Page 292 - Distillation theory
P. 292

P1: JPJ/FFX  P2: FCH/FFX  QC: FCH/FFX  T1: FCH
            0521820928c08  CB644-Petlyuk-v1                                                      June 11, 2004  20:20





                        266    Synthesis of Separation Flowsheets

                               utilities of heat and cold are not used in all columns of each sequence. Besides that,
                               such an approach presupposes that only simple columns are used and the number
                               of such columns is minimum (for the separation of an n-component mixture [n– 1]
                               columns are used in each column there is separation without distributed compo-
                               nents).
                                 Synthesis experiments (Kafarov et al., 1975; Glinos & Malone, 1984) proved
                               that energy expenditures reflected by summary vapor flow V constitute the main
                               share of yearly expenditures, and capital costs do not differ greatly for alternative
                               sequences. Therefore, the flowsheet with smaller summary vapor flow V,asa
                               rule, requires smaller total expenditures on separation. This allows, in the first
                               approximation, to use the summary vapor flow V for estimation of preferability
                               of the sequences at the above-mentioned limitations.
                                 If expensive utilities of heat or cold are used in the columns, for example, at low-
                               temperature separation, then the above-described way of expenditures estimation
                               is not good. In this case, the temperature at which heat should be brought in in the
                               reboilers and withdrawn in the condensers acquires great importance. The total
                               energy expenditures on separation can be estimated by the value of energy of
                               separation that depends on the amount of heat brought in and withdrawn, and on
                               the temperatures at which this heat is brought in or withdrawn (see Chapter 4):


                                    W ≈ T 0   Q k /T k                                          (8.1)
                               where k is number of reboilers or condensers, T 0 is temperature of ambient air,
                               and Q k is amount of heat brought in or withdrawn.
                                 For a column with one reboiler and one condenser in the case of liquid feeding,
                               instead of Eq. (8.1), we have:

                                    W ≈ QT 0 (1/T con − 1/T reb )                               (8.2)
                                 Energy expenditures in the case of the thermodynamically reversible process of
                               separation into pure components (Petlyuk & Platonov, 1964; Petlyuk, Platonov, &
                               Girsanov, 1964):
                                                n

                                    W rev =−RT    x Fi ln x Fi                                  (8.3)
                                               i=1
                                 To estimate preferability of different separation flowsheets, one can use ther-
                               modynamic efficiency instead of energy:

                                   η = W rev /W                                                 (8.4)
                                 Such an approach was used in the works (Petlyuk & Platonov, 1965; Petlyuk,
                               Platonov, & Slavinskii, 1965) and in more detail in Agrawal and Fidkowski (1998,
                               1999). In the latter works the comparison not only of sequences of simple columns,
                               butalsoofdistillationcomplexes,wasmadeforthree-componentmixtures.Capital
                               costs can differ considerably for such separation flowsheets, but in this case it is
                               not important because energy expenditures greatly exceed capital ones.
                                 In other cases, if there is a difference of columns number in alternative se-
                               quences, for example, if a flowsheet with prefractionator and flowsheets with
   287   288   289   290   291   292   293   294   295   296   297