The distillation of true binary systems is rarely encountered in real life applications. The presence of other components, whether volatile or not, often complicate the analysis of separating the components by traditional binary calculations.

Consider a quaternary mixture A, B, C, D, with all components having varying volatilities. If this mixture is fed to a fractionating column, we can only purify two of the four components at a time; multiple columns are needed to obtain fractions that are rich in each of the four components. In every column, there will be a light key and a heavy key component, which are the two components to be enriched in the distillate and bottoms, respectively. Take note that the light key is not necessarily the most volatile compound in the mixture; the same goes with the heavy key not necessarily being the least volatile component.

In this module, we will discuss a shortcut iterative method in designing a multicomponent distillation column.

The Fenske-Underwood-Gilliland (FUG) Shortcut Method

For the design of a new column where a target relationship for the operating reflux and minimum reflux ratio is known, the FUG shortcut method may be used. This involves a series of calculations and would be much manageable when done using a spreadsheet where iterative calculations are better handled.

The following are the general steps in the preliminary design of a multicomponent distillation tower:

1. For the first run, assume that all of the light key component is obtained in the distillate and all the heavy key component is obtained in the bottoms. Assume arbitrary preliminary composition values for the other components.
2. Determine the dew point temperature at the top of the column.
3. Determine the bubble point temperature at the bottom of the column.
4. Determine the minimum number of stages N_min required (for total reflux) using the Fenske equation.
5. Using the solved minimum number of stages, determine the distribution of the other components. If this is not the same as the previously assumed values, go back to step 2 (calculation of dew point temperature at the top of the tower).
6. Determine the minimum reflux ratio R using the Underwood equations.
7. With a known ratio between R and R_min, determine R (actual reflux ratio).
8. Using values of R, R_min, and N_min, determine the actual number of stages N using either the Gilliland or Erbar-Maddox correlation, or the Molokanov equation.
9. The ideal feed plate location can be determined from the Kirkbride equation.
10. Sit back and marvel at your creation.

Easy, right? The name of the method indicates the developed equations to be used in order to complete the design of the column. If you were not able to design your column using the above list, fear not because the next sections will describe all the steps in detail. Be sure to follow along with your own spreadsheet to create your program as described in this guide.