Page 306 - Process Equipment and Plant Design Principles and Practices by Subhabrata Ray Gargi Das
P. 306

308    Chapter 11 Distillation




             separated by other components, the group of non-key components which have their volatility between
             the light and heavy key are termed as split key components.
                As a split key example we may consider the same feed being separated into LPG as top product
             having an allowable maximum limit of 5% total C 5 (nC 5 H 12 þ iC 5 H 12 ) content. Such a limit actually
             exists to limit heavy components in LPG and ensure that empty LPG cylinders contain no residual liquid.
             In this case the separation target needs to be set in terms of separation between nC 4 H 10 and nC 5 H 12 (the
             heavier of the two pentanes). Therefore, the light and the heavy keys are nC 4 H 10 and nC 5 H 12 ,
             respectively. The keys in this case are separated by iC 5 H 12 (split key) having an intermediate volatility.
                To summarise, one may state that the selection of key components depend on the separation targets
             for the design problem and the features of key components are

             •  Both light and the heavy keys will be present in distillate and the bottom stream
             •  Split key components will also be present in both the streams
             •  None of the components lighter than the light key will be present in the bottom product
             •  None of the components heavier than the heavy key will be present in the top product
                Separation sequencing for multicomponent system
                The question of sequencing arises only when the feed has to be separated into more than two
             streams. Different options for separating a feed to three different streams have already been presented
             in Fig. 11.2. The correct way to arrive at the most appropriate sequence for separation is through
             economic analysis. The economic criterion can be the minimum total annualised cost, i.e., the sum-
             mation of the annualised cost of investment and the annual operating cost. This is often a complicated
             and difficult procedure as the configurations to be compared are with optimally designed individual
             columns. Procedure for optimum design of individual column configurations for binary fractionation
             has already been discussed (Section 11.4.2) and the same approach is valid for multicomponent
             system.
                As a preliminary/quick evaluation alternative, the following heuristic rules are often applied for
             non-heat integrated distillation columns:
              (i) Separations where the key components have relative volatility close to 1 or form azeotrope
                 should be performed in absence of nonkey components. In other words, the easier separation
                 should be performed in the first column.
              (ii) Sequences that remove the lightest components one by one in order of decreasing volatility may
                 be favoured. This means that the direct sequence is preferred.
             (iii) A component present as a large fraction in feed should be removed first.
              (iv) Column configurations with as similar as possible molar flow of the top and bottom products are
                 favoured.
                Shortcut design methods
                The steps in shortcut design of multicomponent distillation column involves a few empirical
             equations, namely the Fenske equation, Underwood equation and the Gilliland equation, each being
             known with their proposer’s name. The method is called F-U-G method in short and is shown in
             Fig. 11.12 in the form of a flow chart. In addition, Kirkbride equation is used to decide the location of
             the feed tray in the column.
   301   302   303   304   305   306   307   308   309   310   311