(79c) On the Improper Implementation of the Murphree and Vaporization Efficiencies in Process Simulators

Many of the commonly used chemical process simulation tools today offer the capability to apply efficiency adjustments to individual equilibrium stages or sections, aiming to better match the separation performance of those stages with real-world tray performance. Some simulators even permit the use of component efficiencies at the stage level or within stage sections. However, we believe that two prevalent efficiency models, namely vaporization efficiency and Murphree vapor phase efficiency, have often been implemented incorrectly in these simulators. The resulting errors in the column temperature, composition, and liquid/vapor phase profiles can be profound. The errors arise from three significant conceptual inconsistencies:

1) The assumption that liquid and vapor streams always leave a stage at the same temperature, even when considering stage or component efficiencies.

2) Mandating the specification of efficiencies for all "C" components when only "C-1" are truly independent.

3) In the case of vaporization efficiency, assuming that equal component efficiencies for all components is mathematically and thermodynamically consistent.

These three assumptions can result in unfeasible column composition and temperature profiles - the result of serious violations of fundamental thermodynamic principles. To rectify these mathematical and thermodynamic issues, we propose the following adjustments:

1) Always treating the liquid and vapor streams leaving a stage as bubble point liquids and dew point vapors.

2) Specifying efficiencies for only "C-1" components.

We have created a rigorous column model in Excel using the Aspen Properties V12 Excel add-in that treats vaporization and Murphree vapor efficiencies properly. We will compare the results of several simulations performed with this corrected model to results from a number of commercial simulators available today.