Multistage separations

The simplest form of condensate stabilization is to install a low-pressure separator downstream of an initial high-pressure separator. Unless the gas well produces at low pressure (less than 500 psi) and the gas contains very little condensate (less than 100 bpd), the additional expend - 

---

The calculation of single-stage equilibrium separations in multicomponent systems is implemented by a series of FORTRAN IV subroutines described in Chapter 7. These treat bubble and dewpoint calculations, isothermal and adiabatic equilibrium flash vaporizations, and liquid-liquid equilibrium "flash" separations. The treatment of multistage separation operations, which involves many additional considerations, is not considered in this monograph. 

Solids may be processed continuously or semicontinuously by pumping slurries or by using lock hoppers. An example is the separation of insoluble polymers by floatation with a variable-density SCF. For liquid feeds, multistage separation may be achieved by continuous counter-current extraction, much like conventional liquid-hquid extraction. The final produces may be recovered from the extract phase by a depressurization, a temperature change, or by conventional distillation. 

In almost all cases the molecules have a higher value as liquid than as gas. Crude oil streams typically contain a low percentage of intermediate components. Thus, it is not normally economically attractive to consider other alternatives to multistage separation to stabilize the crude. In addition, the requirement to treat the oil at high temperature is more important than stabilizing the liquid and may require the flashing of both intermediate and heavy components to the gas stream. 

Figure 6-1 shows a multistage separation process. By removing molecules of the light components in the first separator they are not available to flash to gas from the liquid in the second separator, and the partial pressure of intermediate components in the second separator is higher than it would have been if the first separator did not exist. The second separator serves the same function of increasing the partial pressure of the intermediate components in the third separator and so forth. 

The sample volume initially introduced onto the sorbent, the choice of sorbent and solvent system and careful control of the amount of solvent used are of paramount importance for effective pre-concentration and/or clean-up of the analyte in the sample. The number of theoretical plates in an SPE column is low (/V = 10-25). SPE is a multistage separation method and as such requires only a reasonable difference in extractability to separate two solutes. In SPE concentration factors of 1000 or more are possible, as compared to up to 100 for LLE with vortex mixing. 

Smith and Brinkley developed a method for determining the distribution of components in multicomponent separation processes. Their method is based on the solution of the finite-difference equations that can be written for multistage separation processes, and can be used for extraction and absorption processes, as well as distillation. Only the equations for distillation will be given here. The derivation of the equations is given by Smith and Brinkley (1960) and Smith (1963). For any component i (suffix i omitted in the equation for clarity)... 

Continuous multistage separation by zone melting. J. Metals 7, 297 (1955). 

Equations 9.109 and 9.110 are valid for a boiling process in a closed system the gaseous phase develops from a liquid with initial concentration Cq, is the mass fraction of developed gas, and K is the mass distribution constant of the component of interest between gas and liquid. Equations 9.111 and 9.112 describe a multistage separation process in which n is the number of separation stages, AFg is the mass fraction of gas separated in each stage, saiAK is the mean mass distribution constant of the process. Equations 9.113 and 9.114 refer to a boiling process in an open system the gas is continuously removed from the system as the process advances. 

The description of any operation or design problem in a multistage separation process requires assigning numerical values to, or setting, a certain number of independent variables. The number of variables to be set depends on the process, and is usually determined easily by the method the authors have called the description rule (HI). Alternatively the number of variables to be set may be determined by writing all of the independent equations which define the process, then counting the number of variables and the number of equations. In order to solve the equations, a sufficient number of independent variables must be set so that the number of dependent variables remaining equals the number of equations. 

Relaxation solutions are conceptually the most simple methods of solution for any multistage separation process. They were first proposed by Rose et al. (R2) in 1958 and Duflin (Dl) in 1959. Both authors in... 

Dl. Duffin, J. H., Solution of Multistage Separation Problems by Using Digital Computers. Ph.D. Thesis, University of California, Berkeley, 1959. 

D.N. Hanson, J.H. Duffin and G.F Sommerville, Computation of Multistage Separation Processes, Reinhold, New York, 1962. 

We have no better starting point to offer. We will discuss some rearrangements of flow schemes which are possible with multistage separation processes, and the accompanying recompression and cooling of flash vapors, which serve to partially simulate the arrangement of a conventional fractionation process. Several aspects of thfese processes which influence their selectivity will be discussed. 

It has long been known in the industry that increasing the number of stages In a multistage separation system vill improve recovery in terms of barrels of oil per unit volume of well stream. 

It is of Interest that these are essentially the same results that have historically been obtained by the industry for multistage separation systems in onshore areas. 

Beyond a certain complexity these analytical relations between vapor and liquid compositions lose their utility. The simplest one, Eq. (13.14), is of value in the analysis of multistage separating equipment. When the relative volatility varies modestly from stage to stage, a geometric mean often is an adequate value to use. Applications are made later. Example 13.1 examines two ways of interpreting dependence of relative volatility on composition. 

Figure 13.20. Algorithm of the SC (simultaneous correction) method for all multistage separations of fluid mixtures [Naphthati and Sandholm, AIChE J. 17, 148 (1971), Henley and Seader, 1981].

In rate-based multistage separation models, separate balance equations are written for each distinct phase, and mass and heat transfer resistances are considered according to the two-film theory with explicit calculation of interfacial fluxes and film discretization for non-homogeneous film layer. The film model equations are combined with relevant diffusion and reaction kinetics and account for the specific features of electrolyte solution chemistry, electrolyte thermodynamics, and electroneutrality in the liquid phase. 

Success of rate-based multistage separation modeling is ultimately tied to underlying equipment hydrodynamics performance correlations for tray or packed columns. For example, the thickness of the film... 

---