Distillation, King in separations, will remain m the workhorse separations device of the process industries. Even though it is old in art, with a relatively mature technology mpport base, it attracts research and ^afessional interest. Without question, ¿sanation will sail into the future with dear skies and a strong wind. It will remain Ac key separation method against which mitcrnate methods must be judged.
Distillation is the most important and most stable separation technology. The skyline of many refineries and chemical plants is dominated by tall distillation towers, and it «wot be denied that their spatial arrangements often amounts to architectural bmuty. Less spectacular, but also visible are the smoke stacks of these industrial complexes; they represent the energy used in processing, the mqjor part of which is consumed by distillation processes.
DR. FRITS J. ZUIDEKWBO, 1988
As we move through the 1980's, with full recognition of the energy inten&iveness of distillation, we can expect to see relatively BOle displacement of distillation by tOernative separation methods, at least for &e large scale process throughputs. Thus, development of distillation devices will continue. The result will be improved wqnration efficiency at lower pressure drop mud lower cost. dr. james r. fair, 1983
1.1 Distillation Background
1.1.1 What is distillation?
Distillation is an ancient unit operation, and has been widely practiced for thousands of years. Early applications used crude vaporization and condensation equipment, often for concentrating the alcoholic content of beverages. The first vertical columnar continuous distillation still was developed by Cellier-Blumenthal in France in 1813. Perrier introduced an early version of the bubble-cap tray in England in 1822. Packings were used as early as 1820 by a technologist named Clement who used glass balls in an alcohol still. Coffey developed the first sieve tray column in 1830. The first book on fundamentals of distillation was La Rectification de l'alcohol by Ernest Sorel in 1893.
During the first quarter of the twentieth century, the application of distillation expanded from a tool for enhancing the alcohol content of beverages into the prime separation technique in the chemical industry. This expansion accelerated once distillation was recognized as an effective means of separating crude oil into various products. From there, the application of distillation spread into the majority of chemical processes. Detailed descriptions of the history of distillation, including illustrations of historic exhibits, are given by Fair (1), Underwood (2), and Forbes (3).
Distillation is a process of physically separating a mixture into two or more products that have different boiling points, by preferentially boiling the more volatile components out of the mixture. When a liquid mixture of two volatile materials is heated, the vapor that comes off will have a higher concentration of the more volatile (i.e., lower-boiling-point) material than the liquid from which it was evolved. Conversely, if a vapor is cooled, the less volatile (i.e., higher-boiling-point) material has a tendency to condense in a greater proportion than the more volatile material.
1.1.2 Why distillation?
Distillation is a unit operation that has been around for a long time and continues to be the primary method of separation in processing plants, in spite of its inherently low thermodynamic efficiency. The preeminence of distillation for the separation of fluid mixtures is not accidental, but fundamental, and therefore unlikely to be displaced. The reasons are both kinetic and thermodynamic.
From a kinetic standpoint (4), mass transfer per unit volume in distillation is limited only by the diffusional resistances on either side of the vapor-liquid interface in turbulent phases, with no inerts present. In almost every other separation process, there are inert solvents or
■did matrices present, and these lower mass fluxes. Distillation, therefore, has the potential for high mass transfer rates (low capital costs).
From a thermodynamic viewpoint, a typical thermodynamic efïi-riency of a distillation system is about 10 percent (4). This can be enhanced if intercondensers and interreboilers are used. In fact, it has been shown that conceptually, a distillation system can be devised which requires only the minimum work of separation. Although a thermodynamic efficiency of 10 percent appears low, not many other processes are more efficient (4).
Distillation in general provides the cheapest and best method for separating a liquid mixture to its components (4-8), except when (4):
L The difference of volatility between the components is small.
2. A small quantity of high-boiling-point component is to be recovered from the feed. Distillation requires that the whole feed be vaporized in order to recover this small quantity.
3. A compound is thermally unstable even under vacuum conditions.
4. The mixture is extremely corrosive or highly fouling.
12 Vapor-Liquid Equilibrium (VLE)
H is difficult, perhaps altogether impossible, to do justice to the wide topic of vapor-liquid equilibrium (VLE) in a modest amount of space. Many texts are devoted entirely to this topic, or even to fractions of it. The numerous published texts and reviews (e.g., Refs. 9 through 24) «an testify to the large volume of information available.
It is also difficult to discuss distillation without addressing some of flhe implications of VLE to distillation design. For this reason, some Ascussion of VLE is included in this text. This discussion focuses on those VLE principles that in the author's opinion must be understood ^ distillation practitioners. Extensive theoretical discussions and presentation are excluded, and left to the thermodynamic texts. Discussions on predictive models and procedures are far too bulky and have also been excluded. The author recommends Refs. 9 to 26 to those •eeking further information. References 25 and 26 are brief, practical, ad very useful state-of-the-art reviews.
1.2.1 K-value and relative volatility jç _ Mole of fraction of component i in vapor phase ^
' Mole of fraction of component i in liquid phase
U»e if-value is a measure of the tendency of component i to vaporize.
If the if-value is high, the component tends to concentrate in the vapor; if low, it tends to concentrate in the liquid. If the K-value is unity, the component will split equally between the vapor and the liquid.
The /f-value is a function of temperature, pressure, and composition. At equilibrium, whenever two of these three variables are fixed, so is the third. The if-value can therefore be regarded as a function of pressure and composition, or temperature and composition (or temperature and pressure).
The relative volatility of components i and.; is defined as
X-value of component j
Distillation is a technique of separating components according to their relative volatility. The relative volatility is a measure of the ease of separation. This definition makes the relative volatility the ratio between the tendency to vaporize of the two components. If relative volatility is high, one component has a much greater tendency to vaporize (i.e., is more volatile) than the other, and it will be easy to separate the two by vaporizing one from the other (i.e., by distillation). On the other hand, when one component has as high a tendency to vaporize (i.e., is as volatile) as the other, relative volatility will approach unity, and the components will be difficult to separate from each other by distillation. If relative volatility is unity, each component is as volatile as the other, and they cannot be separated by distillation.
By convention, relative volatility is defined as the ¿¡f-value ratio of the more-volatile to the less-volatile component, and therefore, relative volatility is always greater or equal to unity.
For binary system, Eqs. (1.1) and (1.2) can be combined to give
This equation can be rearranged to give
Equation (1.4) expresses the more-volatile component (MVC) mole fraction in the vapor as a function of the mole fraction of the MVC in the liquid and the relative volatility. This relationship is plotted in Fig. 1.1.
Figure 1.1a is a plot called an x-y diagram. The x and y axes show the concentration of the MVC in the liquid and in the vapor, respectively. The 45° diagonal represents points at which vapor and liquid
Rgur» 1.1 The relative volatility concept, (a) Concentration <rf the more-volatile component (MVC) in the vapor. (b) Effect of relative volatility on the concentration of the MVC in the vapor.
compositions are the same. The curve on Fig. 1.1a is an equilibrium relationship. Figure 1.1c illustrates how the MVC concentrates in the vapor. A liquid mixture containing x1 mole fraction MVC = 0.45 in Fig. 1.1a) is in equilibrium with vapor containing mole fraction MVC = 0.71 in Fig. 1.1a). If this vapor is collected and condensed, one will end up with a mixture in which the MVC mole fraction has been enriched from 0.45 to 0.71.
Figure 1.16 illustrates the effect of relative volatility on the tendency of the MVC to concentrate in the vapor. When volatility is high, the enrichment is large. For instance, when relative volatility is 10, a liquid mixture containing 0.45 mole fraction MVC is in equilibrium with vapor containing 0.88 mole fraction of MVC; it would take only a few steps to convert the liquid mixture into pure components. Conversely, when relative volatility is vety low, say 1.1, a liquid mixture containing 0.45 mole fraction MVC is in equilibrium with vapor containing 0.47 mole fraction MVC. Under these conditions, it will take a very large number of steps to separate the mixture into the pure components.
1.2.2 Ideal and nonideal systems
An ideal system is one where the vapor obeys the ideal gas law and the liquid obeys Raoult's law. An ideal gas mixture obeys Dalton's law, i.e.,
An ideal solution obeys Raoult's law, which states that the partial pressure of a component in solution is equal to the product of its mole fraction and of the vapor pressure of the pure component; thus,
From Eqs. (1.5) and (1.6) and the definition of K-value (Sec. 1.2.1), one obtains
For nonideal systems, the fugacities of component i in the vapor and in the liquid play the same role as the component partial pressure in the vapor and the component vapor pressure in the liquid. The fugacity can be regarded as a thermodynamic pressure. For equilibrium, vapor fugacity is equal to liquid fugacity, i.e.,
The vapor fugacity can be regarded as a corrected partial pressure, given by the equation fi = 4>,v"(Py.) (1.9)
Similarly, the liquid fugacity can be regarded as a corrected vapor pressure, given by ft = 4-b,<i<,0,p;) (1.10)
A detailed derivation of these equations from the thermodynamic relationships is presented in most thermodynamics texts (e.g., Refs. 9-12, 15). The various coefficients in Eqs. (1.9) and (1.10) are discussed below.
<!>Y The vapor fugacity coefficient. It accounts for the effect of vapor nonideality on vapor fugacity. It is usually estimated from an equation of state and is based on system temperature, pressure, and vapor mole fraction.
cbf The liquid fugacity coefficient. It accounts for the effect of vapor nonideality on liquid fugacity. This coefficient is estimated in a similar manner to the vapor fugacity coefficient, but it is based on the system temperature and the pure component vapor pressure.
i), The Poynting correction factor. It accounts for the effect of pressure on liquid fugacity. Since 4>f is evaluated at the vapor pressure of the pure component, ^ is used to account for the difference between the pure component vapor pressure and the mixture pressure. This effect is small and can be neglected at low pressures (11,27), but is important at high pressures.
•y, The liquid activity coefficient. It corrects the liquid fugacity for the effect of composition. Its value depends on how similar the components are. For two similar components, such as an isobutane-normal butane mixture, the liquid activity coefficient is close to unity. If the components are different, activity coefficients deviate from unit.
Combining Eqs. (1.1), (1.8), (1.9), and (1.10) one gets
1.2.3 Effect of temperature, pressure, and composition on K-values and volatility
For the purpose of this discussion, Eq. (1.11) is simplified by omitting the Poynting correction, which is usually small at low pressures. Combining Eq. (1.11) with the definition of relative volatility, Eq. (1.2) gives
Was this article helpful?