Evolving New Zeolites with Controllable Pores on a Molecular Level

---

By Steven M. Kuznicki, PhD

Molecular Sieves Touch Your Life
Forming the heart of thousands of processes as catalysts, adsorbents, ion-exchange and purification agents, crystalline zeolite molecular sieves are among the most important and valuable classes of inorganic materials available to the chemical industry. Utilized in petroleum cracking for gasoline production, oxygen separation from air, water purification by softening and removing heavy metals and much more, these materials directly impact our lives.
Crystalline molecular sieves derive their name from the extremely uniform pores inherent to their structures. As is evident in Figure 1, these holes are so uniform that they may be used to separate molecules by size. The TEM images shown in this figure represent a molecular sieve with pores of about eight angstroms (0.8 nanometres) in diameter — about the size of a one, three, five-trimethylbenzene molecule. Molecules smaller than the pores are able to penetrate them and be actively adsorbed by the crystals, while those larger than the pores are excluded and rejected.
While some templated structures exist with pores as large as 100 angstroms or more, traditional aluminosilicate molecular sieves may be prepared with pore sizes ranging from about 2.5 to 10 angstroms, a very convenient range for molecular separations. Unfortunately, since they are constructed from fixed cages with openings formed by rings of six, eight, 10, or 12 interconnected metal oxide units, only certain, incremental pore sizes have been available, even with modifications, such as placing different size cations in the pore openings.
While there would appear to be limitless opportunities for molecular separations based on the molecular sieve effect, very few processes utilize the elegant property of size discrimination on a molecular level due to the limited number of pore sizes available. Most zeolite separation processes rely on differential adsorbate — adsorbent interaction energies. Expanding the scope of available pore increments available for molecular sieves could impact many areas by expanding the available “sieves” that could be employed.

A New Concept in Molecular Separations
While traditional molecular sieves are constructed from fixed cages, a new class of molecular sieves has been discovered that are constructed from parallel chains of certain metal oxides, most notably titanium (IV), strung together with rings of other units, most notably silicon (IV). The metal atoms in the chains are six co-ordinated while the atoms in the rings are four co-ordinated. These “mixed co-ordination” molecular sieves possess uniform pores in the same size range as traditional zeolites, but also manifest a unique property — they shrink when calcined. As the overall crystal shrinks, the pores shrink in unison and essentially uniformly. This pore shrinkage can be controlled to a precision of 0.1 angstrom (0.01 nanometres) or better. As depicted in Figure 2, this controlled pore shrinkage can produce adsorbents that can be employed to resolve and completely separate molecules with size differences as small as 0.1 angstrom. The size range can be “dialed in” by controlled shrinkage. This phenomenon forms the basis of the so-called Molecular Gate® effect.

First Commercial Application
ETS-4 is a small-pored, mixed co-ordination titanium-silicate molecular sieve. With an initial pore size of about four angstroms, it can be shrunk down to about 2.5 by controlled calcinations — a convenient size range to separate many commercially important gas mixtures. The pores can be “frozen” at sizes between 2.5 and four angstroms by controlling the calcination process.
Nitrogen is a common, and often difficult to remove, contaminant of natural gas. It is also marginally smaller than methane, typically listed as being between 3.6 and 3.7 angstroms versus 3.8 for CH4. This incredibly small difference, about 1/20 the diameter of a hydrogen molecule, is sufficient to facilitate a separation by molecular size using the Molecular Gate effect. The adsorbent pores can be “tuned” by controlled shrinkage to readily adsorb nitrogen while “rejecting” methane. The size discrimination can be made so complete that the separation can be conducted at wellhead pressures. It is claimed that an additional small molecule (CO2 at a diameter of about 3.3 angstroms), which also commonly contaminates natural gas, can be removed simultaneously with nitrogen using the same effect. This effect forms the basis of a growing commercial natural gas purification process in the U.S.

Separation of the Constituents of Air
An interesting laboratory demonstration of the ability of the controlled pore size separation of molecules of similar size can be found in the separation of the constituents of air. Nitrogen, oxygen and argon are all generally listed as within 0.1-0.2 angstroms of 3.5. As presented in Figure 3A, if air is injected into a GC column with an adsorbent featuring a large enough pore to accommodate all three species, nitrogen interacts with the crystal’s interior more strongly than oxygen or argon, leading to greater retention. This sort of differential interaction of air constituents with more traditional zeolites forms the basis of the huge adsorptive air separation industry. In
Figure 3B, a slightly smaller pore results in argon being “rejected” by being larger than the available pore size and being eluted quickly. Note that nitrogen retention is also declining as that molecule — now only slightly bigger then the available pores — is experiencing reduced adsorption. Finally, in Figure 3C, where the pore is reduced slightly to an even smaller size, nitrogen is “rejected” along with argon and the smallest of the three species, oxygen is preferentially retained.
While intended as a demonstration of the separation of molecules of nearly identical size, it can be argued that the separation of air by adsorption of the 20% oxygen may offer advantages over the current removal of the 79% nitrogen, if the proper cycle were constructed. Alternatively, it could provide a path toward more efficient and economical N2 separation from air.

The Future – New Adsorbents, New Opportunities
The potential applications for controllable pore-sized adsorbents in the areas of separations and purifications seem limitless. This would be especially true if the range of pore sizes could be expanded and pore-size control made even more precise.
Very recently, a means of generating such controlled-pored materials without calcination was discovered at the University of Alberta (Calgary, AB). This new discovery allows the creation of larger and smaller pores. It also appears to produce even more perfect pores than previously available, that is to say adsorbents with even narrower pore size distributions than the already impressive Molecular Gate adsorbents. This new method of generating controlled pore size adsorbents is the subject of several patent applications and application specific development efforts with several Canadian companies.
Clearly, the potential for such an enabling concept, which is already seeing real-world applications in natural gas purification, extends well beyond the areas mentioned or envisioned. It truly does add a new dimension to the concept of molecular sieving.

Acknowledgement
Elements of this work are being conducted using funding from the Alberta Ingenuity Fund (Edmonton, AB), the Natural Sciences and Engineering Research Council of Canada (Ottawa, ON) Discovery Grant Program, the University of Alberta, and industrial sponsors such as QuestAir Technologies Inc. (Burnaby, BC).
A special thank you is extended to Christopher Lin and Andree Koenig of the University of Alberta and James Sawada, PhD of QuestAir Technologies.

Steven M. Kuznicki, PhD is a professor in the department of chemical and materials engineering at the University of Alberta, an Alberta Ingenuity Fund scholar and the Canada Research Chair in Molecular Sieve Nanomaterials. He can be reached at
Kuznicki@ualberta.ca.