Supercritical Fluid Science and Technology - American Chemical


Supercritical Fluid Science and Technology - American Chemical...

1 downloads 150 Views 1MB Size

Chapter 1

New Directions in Supercritical Fluid Science and Technology Keith P. Johnston

Downloaded by 80.82.78.170 on December 10, 2016 | http://pubs.acs.org Publication Date: August 29, 1989 | doi: 10.1021/bk-1989-0406.ch001

Department of Chemical Engineering, The University of Texas, Austin, TX 78712

An overview of newresearchdirections is presented. The domain of this field has grown significantly with advances in separations, reactions, and materials processing of complex substances such as polymers, surfactants, and biomolecules. The field has encompassed a large number of areas in engineering and the chemical, physical, and biological sciences, which will be discussed. In the U.S. new commercial processes include coffee decaffeination, hops extraction, catalyst regeneration, extraction of organic wastesfromwater, and supercritical fluid chromatography. These applications complement older technologies such as residuum oil supercritical extraction (ROSE), carbon dioxide enhanced oil recovery, and reaction processes for the production of polyethylene and primary alcohols in supercritical fluid ethylene. The interest in environmental applications is increasing rapidly. Given the experience gained in developing commercial plants in Europe, the U.S., and now Japan and Korea, it would be expected that the time lag between research and commercialization will diminish. Not long ago, research in supercritical fluid (SCF) science and technology emphasized energy savings, synfuels processing, and the thermodynamics of phase equilibria for commodity chemicals. Thefielddealt primarily with extractions of relatively simple molecules that dissolve influidssuch as carbon dioxide. In the renaissance of the last five years, attention has shifted to significantly more complex molecules, undergoing much broader types of physical and chemical transformations. A key innovation has been to explore the effect of a SCF on other phases besides the SCF phase, which often contain higher value substances, such as polymers, biomolecules, or growing crystals. This has extended the scope of the field markedly, as only a limited number of substances are soluble in carbon dioxide. 0097^156/89/O406-0001$06.00A) ο 1989 American Chemical Society

Johnston and Penninger; Supercritical Fluid Science and Technology ACS Symposium Series; American Chemical Society: Washington, DC, 1989.

Downloaded by 80.82.78.170 on December 10, 2016 | http://pubs.acs.org Publication Date: August 29, 1989 | doi: 10.1021/bk-1989-0406.ch001

2

SUPERCRITICAL FLUID SCIENCE A N D T E C H N O L O G Y

A number of examples illustrate the great deal of innovation that has occurred in the lastfiveyears. These include challenging fractionations of foods (such as CD-3 fatty acids) and pharmaceuticals (for example sterols), crystallization (including rapid expansion from supercritical solutions(l,2), retrograde crystallization (2) and gas anti-solvent crystallization(4)), purification of polymers, impregnation of polymers, manipulation of polymer morphology and porosity, manipulation and control of chemical reactions (even in SCF water), extraction and oxidation of hazardous wastes, spectroscopic studies of solvent strength and structure, computer simulation, separations and reactions in reverse micelles, and electrochemical synthesis. This extensive research activity has been complemented by advances in plant design, which are discussed in detail in this symposium. Together, these developments in research and design should lead to some innovative applications, particularly in the food, polymer and pharmaceutical industries, and in environmental protection. For most applications, it is not possible to treat supercritical fluid separations, for example extraction, as a well-defined unit operation as is the case for simpler processes such as distillation. Instead, research is often needed to characterize the important properties for each specific separation or reaction. However, the recent advances in the molecular understanding of SCF solutions provide some general themes that can be utilized in a semi-quantitative manner to evaluate the potential of both research and applications. Characterization of supercriticalfluids:solvent strength and selectivity The solvent strength of a supercritical fluid (compressed gas) may be adjusted continuouslyfromgas-like to liquid-like values, as described qualitatively by the solubility parameter. The solubility parameter, δ (square root of the cohesive energy density) (5), is shown for gaseous, liquid, and SCF carbon dioxide as a function of pressure in Figure 1. It is a thermodynamic property which can be calculated rigorously as 1/2

1/2

(1)

although this point has been missed in many papers. For a given fluid, this plot resembles that of density versus pressure. The δ for gaseous carbon dioxide is essentially zero whereas the value for liquid carbon dioxide is comparable with that of a hydrocarbon. At -30 C, there is a large increase in δ upon condensation from vapor to liquid. Above the critical temperature, it is possible to tune the solubility parameter continuously over a wide range with a small isothermal pressure change or a small isobaric temperature change. This ability to tune the solvent strength of a

Johnston and Penninger; Supercritical Fluid Science and Technology ACS Symposium Series; American Chemical Society: Washington, DC, 1989.

Downloaded by 80.82.78.170 on December 10, 2016 | http://pubs.acs.org Publication Date: August 29, 1989 | doi: 10.1021/bk-1989-0406.ch001

1.

JOHNSTON

New Directions in Supercritical Fluid Science and Technology

10.5 ι

10°

1—ι Mini]

10

1—ι f ι mi|

1

10

1—r ι ι η ι n

2

10

3

ι—ι ι ι ι un

10

4

pressure (bar) β

β

β

Figure 1. Solubility parameter of C 0 (•: -30 C, Q: 31 C, Δ: 70 C). 2

Johnston and Penninger; Supercritical Fluid Science and Technology ACS Symposium Series; American Chemical Society: Washington, DC, 1989.

4

SUPERCRITICAL FLUID SCIENCE A N D T E C H N O L O G Y

supercritical fluid is its unique feature, and it can be used to extract, then recover, products as explained in the book by McHugh and Krukonis which provides an excellent introduction to the field®. In order to understand solubilization of a solid in a SCF, consider the very similar tetracyclic sterols: cholesterol, stigmasterol, and ergosterol, each of which contains a single OH group (see Figure 2). It may seem surprising that the selectivity of carbon dioxide for cholesterol versus ergosterol (ratio of solubilities) is almost two orders of magnitude©. To explain this result, it was necessary to measure their vapor pressures, which are as low as 10' bar (1(H ton), using a gas saturation technique. The selectivity simply follows the ratio of vapor pressures, even though the vapor pressures are negligibly small. To further understand theroleof solute-solvent interactions on solubilities and selectivities, it is instructive to define an enhancement factor as the actual solubility, y2, divided by the solubility in an ideal gas, so that Ε = y2 P/Pz^ This factor is a normalized solubility, because it removes the effect of the vapor pressure, providing a means to focus on interactions in the SCF phase. In carbon dioxide at 35 C and 200 bar, Ε is about 7 for all three of the above sterols. In fact, enhancement factors do not vary much for many types of organic solids. As shown in Figure 3, E's fall within a range of only about 1.5 orders of magnitude for substances with a variety of polar functional groups, even though the actual solubilities (not shown) vary by many orders of magnitude. This means that solubilities, and also selectivities, in carbon dioxide are governed primarily by vapor pressures and only secondarily by solute-solvent interactions in the SCF phase. An exception is strong bases such as ammonia that canreactwith carbon dioxide. SCF carbon dioxide is a lipophilic solvent since the solubility parameter and the dielectric constant are small compared with a number of polar hydrocarbon solvents. Co-solvents(also called entrainers, modifiers, moderators) such as ethanol have been added to fluids such as carbon dioxide to raise the solvent strength while maintaining it's adjustability. Most liquid cosolvents have solubility parameters which are larger than that of carbon dioxide, so that they may be used to increase yields, or to decrease pressure and solvent requirements. A summary of the large increases in solubility that may be obtained with a simple cosolvent is given at the top of Table I. Cosolvents, unlike carbon dioxide, can form electron donoracceptor complexes (for example hydrogen bonds) with certain polar solutes to influence solubilities and selectivities beyond what would be expected based on volatilities alone. Several thermodynamic models have been developed to correlate and in some cases predict effects of cosolvent on solubilities(8,2). They are used extensively in SCFresearchand development.

Downloaded by 80.82.78.170 on December 10, 2016 | http://pubs.acs.org Publication Date: August 29, 1989 | doi: 10.1021/bk-1989-0406.ch001

10

Molecular structure of supercritical fluid solutions The understanding of the molecular interactions and structure in SCF solutions has grown considerably in the last five years, and isreviewedin greater

Johnston and Penninger; Supercritical Fluid Science and Technology ACS Symposium Series; American Chemical Society: Washington, DC, 1989.

Downloaded by 80.82.78.170 on December 10, 2016 | http://pubs.acs.org Publication Date: August 29, 1989 | doi: 10.1021/bk-1989-0406.ch001

1.

JOHNSTON

New Directions in Supercritical Fluid Science and Technology 5

14

16

18

DENSITY

20

22

(MOL/L) β

Figure 2. Solubility of sterols in pure C 0 at 35 C (data are correlated with the HSVDW [7]). 2

7.0

4.0

,

1

16

,

,

18

" 20

1

,

1

22

DENSITY (MOL/L)

Figure 3. Enhancement factor for solids with a variety of polar functionalities in OCX at 35 * C (O: hexamethylbenzene, Δ: 2-naphthol, • : phthalic anhydride, O; anthracene, © : acridine). (Reprinted from ref. 8. Copyright 1987 American Chemical Society.)

Johnston and Penninger; Supercritical Fluid Science and Technology ACS Symposium Series; American Chemical Society: Washington, DC, 1989.

6

SUPERCRITICAL FLUID S C I E N C E A N D T E C H N O L O G Y

Downloaded by 80.82.78.170 on December 10, 2016 | http://pubs.acs.org Publication Date: August 29, 1989 | doi: 10.1021/bk-1989-0406.ch001

Table I. Effects of cosolvents including simple liquids, complexing agents and surfactants on solubilities in SCFs at 35 C Solute

co-solvent

acridine

3.5% CH OH

2.3

2-aminobenzoic acid

3.5%

CH3OH

7.2

cholesterol hydroquinone

3

9% C H 3 O H

2% tributylphosphate

ybinary

100

>300

hydroquinone

.65%AOT,W =10* 6% octanol

>200

tryptophan

.53% AOT, W = 10* 5% octanol

»100

o

0

SOURCE : Data are from references 7 and 8. * W is the water-to-surfactant ratio. Q

detail in a following chapter(2). In the highly compressibleregion,a SCF solvent condenses about a solute (which is usually much larger and more polarizable) so that the solute's partial molar volume can be on the order of -10,000 cc/mole. Thermodynamically, this explains a key unique feature of SCF solutions, that is the large pressure effects on phase behavior and chemicalreactionrates. For a solvent with a molar volume of 100 cc/mole, a V2 of -10,000 cc/mole indicates directly that an average of 100 solvent molecules condense about each solute. This solvent condensation or clustering may be described by a simple equation in terms of the isothermal compressibility of the solvent using Kirkwood-Buff solution theory(2). It is quite satisfying that the conclusions of a number of studies of clustering are consistent, given the variety in the methodologies: partial molar volume data, Kirkwood-Buff theory, UV-visible and fluorescence(spectral shift) data, integral equations, and computer simulation data. Another type of clustering

Johnston and Penninger; Supercritical Fluid Science and Technology ACS Symposium Series; American Chemical Society: Washington, DC, 1989.

1.

JOHNSTON

New Directions in Supercritical Fluid Science and Technology 7

in SCF solutions, solute-solute clustering, has been discovered recently using fluorescence studies(lO) and integral equations(ll).

Downloaded by 80.82.78.170 on December 10, 2016 | http://pubs.acs.org Publication Date: August 29, 1989 | doi: 10.1021/bk-1989-0406.ch001

Reactions in and with swpwritical flwds Many of the desirable properties of supercritical fluids which are used in separation processes can be exploited in systems undergoing chemical reaction. Less than 5% of the SCF technology literature deals with chemical reactions, although this relatively new area is growing considerably because of applications such as oxidation of hazardous wastes in supercritical water. A number of interesting opportunities for performing reactions in supercritical fluid solvents will be discussed: 1. The solvent strength of a SCF may be manipulated using pressure and/or temperature to adjustreactionrates by changing rate and equilibrium constants, or concentrations of reactants and products. The latter is due to the large changes in concentrations that occur in the critical region. A small change in the pressure of a SCF can produce a large change in the solvent strength (as measured by solubility parameter or solvatochromic polarity scales(2)), which can cause a large thermodynamic solvent effect on a rate or equilibrium constant. This phenomenon is unique to SCFs. An extremely pronounced pressure effect was discovered for the rate constant of the unimolecular decomposition of α-chlorobenzyl methyl ether in supercritical 1,1 -difluoroethane(12) C6H5CTO-O-CH3 ~> C6H5CHO + CH3CI

(2)

A change in pressure of only 15 bar increased the rate constant by an order of magnitude, because the density and thus the solvent strength increased significantly. This solvent effect can be explained in an alternative but more complex manner by using transition state theory. For a unimolecularreaction,A = A* —» products, the activation volume may be expressed as •RTOlnk^P^r^VA^-VA

(3)

where k is the rate constant based on mole fraction units. In conventional liquid solvents activation volumes are up to ± 30 cc/mol. In a highly compressible SCF, vis of solutes can reach thousands of cc/mol negative. As aresult,activation volumes, which are differences in VJS, can also be pronounced. For the above ether decompositionreaction,activation volumes were observed as low as -6000 cc/mol, demonstrating the capability of SCFs to adjust rate constants. Large pressure effects were also found for the Diels-Alder reaction of isoprene and maleic anhvdride(13). the tautomeric equilibria of 2-hydroxypyridine x

Johnston and Penninger; Supercritical Fluid Science and Technology ACS Symposium Series; American Chemical Society: Washington, DC, 1989.

8

SUPERCRITICAL FLUID SCIENCE A N D T E C H N O L O G Y

and 2-pyridone at infinite dilution in propane and 1,1 -difluoroethane( 14). for the reversible redox reaction of the hfl- couple in supercritical waterQS). and for other examples in the chapter on reactions. Although these studies have made considerable progress in understanding these pressure effects in terms of solutesolvent interactions and solvent compressibility, further work is needed in this new area. 2. Selectivities to certain products may likewise be adjusted using pressure or temperature. In SCF water, reaction chemistry is governed by free-radical (homolytic) mechanisms for an ion product of water, K , less than 10" , and by ionic (heterolytic) mechanisms for larger values of K (16). This describes the observation that decomposition reactions shiftfrompyrolysis to hydrolysis as the density of water and thus K is increasedQl). In the near-critical region, K can be manipulated by small changes in temperature and pressure to control the reaction selectivity. In the photochemical dimerization of isophorone in SCF fluoroform and carbon dioxide, it is possible to study pressure effects on regioselectivity as well as stereoselectivity by observing the concentrations of the three primary dimersQfi). Here regioselectivity refers to head-to-head versus head-to-tail oxygen configurations for the dimers, while stereoselectivityrefersto syn versus anti configuration. The dipole moments of the regio-isomers are very different, while those of the stereoisomers are about the same. As aresult,the regioselectivity is influenced by both solvent polarity (solute-solvent attractive interactions), and solventreorganization(solute-solventrepulsiveinteractions), while the stereoselectivity is influenced only by solventreorganization.Studies of additional reactions would help elucidate the emerging understanding of pressure effects on selectivity. 3. Reactions may be integrated with SCF separation processes to achieve a large degree of control for producing a highly purified product Reaction products could be recovered by volatilization into, or precipitation from, a SCF phase. The classic example is the high pressure production of polyethylene in the reacting solvent SCF ethylene(®. The molecular weight distribution may be controlled by choosing the temperature and pressure for precipitating the polymerfromthe SCF phase. 4. The use of a supercritical fluid may provide a means to perform a single phase homogeneousreaction,instead of a multiphasereaction(12).For example, organics may be oxidized with oxygen in SCF water, in the one phaseregionwith minimal mass transfer limitations. This improves both energy and destruction efficiencies. 5. Transport properties are improved in that diffusivities are higher and viscosities are lower than in liquids. 6. SCFs offer several benefits in heterogeneous catalysis. They can act as a vehicle to improve transport in catalysts byreducingcapillary condensation and pore pluggingQQ). SCFs have been used to accelerate gas-solid reactions by 14

w

Downloaded by 80.82.78.170 on December 10, 2016 | http://pubs.acs.org Publication Date: August 29, 1989 | doi: 10.1021/bk-1989-0406.ch001

w

w

Johnston and Penninger; Supercritical Fluid Science and Technology ACS Symposium Series; American Chemical Society: Washington, DC, 1989.

w

Downloaded by 80.82.78.170 on December 10, 2016 | http://pubs.acs.org Publication Date: August 29, 1989 | doi: 10.1021/bk-1989-0406.ch001

1.

JOHNSTON

New Directions in Supercritical Fluid Science and Technology

removing reactants or products from a solid phase(21). In the chlorination of alumina in SCF CCI4, the reaction accelerates as density increases due to dissolution of the product AICI3 from the reactive surface. SCFs offer a nonaqueous environment which can be desirable for enzymatic catalysis of lipophilic substrates. Hie lipophilic substance cholesterol is 2 to 3 orders of magnitude more soluble in CC>2-cosolvent blends than in water©. In CO2 based blends, it may be oxidized to cholest-4en-3one, a precursor for pharmaceutical production using an immobilized enzyme(22). The enzyme polyphenol oxidase has been found to be catalytically active in supercritical CO2 and fluoroform (23). The purpose of using a SCF is that it is miscible with one of the reactants-oxygen. Lipase may be used to catalyze the hydrolysis and interesterification of triglycerides in supercritical CO2 without severe loss of activity(24). These reactions could be integrated with SCF separations for product recovery. 7. SCFs may be used to perturbreactionsgently in order to study reaction mechanisms and solvent effects, simply by a modest change in the pressure. Here the SCF would be used as a tool to obtain information about areaction,whereas, it is used to improve the performance of a reaction in each of the above cases. The perturbation could be due to changes in solvent polarity or viscosity(25). The advantage of using a SCF is that the thereactionmay be perturbed over a continuum with a single solvent instead of with a series of liquid solvents with differing molecular functionality. Complex molecules in supercritical fluid science and technology: polymers. surfactants and biomolecules SCF technology has spread quicklyfrommolecules such as naphthalene to more complex substances such as polymers, biomolecules, and surfactants. Supercritical fluids can be used toreducethe lower critical solution temperature of polymer solutions in order to remove polymers from liquid solvents(6.26). The technology has been extended to induce crystallization of other substances besides polymersfromliquids, and has been named gas recrystallization(4). In other important applications, SCF carbon dioxide has been used to accomplish challengingfractionationsof poly(ethylene glycols) selectively based on molecular weight as discussed in this symposium, and of other polymers(â). A new use for supercritical fluids is to swell glassy polymers so that they may be impregnated rapidly with additives such as plasticizers, pharmaceuticals, pigments, and other additives(27). The inverse of this concept is also important, that is SCFs may be used to purify polymers by extractingresidualsolvent and monomers (©. Both the impregnation and purification processes take advantage of enhanced solute diffusion by up to 9 orders of magnitude, whichresultsfrom plasticizing a glassy polymer with a SCF(27). This increase is due to the increased freedom of motion of the swollen polymer chains. The swelling of polymers using

Johnston and Penninger; Supercritical Fluid Science and Technology ACS Symposium Series; American Chemical Society: Washington, DC, 1989.

Downloaded by 80.82.78.170 on December 10, 2016 | http://pubs.acs.org Publication Date: August 29, 1989 | doi: 10.1021/bk-1989-0406.ch001

10

SUPERCRITICAL FLUID SCIENCE AND T E C H N O L O G Y

pure SCFs has been measured onlyrecently(28.29.30):this information is useful for the extension to multicomponent systems, which contain solutes in addition to the polymer and SCF solvent. A final application is the manipulation of polymer morphology or porosity by tailoring the depressurization of a SCF. An important factor that dictates the feasibility of both polymer impregnation and purification processes is the equilibrium distribution coefficient of the solute between the polymer phase and the SCF phase. The distribution coefficient of toluene between cross-linked silicone rubber and carbon dioxide has been explored quantitatively as a function of temperature over a wide range in pressure up to 250 barQl). It is adjustable over a continuum in the highly compressible region of carbon dioxide as shown in Figure 4. The behavior was predicted quantitatively using only information from binary systems. This type of thermodynamic understanding of phase behavior in ternary systems containing polymers and SCFs is rare, but very beneficial for guiding practical applications. Surfactants are another example of complex molecules that have interesting properties in systems containing SCFs. Carbon dioxide, even when doped with a co-solvent such as ethanol, is incapable of dissolving a measurable amount of hydrophilic substances such as proteins. The same is true for other SCFs with critical temperatures under 100 C. However, SCFs can solubilize hydrophilic molecules by forming complexes, aggregates, reverse micelles, or microemulsions. An example is given in Table I where an amino acid, which is very insoluble in ethane, can be solubilized with an anionic surfactant, sodium di-2-ethylhexyl sulfosuccinate (AOT). In other breakthroughs described in this symposium, viscous gels have been formed in SCF solutions, and emulsion polymerization has been carried out inreversemicelles. The recent advances involving polymers, surfactants, and biomolecules open up large new domains for research in SCF science and technology.

û

100

200

300

P R E S S U R E , bar

Figure 4. Large change in the distribution coefficient of toluene between silicone rubber and C 0 based on volume fraction in the critical region (—: predicted value). 2

Johnston and Penninger; Supercritical Fluid Science and Technology ACS Symposium Series; American Chemical Society: Washington, DC, 1989.

1.

JOHNSTON

New Directions in Supercritical Fluid Science and Technology

Acknowledgment Acknowledgement is made to the donors of the Petroleum Research Fund, administered by the American Chemical Society, the Camille and Henry Dreyfus Foundation for a Teacher-Scholar Grant, and the Separations Research Program at the University of Texas.

Downloaded by 80.82.78.170 on December 10, 2016 | http://pubs.acs.org Publication Date: August 29, 1989 | doi: 10.1021/bk-1989-0406.ch001

Literature Cited 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.

Krukonis, V. 1984, AIChE Annual Meeting, San Francisco Mohamed R.S.; Halverson, D.S.; Debenedetti, P.G. this symposium. Chimowitz, E.H.; Pennisi, K.J. AIChE J., 1986,32,1665. Gallagher, P.M.; Coffey. M.P.; Krukonis, V.J.; Klasutis, N. this symposium. Barton, A. Handbook of Solubility Parameters and Other Cohesive Parameters, CRC: Boca Raton, FL, 1983. McHugh, Μ. Α.; Krukonis, V. J. Supercritical Fluid Extraction: Principles and Practice, Butterworths; Boston, Mass, 1986. Wong, J. M.; Johnston, K. P. Biotech. Progress 1986, 2, 29. Dobbs, J.M.; Johnston, K.P. Ind. Engr. Chem. Res. 1987, 26, 1476. Johnston, K.P.; Kim, S.; Combes, J., this symposium. Brennecke, J.F.; Eckert, C.A.; this symposium. Cochran, H.D.; Lee, L. L.; this symposium. Johnston, K. P.; Haynes, C. AIChE J., 1987,33,2017. Paulaitis, M.E.; Alexander, G. C. Pure & Appl. Chem. 1987, 59, 61. Peck, D.G.; Mehta A. J.; Johnston K. P. J. Phys. Chem. 1989, in press. Flarsheim, W.M.; Bard, A.J.; Johnston, K.P. J. Phys. Chem. 1989, in press. Antal M . J.; Brittain, Α.; DeAlmeida C.; Ramayya, S.; Roy J.C. ACS Symp. Series 329, 1986, 77. Townsend, S.H.; Abraham, M.A.; Huppert, G.L.; Klein, M.T.; Paspek, S.C. Ind. Eng. Chem. Res. 1988, 27, 143. Hrnjez, B.J.; Mehta, A.J; Fox, M.A.; Johnston, K.P. J. Am. Chem. Soc. 1989, 111 2662. Subramaniam B.; McHugh M . A. Ind. Eng. Chem. Process Des. Dev. 1986, 25, 1. Barton, P.; Kasun, T.J. In Supercritical Fluid Technology, Penninger, J. et al. Ed.; Elsevier, Amsterdam, 1985, 435. Herrick, D.E.; Holder, G.D.; Shah, Y.T. AIChE. J. 1988,34,669. Randolph, T.W.; Blanch, H.W.; Prausnitz, J.M. AIChE J. 1988,34,1354. Hammond, D.A.; Karel, M.; Klibanov, A.M. Applied Biochemistry and Biotechnology 1985, 11, 393 Nakamura, K.; Chi, Y.M.; Yamada, Y.; Yano, T. Chem. Eng. Comm. 1986, 45, 207.

Johnston and Penninger; Supercritical Fluid Science and Technology ACS Symposium Series; American Chemical Society: Washington, DC, 1989.

12

SUPERCRITICAL FLUID SCIENCE AND TECHNOLOGY

25. 26. 27. 28. 29.

Aida T.; Squires T. G. ACS Symp. Series 329; 1987, 58. McHugh, M.A.; Guckes, T.L. Macromolecules 1985,18,674. Berens, A. R.; Huvard, G. S.; Korsmeyer, R.W. this symposium. Fleming, G. K.; Koros W.J. Macromolecules 1986, 19, 2285. Liau, I. S.; McHugh, M.A. Process Technology Proceedings 3 (Supercritical Fluid Technology). J. M. L. Penninger, et al.,Ed.; Elsevier, Amsterdam, 1985, 415. 30. Wissinger, R. G.; Paulaitis, M.E. J. Polym. Sci., Part B: Polym. Phys. 1987, 25, 2497. 31. Shim, J.J.; Johnston, K.P. AIChE J. 1989, in press.

Downloaded by 80.82.78.170 on December 10, 2016 | http://pubs.acs.org Publication Date: August 29, 1989 | doi: 10.1021/bk-1989-0406.ch001

RECEIVED May 1, 1989

Johnston and Penninger; Supercritical Fluid Science and Technology ACS Symposium Series; American Chemical Society: Washington, DC, 1989.