Catalytic nanoliths

Abstract

The nanoporous anodic aluminum oxide (AAO) structure is shown to be a useful platform for heterogeneous catalysis. By appropriately masking the perimeter during anodization and etching, the AAO can be formed at the center of an aluminum disc. The remaining aluminum ring connects seamlessly to the AAO and provides mechanical support for convenient handling. The supported AAO can be sealed in a standard fitting so that the nanopores in the structure function as an array of tubular reactors, i.e. a nanolith. Coating the walls with catalytically active materials turns the nanolith into a novel catalytic system. For the oxidative dehydrogenation (ODH) of cyclohexane, the nanolith catalytic system is superior to a conventional powdered catalyst in terms of both efficiency and in reducing over oxidation. A simple analysis of the flow through the nanolith combined with experimental data indicates that mass transfer through the nanopores follows a mixed flow model.

Introduction

Porous anodized aluminum oxide (AAO) structures feature a highly ordered array of straight nanopores (Patermarakis and Pavlidou, 1994; Masuda and Fukuda, 1995). For the last half a century this material has been widely used as micro-filters, anti-corrosion barrier layers, and as substrates for various coatings (Kuo and Xu, 2006; Kyotani et al., 2002; Tomioka and Yoshida, 2001). In recent years, AAO structures have found application in the nanofabrication field because the ordered nanopores form an ideal template for synthesizing a variety of nanostructures (Martin, 1994; Masuda et al., 1995, 1997; Gao et al., 2002). However, the application of AAO structures as a framework for catalysts remains an intriguing but little investigated area. The functionality of a heterogeneous catalyst depends critically on its structure over a range of length scales (Goodwin et al., 2004; Weitkamp et al., 1999). At the length scale of 10s–1000s of nanometers, the dimensions and topology of the catalyst pore structure can influence reagent flow, the sequencing of catalytic active sites, and the contact time between reagents and catalyst (Roy et al., 2003). Conceptually, the AAO structure consists of an assembly of identical pores having nanometer dimensions that span a flow reactor so as to produce an array of nanoreactors that we call the catalytic nanolith, i.e. a nanometer scale version of the catalytic monolith. With this assembly, each reagent molecule must traverse an identical pore and, ideally, each diffusion path can be engineered through control over the pore diameter, wall composition, and length. This results in more uniform and tunable contact times than are possible with a conventional fixed bed powder catalyst.

Besides fine control over diffusion paths and contact times, this catalytic system also provides a number of additional possibilities. For example, its open, one-dimensional channel structure facilitates the deposition of coatings and catalytically active materials. One good example is the application of atomic layer deposition (ALD) in fabricating a catalytically active nanolith. Traditionally this powerful deposition technique has been used in the coating of flat surfaces; its application in the catalysis field can be limited by diffusion into powder supports with very small irregular pores (McCormick et al., 2007). In our studies, ALD has been shown to be a powerful tool for adjusting the pore size, controlling chemical composition, and synthesizing interesting catalytic structures inside the AAO nanopores (Stair et al., 2006). Another feature of this nanolith catalytic system is the possibility to study the influence of reagent delivery on catalytic performance. For example, with an asymmetric reactor it is possible to deliver two reagents on opposite sides of the nanolith structure or to sweep reactants or products away from the entrance and exit of the structure (as pictured in Fig. 1(a)) in order to control the flow of molecules that enter and exit the nanolith. Catalytic results from such experiments may be a valuable test of theoretical predictions regarding the location of catalytic transformations in the pores (Albo et al., 2006). Furthermore, the influence of the flow pattern in the nanopores on the catalytic performance can also be studied. Using ALD, the sizes and shapes of the nanopores can be finely tuned. The pore diameter can be adjusted in the range of 2–40 nm (using a 40 nm pore AAO as the starting point) with Ångstrom scale precision. By controlling the preparation conditions, it is even possible to synthesize asymmetric nanoliths with pores of different diameters and compositions open on the two sides (Li et al., 1999). Since the dimensions of the nanopores significantly influence the flow pattern of the reagents inside the pores, a measure of control over the catalytic properties of the nanolith system can be achieved (Albo et al., 2006). We can also investigate how the spatial location of the catalytically active materials in the pores influences the catalytic performance. For example, it is possible, using ALD diffusion techniques, to locate the catalytic material at the entrance or exit to the pores or even to arrange the catalytic material as a series of bands down the length of the pore. Similarly, the formation of a controlled sequence of different catalytic materials is a topic of interest. Schematic pictures of some catalytic structures of interest are depicted in Fig. 1(b).

The arrays of nanoreactors on the AAO structure make a promising, highly selective catalytic system for some sequential reactions. When the reactants flow through the nanolith, the limited fraction of the pore length and precise position over which the catalytic film is deposited may dramatically reduce the contact time between reactants and catalyst so that undesirable over reactions can be limited to a great extent. The oxidative dehydrogenation (ODH) of alkanes to olefins is a sensitive probe for the performance of the nanolith because the partial oxidation products are easily oxidized to oxygenated products, which are thermodynamically more stable (Corberan, 2005; Mamedov and Corberan, 1995; Kung, 1994). The ODH of cyclohexane catalyzed by supported vanadium oxides is an example, in which cyclohexane is first converted to cyclohexene and then further oxidized to benzene and oxygenated products such as CO and CO₂. In this reaction it is difficult to achieve a high yield of cyclohexene as the partial oxidation product (Panizza et al., 2003). In our earlier work we have reported that VO_x supported on the AAO structure exhibits much better performance than a conventional vanadia powder in terms of the selectivity to cyclohexene (Stair et al., 2006). However, those experiments were carried out in a reactor system where the AAO structures were not sealed to the reactor walls. In that configuration, bypass of the reactant around the catalyst is inevitable. This motivated the desire to study the sealed, gas-tight AAO structure in a flow reactor, in which the reactant is forced to flow through the nanopores, forming a nanolith. However, due to the fragile nature of this ∼100 μm thick material, little sealing force can be directly applied without breaking the AAO. Our solution to this problem is to mount the AAO structure onto a supporting material that can accept the force of the seal in a conventional gas-tight fitting.

In this paper we report a novel method of synthesizing an AAO structure embedded in an aluminum ring support. The concept of the nanolith is realized by sealing the AAO in a flow reactor adapted from a standard Swagelok VCR fitting so that the reactants must flow through the nanopores. The performance of the nanolith catalytic system is compared to that of a powder bed for the ODH of cyclohexane catalyzed by supported vanadium oxides. The performance of the nanolith is analyzed using a physical model that includes factors such as the mass flow, diffusion, and the chemical reactions inside the nanopores.