Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Chemical & Biochemical Engineering

First Advisor

Paul L. Barrick

Second Advisor

Lee F. Brown


The low temperature ortho-parahydrogen shift reaction has long been considered to be an ideal one for the study of catalytic reaction mechanisms. This is due to its being a simple reaction in which the reactant is easily purified, side reactions are non-existent, and heat effects are quite small. For these reasons a large number of studies have been conducted on this reaction. Most investigators have concluded that their kinetic data at temperatures below 100°K were explained by a first-order reaction rate law. However, there have been a few investigators who did not agree that the reaction is a simple first-order reaction and theoretical investigation showed that, except under certain conditions, the reaction should not be of first order type.

An apparatus was therefore devised which could be used to collect kinetic data at 76 °K for the reaction. A hydrous ferric oxide gel catalyst was used to promote the reaction and a pressure range of 30 to 1010 psia in geometric intervals was covered. Since the ortho-parahydrogen reaction is reversible at 76 °K, a study of the reaction in both directions is quite useful in determining a reaction mechanism--according to the principal of microscopic reversibility, the mechanism must be the same in both directions. No single investigation on the reaction in both directions using the same catalyst and covering identical pressures and temperatures has previously been reported.

Rate expressions based on different sets of theoretical assumptions are derived and tested against the data. Models which assume a three-step mechanism consisting of adsorption, surface reaction, and desorption are postulated and an expression is derived when each of the three in turn controls the rate of reaction. Four different types of adsorption laws are used in developing mechanisms. The names of Langmuir, Temkin, Elovich, and Freundlich are associated with these adsorption laws. In addition, for the Langmuir case, an expression is derived which postulates that none of the three steps controls the rate.

The results show that a good correlation of the data is obtained from both the Langmuir and Elovich models. However, there is a deviation with pressure which is not predicted by the theory in the rate constants of the Elovich expressions, and there is a difference in value between the rate constants of the Langmuir expression for the forward and reverse directions. It is shown that if some of the basic assumptions of the Langmuir model are modified,e.g., if adsorbent surface heterogeneity or interactions between adsorbate molecules are assumed, both of which result in a non-uniform heat of adsorption, then the experimental variation in the Langmuir rate constants can be predicted. It is concluded that a model in which the reactant adsorbs onto the surface with a non-uniform heat of adsorption, reacts, and then the product desorbs from the surface, with the surface reaction being the rate-controlling step, is the best explanation of the ortho-parahydrogen shift reaction.