The effect of diffusion in the isomerization of normal butane

It Is the purpose of this thesis to develop a method whereby diffusion effects for a solid catalyzed gaseous reaction system can be deduced directly from experimental data.

The system chosen for study is the catalytic isomerizatlon of normal butane over an aluminum chloride catalyst. The reaction is first-order with no significant side reactions at ordinary operating conditions.

By placing certain limitations on the reacting system and through mathematical manipulations of the basic design equation for catalytic reactors and of various reaction rate equations, a method is obtained which allows the effect of diffusion to be determined by measuring changes in concentrations at various flow rates. If the over-all reaction is first-order, each step in the process, including diffusion (or mass transfer), surface adsorption, and surface reaction, may be considered as a resistance in series. The reciprocal of the over-all reaction rate constant, l/Kov, can be expressed as the sum of the individual resistances. l/Kov= 1/Kp +1/kc+1/ka. One of the initial stipulations is that adsorption is rapid when compared to the other steps: therefore, 1/Kov = 1/kp+1/kc. Since the mass transfer coefficient is proportional to G^n and since the surface reaction rate is independent of the flow rate, a plot of 1/Kov versus 1/G^n should be linear. By extrapolating the straight line to zero (or infinite mass velocity), the effect of diffusion can be separated from other resistances, n* is the constant which gives the best fit of the experimental points.

Experimental results indicate that diffusion Is the controlling process; however, a break is indicated in the curve as a probable result of changing from laminar to turbulent flow. The value of n' for the lower velocities is .76 while n’ equals .70 at the higher velocities. Because of the break in the curve, two separate curves are required to present the findings.