The geometry of catalyst-containing bale packing is characterized in this paper. The calculated packing parameters (specific surface area and void fraction) are employed in conjunction with a model to predict two-phase pressure drop, maximum capacity, and height equivalent to a theoretical plate (HETP). Experimental data obtained in a 5.3-cm (2.1-in.) column, operated at total reflux, are presented for two systems (cyclohexane/n-heptane and acetone/methyl ethyl ketone) at pressures of 138 and 241 kPa (20 and 35 psia). Model predictions for pressure drop and HETP are validated with experimental data obtained under nonreactive conditions. An appropriate procedure for scaleup of HETP and pressure drop, with associated limitations, is also discussed.

No abstract

Reactive distillation has been proposed as a means of enhancing the conversion for reactions whose progress is limited by chemical equilibrium. In the present work, reactive distillation tests were carried out for the hydration of isoamylene (2-methyl-2-butene) to 2-methyl-2-butanol (tert-amyl alcohol or TAA) in a 5.3-cm column using a bale-type catalytic packing. Acetone was used as a solvent to avoid formation of a second liquid phase and also to enhance the rate of reaction. The experimental results showed TAA yields approaching 100% when the distillate stream from the top of the column was essentially suppressed. This yield is more than double that possible thermodynamically in a separate reactor. It was also found that an excess of catalytic packing can generate operational problems. The column was simulated using an equilibrium stage model together with the kinetic expression developed for this reaction and reported in Part 1 of this paper. The simulation and the experimental results matched reasonably well. Model results indicate that the TAA yield can be greater than 95% (at a pressure of 240 kPa) when the weight hour space velocity is lower than 30 kg/(h‚kg). A parametric study elucidated the effect of varying the feed composition and key operating variables.

A model was developed to predict the flowdynamic behavior of bale-type catalytic packings, which are widely used in the industry for several applications. The new model takes into account the effect of the catalyst properties and packing characteristics as well as the properties of the fluids, flow rates, and column diameter to predict liquid hold-up, pressure drop, and the loading and flooding points. The particle model concepts were used as the basis for the generation of the model, introducing considerations that account for a proper representation of a wide variety of data obtained in tests performed in this work or reported previously in the literature. It was found that the nature of the catalyst contained in the "bales" strongly affects the liquid hold-up due to solid-liquid interactions. A factor was also introduced in the model to account for the strong effect of the wall on small-diameter columns. The new model 855 ORDER REPRINTS significantly improved the prediction of the flowdynamic parameters such as hold-up and pressure drop as well as loading and flooding points, which showed deviations smaller than 10 -30%.

No abstract