PlGUREf 12.11.1 SOLVENT RECOVHRy COLUHN
cftse slvdy HO. 11
CASE STUDY NO. 12 (Contributed by F. Wetherill, Santa Fe Braun Inc., Alhambra, Calif)
INSTALLATION A specialty chemical plant producing a heavy, water soluble alcohol. Effluent from a front-end hydrogenation reactor was purified into the desired product in a four column separation plant, shown as the heavy lines on Figure 12.12.1.
The first two columns separated water from the feed. Final traces of water were removed in the flash drum overheads. The intermediate column removed the low boiling organic impurities. Finally, the product was separated from the high boiling residues in the product column. The high-boilers were mostly other alcohols and aldehydes as well as some inorganic salts.
I PROBLEM Although the water content of the feed was removed in the
! first two columns and flash drum, the final product consistently showed a small water content. It lowered the product freeze point and was confirmed by chemical analysis.
INVESTIGATION The feed to the product column was analyzed and was found to contain no water. The column was reboiled by high pressure steam and condensed by cooling water, and it was speculated that one of these exchangers could have been leaking. In order to check this, the product column was placed on total reflux, and the overhead analyzed periodically for water. These tests showed an increasing quantity of water in the product with time, even though no feed entered the column. It was concluded that the problem occurred in the product column.
A leak in either the condenser or the reboiler was suspected. The product column was shut down, and the exchangers were pressure tested. Neither exchanger showed any signs of leakage.
SOLUTION Following the investigation, a new theory was formulated. The bottom temperature wa,s run at about 360°F. It was economical to run this temperature as hot as possible in order to maximize alcohol product recovery from the residue. As most of the heavy impurities were alcohols and aldehydes, it was believed that these might undergo condensation reactions to form acetals and water. Such a reaction would tend to be promoted by higher temperatures.
In order to suppress these possible reactions, column base temperature was lowered to 320°F. At this lower temperature, the problem disappeared. However, product recovery also dropped because the lower bottom temperature allowed more product loss with the residue.
Was this article helpful?