(26b) Hydrocarbon Strippers - the Importance of Water Solubility | AIChE

(26b) Hydrocarbon Strippers - the Importance of Water Solubility

Authors 



Many control room operators and design engineers assume that stripping steam added to a distillation column, will infallibly end up in the column overhead. This is normally not the case, especially not for columns operating at high pressures. Since the water flow from overhead reflux drums is rarely measured, there is the chance that a column can run for an extended period of time with no steam reaching the column top, without anyone noticing.

There are two situations when no stripping steam goes through the column and hence, the added stripping steam does not do any stripping, but does only wet the bottom product. The first situation is when the amount of stripping steam used is less than the amount of steam that can be dissolved in the hydrocarbons in the bottom of the column. This situation might seem a bit unrealistic, but for all three high pressure strippers (5,8 barg or higher) in one refinery, 50% or more of the stripping steam used, was dissolved in the hydrocarbons in the bottom of the columns. There have even been incidents in the past where all or nearly all of the stripping steam ended up in the bottom of the column due to this fact. For one of the columns, the amount of stripping steam that dissolved in the bottom hydrocarbons, was more than 75% of the design flow rate of stripping steam.

The second situation that can render stripping steam useless, is when the inlet temperature to the column is lower than the condensation temperature for steam at the column bottom pressure. This situation might seem even more unrealistic. Even so, a column in a HDS unit running in block mode, switching back and forth between processing kerosene and light gasoil every week, had complete condensation of all added stripping steam every time the column was running in kerosene mode, for at least 10 years. This led to very poor stripping of light gases in the column, which led to excessive flaring from a downstream fractionator every time the unit was running in kerosene mode. Since excessive flaring is not allowed, this limited the unit throughput. Increasing the column inlet temperature by about 15 °C was enough to avoid that the stripping steam condensed and the flaring problem disappeared.  

Many simulation tools using equation of state models to estimate vapour-liquid equilibrium, have problems estimating three phase equilibrium when water is present. A simple equation for maximum water solubility in hydrocarbons has been suggested in literature. The equation states that the water solubility does only depend on temperature, pressure and average molecular weight for the hydrocarbon. That equation was compared with the observed solubility in three columns with oil fractions with very different molecular weights. The estimated solubilities are in good agreement for all fractions, and can be used to check the simulation results from more rigorous models in the design of new units.