The heat duties associated with these fractions can be estimated by the Clausius-Clapeyron equation (20)

For R, for example, vapor pressure is 5, 10, and 20 mmHg at temperatures of 411.2, 425.3, and 441.2°

1.987 Btu


20 _ AJfvap 10 1.987

(425.3 X 1.8) (441.3 x 1.8) 0.6931 x 1.987 = AHmp (0.00004733) AHva? = 29,100 btu/lb-mole AHvap ave = 29,900 Btu/lb-mole

Similarly for P, and W

Affvapav, = 25,540 Btu/lb-mole

AHV8P = 18,180 Btu/lb-mole

Total heat required {Note: For distillation duty only)

•The relative volatility of W with respect to P is so large that it can be considered to be capable of being separated without reflux.


With these calculations and the time allowable for the distillation, the still system can be specified. A batch must be distilled every day to meet plant capacity requirements, so a reasonable schedule might be

Charge still 0.5-1 h Empty still 0.5-2 h Heatup 2 h

Distillation 18 h Total say 22 h

Therefore, distillation heat duty = 7,930,000/22 = 360,000 Btu/h. Of course, the condenser must be capable of removing this amount of heat.

Because a still pot for such a system should only be filled to about 70 percent of its capacity, the still volume should be

The 18-h distillation time means that the rectifying column must be designed for a vapor loading of

The most drastic condition which will exist in the column will occur during the foreruns cut and during the finishing of the product cut when reflux ratio will be 5:1. At these points, the L/V ratio will be at its maximum value, and the column must be capable of adequately handling such a loading (see Chap. 12).

The still system will consist of a receiver for W, a receiver for P, a foreruns-tailing receiver, and a product receiver. The method of operation of the system will be:

1. Charge system

2. Start heat to still

3. Start vacuum system

4. Divert product to W receiver

-5. Observe temperature profiles in the column to change delivery from a. W receiver to P receiver (increase reflux ratio)

b. P receiver to foreruns-tailings receiver (increase reflux ratio)

6. Based on product appearance and freezing point, divert to product receiver (decrease reflux ratio)

7. Based on product appearance (mainly) and freezing point, divert to foreruns-tailings receiver

8. Cut off heat, break vacuum.

The discussion above leaves many things unsaid. It would be extremely unlikely that a competent research department that had "Thoroughly investigated the manufacture of R" would not have developed better information about C than is shown. Of course, better information, and consequent better understanding, would make the process engineer's task easier, but the real point is to show how some otherwise-clumsy materials can be accounted for.

The suggestion to start the heating system and vacuum-producing system and to begin removing W immediately is a departure from the conventional idea of operating a batch system at total reflux "to equilibrate" it before starting product removal. If the first material to come out of the distillation system is a valuable product, the system should be equilibrated. In a case like the one discussed above, where W, P, and foreruns are distilled off before the product cut is started, equilibration is unnecessary. After all, batch distillation is always a transient process, so why worry about achieving equilibrium before starting a strictly nonequilibrium process?

"Observing temperature profiles" in the column is a clear recommendation of installing thermocouples in at least the bottom, middle, and top of the column. Further, use of the vapor pressure regressions and a little imagination permits the construction of graphs which allow the operators to infer product composition from temperature and pressure readings.

The material collected in the foreruns-tailings receiver would normally be recharged to the still as part of the next cycle. One should not be surprised to observe that the amount of foreruns and tailings separated land returned) tends to become a fairly stable quantity, even though "logic" would dictate otherwise.

Product quality was easy to assess in the example inasmuch as color is easy to detect and freezing points are easy to measure—even by relatively unskilled operators. For operation with other systems, product ^nality needs to be easily assessable to aid operator decisions—finding the exact analytical tool required for such assessment might not be so

Neither trays nor packing have been discussed in any of the above. If trays are used, tray efficiency must be considered, and if packing is specified, the HETP of the packing must be known. The vendors of column internals can be most helpful here, but also see Chaps. 10 and 12.

"Safety factors" have not been discussed inasmuch as these factors should reflect the engineer's confidence in the data he or she has, the distillation system performance deduced in manipulating these data, the reliability the equipment must demonstrate, etc., in short, "engineering judgment." There is never anything better than reliable experimental data on the system which is to be distilled, of course. In other, less-well-defined circumstances, conservatism is advised (9). Batch stills are usually (at least relatively) low-cost items, and increased capabilities represent small incremental costs.

5.3 Special Note to Readers

As stated earlier, the development above is based on a "real-life" design, but does not reflect that design identically. The crude actually had a nondistillable residue associated with it, and it was necessary to remove this material from the still after each distillation. However, the residue had no other effect on the distillation and, for this reason, was not included in the discussion above. The reflux ratio actually used between foreruns and product cuts was 10:1, and for reasons that are now nonretrievable, the column contained about 15 feet of Intalox8 saddles—equivalent to five to eight theoretical stages. The operating procedures which evolved with practice were those recommended above, and the quantities of foreruns and tailings recovered and recycled leveled out fairly well. The only reaily serious problem not anticipated in the design was the gradual increase in viscosity in the still residue to the point where it was no longer transferable by the centrifugal pump provided. Substitution of a positive-displacement pump remedied this problem.

5.4 References

1. Barb, D. K„ and C. D. Holland, 7th World Congress, Mexico City, April 1967.

3. Block, B., in Perry, R. H„ and C. H. Chilton (eds.), Chemical Engineers'Handbook, 5th ed„ McGraw-Hill, New York, 1973.

4. Bogart, M. J. P., Trans. AIChE, 33, pp. 139-151, 1937.

5. Boston, J. F., in R. S. Mah, and W. D. Seider (eds.), Foundations of Computer-Aided Chemical Process Design, Vol. 2, American Institute of Chemical Engineers, New York, 1981.

6. Coates, J, A., and B. S. Pressburg, Chem. Eng., 68(2), pp. 131-136, 1961.

7. Distefano, G. P., Am. Inst. Chem. Engrs. J., J-i(l), pp. 190-199, 1968.

8. Ellerbee, R. W., Chem. Eng., «0(12), pp. 110-116, 1973.

9. Ellerbee, R. W., in P. A. Schweitzer (ed.), Handbook of Separation Techniques for Chemical Engineers, McGraw-Hill, New York, 1979.

10. Frank, O., Chem. Eng., 84(6), pp. 110-128, 1977.

11. Houtman, J. P. W., and A. Husain, Chem. Eng. Sei., 5, pp. 178-187, 1956.

12. Huckaba, C. E„ and D. E. Daniy, Am. Inst. Chem. Eng. J., 6, pp. 335-342, 1960.

13. Meadows, E. L„ Chem. Eng. Prog. Sym. Ser. 46, 59(46), pp. 48-55, 1963.

14. Pigford, R. L., J. B. Tepe, and C. J. Garrahan, bid. Eng. Chem., 43, pp. 2592-2602, 1951.

15. Rayleigh, Lord, Phil. Mag., 6th series, 4, pp. 521-537, 1902.

16. Robinson, C. S., and E. H. Gilliland, Elements of Fractional Distillation, 4th ed., McGraw-Hill, New York, 1950.

19. Rose, L. M., Distillation Design in Practice, Elsevier, Amsterdam, 1985.

20. Smith, J. M., and H, C. Van Ness, Introduction to Chemical Engineering Thermodynamics, 4th ed., McGraw-Hill, New York, 1987.

21. Smoker, E. H., and A. Rose, Trans. Am. Inst, Chem. Engrs., 36(2), pp. 285-293, 1940.

Chapter 6

Was this article helpful?

0 0

Post a comment