Research

Novel Catalyst Development

Supported Liquid-Phase Catalysts (SLPC)

In SLPC, the liquid-phase catalyst is coated onto the walls of a porous support as an ultra-thin film, much as in a chromatographic packing with supported liquid and it, thus, effectively combines the attractive features of homogeneous catalysis such as high specificity, mild conditions, and molecular dispersion of catalytic species, with those of heterogeneous catalysis such as large interfacial area, lack of corrosion, use of packed-bedreactors, and facile separation of catalyst from products. The technique is suitable for low temperature gas-phase reactions [1-3] as well as for liquid-phase reactions in which the catalyst and reactant phases are immiscible [4]. We have also developed comprehensive a model of transport and reaction in SLPC that is free of adjustable parameters and allows design and optimization of these systems [1-3].

Supported Molten-Salt Catalysts (SMSC)

The volatility of conventional solvents limits the temperature at which a SLP catalyst may be used to about 200 ºC. While this is not limiting for immiscible two-phase liquid reactions, it is a significant constraint for many gas-phase reactions. This problem may be addressed by using appropriate molten-salts as solvents. Molten salts typically have a broad liquidus range and very low volatility. Thus, we developed a Wacker SMMC with excellent stability and none of the corrosion problems of the industrial process [5]. Further, molten salts can disperse not only homogeneous but also many conventional heterogeneous catalytic species [6]. For instance, we developed an SMSC containing Pt crystallites that is less susceptible to coking than the conventional Pt catalysts. We have also applied the SMMC technique to fuel cells [7].

Supported Molten-Metal Catalysts (SMMC)

The traditional heterogeneous catalysts frequently involve tiny metal crystallites (~5 - 10 nm diameter), e.g., Pt, dispersed on a porous support to provide an extremely large interfacial area in a small volume. This, of course, has been a key success of catalyst technology in the twentieth century, allowing the wide-spread use of precious metals in industry and in pollution abatement. On the other hand, while many non-noble metals from groups 1, 12, 13, 14, 15, and 16, their alloys, or intermetallic compounds possess catalytic properties, their industrial usage has not been possible since many of these are molten at temperatures below 500 ºC. This is due to the limited interfacial areas and severe corrosion problems encountered in conventional bubbled reactors containing molten metals. In the technique of SMMC [8, 9], microdroplets of molten metal catalysts are supported on porous supports, thus providing extremely large interfacial areas and obviating any corrosion problems. Results show that interfacial areas and dispersion rivaling those of conventional heterogeneous catalysts are possible. SMMC has also allowed us to study reactions previously not observed on molten metals due to limited surface areas, e.g., the highly efficient selective catalytic reduction of NO on In, which has a melting point of 150 ºC [10]. Further, these catalysts are fundamentally different from conventional metal catalysts, since the microdroplets are not crystalline with different crystal faces, but are rather atomically uniform. These catalysts also appear to be less susceptible to coking than conventional catalysts.

Fuels and Chemicals from Renewable Resources

We are investigating catalysis of the production of fuels and chemicals from renewable resources, in particular, via biomass-based ethanol. Thus, we are systematically studying the thermodynamics and kinetics of ion-exchange resin catalyzed synthesis of the family of tertiary ethyl ethers from ethanol and C4 to C6 tertiary olefins derived from FCC gasoline [11-16]. These ethers are of commercial significance for use as oxygenate gasoline additives designed to reduce CO emissions and to concomitantly enhance octane. In addition, we are developing a process to produce isobutylene from watery ethanol through ethylene [17], which is also of potential commercial significance. In addition, we are doing a theoretical and experimental study of the non-isothermal liquid-phase packed-bed reactors used industrially for ether synthesis [18]. Since the production of ethanol is energy intensive, we are also studying the novel and energy efficient technique of microporous distillation for the production of anhydrous ethanol [19-20].

Catalytic Reactor-Separators

The combination of reaction and separation in a single device offers advantages such as enhanced rates and conversion for equilibrium restricted reactions, altered selectivities, higher energy efficiency, and reduced capital costs. The following catalytic reactor-separators are being investigated:

Catalytic Membrane Reactor Separators (CMRS)

We developed a novel supported liquid-phase catalytic membrane reactor-separator (SLPCMRS) for homogeneous catalysis [23], that not only immobilizes the liquid-phase catalyst on a porous support sandwiched between two membranes, but also simultaneously effects separation of reactants and products. We have also performed porous-walled reactor separator (PWRS) experiments for the Pt catalyzed methylcyclohexane dehydrogenation [24] along with the development of a sophisticated model of transport and reaction based on the dusty-gas model to account for the combined diffusion and flow.

Catalytic Reactive Distillation (CRD)

The CRD is being studied by us in the production of oxygenate tertiary ethyl ethers [25-27] for single (e.g., MTBE or ETBE) or multiple reactions (e.g., TAME or TAEE). This work is being done in collaboration with Dr. Moses Tadé of Curtin University, Australia, and encompasses modeling, experimentation, and development of control strategies for CRD. Further, although theoretical simulations abound, there is a paucity of experimental thermodynamic data for the simultaneous reaction and vapor-liquid equilibrium. We are in the process of gathering these data for some tertiary ethyl ethers along with a thermodynamic analysis, which should shed light on some interesting, and not intuitively obvious, interactions [27]. For instance, it is believed that non-ideality (Kg) in the liquid-phase can sometimes aid in attaining higher equilibrium conversion. This advantage is lost, however, when a vapor phase, typically ideal, is also present, as in CRD. Thus, in some important cases, CRD may actually provide lower conversion than a simple liquid-phase packed-bed reactor [27].

Catalytic Micro-Kinetics

Detailed kinetic studies are an integral part of most projects. The objective is to develop robust kinetic expressions based on detailed molecular mechanisms coupled with predicted energetics of the elementary steps.

Response Reactions and the UBI-QEP Method

We have developed a theory of reaction routes in catalytic reactions based on a thermodynamic approach leading to the so-called response reactions (RERs), which are a unique set of reactions describing a given chemical system with a specified set of species [21]. This is quite different from the usual approach, in which the overall reactions as well as the mechanistic steps are picked arbitrarily from an infinite set. Algorithms have also been developed for the systematic generation of direct overall reactions and direct mechanisms from the RERs [28]. The energetic characteristics, namely enthalpy changes and activation energies, of the elementary reactions are predicted based on the unity bond index - quadratic exponential potential (UBI-QEP) method. This allows determination of the energetically most favorable pathways [29] as well as a prediction of the reaction kinetics of the overall reactions based on the De Donder approach.

Extrathermodynamic Correlations

For liquid-phase reactions especially, no theoretical methods are available for the prediction of reaction kinetics. Thus, semi-empirical approaches such as extra-thermodynamic correlations (ETC , i.e., relations between kinetics and thermodynamics, must be resorted to. We develop the overall rate expressions based upon the application of the thermodynamic transition-state theory (TTST) to the elementary steps within the conventional catalytic kinetic approach, e.g., the LHHW formalism. This approach shows that it is appropriate to write rate expressions for liquid-phase catalytic reactions in terms of activities, which has been experimentally confirmed for the family of tertiary alkyl ethyl ethers [11, 15, 16, 30]. The TTST formulation also provides a rationale for the extrathermodynamic correlations (ETC) observed in this family of reactions [30].

Diffusional Transport in Porous Catalysts and in Membranes

Transport in Porous Catalysts

We are developing the detailed theory of multicomponent diffusion and reaction in porous media for both liquids and gases and applying it in a number of situations. Thus, research is being conducted in the development of a unified theory of multicomponent diffusional transport in porous media and in membranes that is based on the species linear momentum balance to obtain the diffusion driving force and generalized Maxwell-Stefan diffusion equations [31], followed by volume-averaging to yield the dusty-fluid model (DFM) for fluids of arbitrary state of aggregation. DFM is extended to include d'Arcy flow, surface diffusion, and solid-phase diffusion. Solution procedures involve matrix manipulations. Thus, we have developed sophisticated models of transport and reaction in SLPC [1-3], thermal decomposition of solids [32], SLPCMRS [23], and porous catalysts and PWRS [24].

Transport in Membranes

The DFM has been applied to develop a generalized model of transport of gases in membranes [33]. We are also studying diffusional transport phenomena and fouling in ultrafiltration [34]. Further, we have developed a simple and accurate model for predicting the osmotic pressures of single and binary protein mixtures at high concentrations typical of those found at the membrane surfaces [35, 36] by simply using a free solvent model that excludes the protein hydration layer. Osmotic pressure, of course, needs to be accounted for in the obtaining the real driving force for protein ultrafiltration. This work is in collaboration with Professor Victor G. J. Rodgers of the University of Iowa. A theoretical model has also been developed to predict proton conductivity in proton-exchange membranes used in fuel cells [37].

Maintained by webmaster@wpi.edu
Last modified: October 24, 2007 09:34:52