and unsaturated hydrocarbons, and linear and branched hydrocarbons.

"The Tsapatsis group may claim to be the first to establish absolute control of the crystallographic orientation of MFI-type zeolite membranes," Coronas says. "Their seeded growth strategy allows one to prepare thin zeolite membranes in a reproducible way that can potentially afford the expansion of zeolite membranes to many interesting separation and reaction applications."

Tsapatsis points out that several recent developments by other groups have contributed to this level of control. "Our work benefited greatly from recent progress in thin mesoporous films and zeolite nanoparticle synthesis and extensively uses recently introduced particle monolayer deposition techniques," he says.

The a- and b-oriented films, synthesized from oriented seeds, belong to what the Minnesota group calls the second generation of oriented MFI films. The researchers are currently comparing the permeation properties of the a- and b-oriented films. The aim, they note, is to provide "the first set of data for gas and vapor permeation through zeolite membranes of a given structure type with drastically different preferred orientations."


This article has been sent to the following recipient:

Chemistry matters. Join us to get the news you need.