John M. Prausnitz - US grants

This project is concerned with two areas of importance in the development of bioprocesses for the production of high and intermediate-value biochemicals. The first concerns the use of aqueous two-phase polymer systems for the recovery of fermentation products. The research is directed at obtaining the molecular thermodynamic properties of these solutions and developing predictive models for phase behavior and partitioning of solute molecules between phases. The second area concerns the use of non-aqueous solvents for enhancing substrate solubility in enzymatically catalyzed reactions. Supercritical carbon dioxide, alone with selected entrainers, is used as a solvent for studying the kinetics of several commercially important reactions: cholesterol oxidation, interesterification of lipids, and various hydrolytic reactions which can be run "backwards".

This research concerns phase-equilibrium properites of natural-product mixtures (such as vegetable oils and other agricultural products) as required for process design to manufacture-refined foods (e.g., cocoa butter), speciality chemicals, and pharmaceuticals. To study these properties, it is necessary to separate the agricultural product into fractions according to molecular type and size. A preparative liquid chromatograph is required to obtain such fractions in sufficient quantities for subsequent experimental studies using liquid chromatography and other analytical instruments for measuring thermodynamnic phase equilibrium properties. These properties are needed for process development in a variety of natural-products industries, including biotechnology process innovation.

This renewal project is concerned with two areas of importance for increasing the impact of biological processes on production of specialty chemicals: use of non-aqueous solvents for enzyme catalyzed reactions and development of novel separations processes based on aqueous two-phase systems. The first aspect will focus on enzyme reactions in supercritical solvents: In the previous grant period, the stability of enzymes, both free and immobilized in dense carbon dioxide, has been demonstrated, and kinetic data have been obtained for the oxidation of cholesterol and the interesterification of triacylglycerides. This research expands the possible range of substrate concentrations which can be used with supercritical carbon dioxide by exploring its use with entrainers; it also provides a more detailed study of the kinetic and mass transfer effects of importance for both enzymes studied. In addition, the conformation of enzymes in this non-aqueous medium is examined by the use of EPR spectroscopy. A nitroxide spin-labelled cholesterol probe is employed to monitor the spatial coordinates at the active site of cholesterol oxidase. The second aspect will focus on aqueous two-phase systems for product recovery: This research is directed toward establishing the fundamental molecular thermodynamics of aqueous two-phase systems. An important objective is to determine what polymer-polymer systems may be best suited for bioprocessing. Toward that end, the PIs continue their measurements using light scattering and differential vapor pressure to obtain thermodynamic parameters for aqueous mixtures containing dextran, polyethyleneglycol, proteins and combinations of polymer-polymer and polymer protein pairs. These data, combined with information on electrical-potential effects due to salt partitioning and a molecular-thermodynamic model based upon the osmotic virial equation, allow the PI's to predict polymer-polymer-water phase diagrams and protein partition coefficients for process design and optimization.

This proposal requests funding for the purchase of a low-angle laser light-scattering (LALLS) photometer and a high performance liquid and size exclusion chromatography (HPLC-SEC) system. This equipment is to be used to determine molecular-thermodynamic parameters and experimental ternary polymer-polymer-water phase diagrams required to develop a predictive model for a promising new separation technique in biotechnology, i.e., two-phase aqueous polymer-polymer systems for selective extraction of biomolecules. The PIs are considered to be very well qualified to effectively utilize the proposed equipment. Funding of this proposal is being recommended and the costs are to be shared by the Interfacial, Transport and Thermodynamic Processes and the Biotechnology Programs of the Engineering Directorate within NSF, and the University of California at Berkeley. Biotechnology Program $ 50,932 Interfacial, Transport and $ 30,000 Thermodynamic Processes Program University of California at $ 40,466 Berkeley

The wider use of enzymes in organic syntheses, where their high degree of regio- and stereo-selectivity is difficult to achieve with conventional catalysts, is severely hampered by the low aqueous solubility of many substrates of interest. The encapsulation of enzymes in reverse micelles, which provides a method for stabilization of the enzyme, is a new and particularly promising approach for biological synthesis or transformation of organic compounds in organic solvents. This project concerns fundamental studies for the development of new bioreactors employing such encapsulated enzymes. It includes a study of the effect of encapsulation of dopamine beta- monooxygenase, a non-specific enzyme which hydroxylates ring- substituted phenethylamines, alkenes, aldehydes, amides and sulphides. The effect of micelle structure on the conformation of the enzyme is studied by use of ESR and kinetic measurements. A molecular- thermodynamic model, relating the properties of the aqueous miniphase to prevailing conditions, is also developed. The use of reverse micelles to protect the enzyme from the often denaturing influence of organic solvents is a new approach to enzyme stabilization. Current research has focussed on the use of reverse micelles for protein purification. The inner water pool of a reverse micelle can solubilize large protein molecules and the micelle provides a large surface area for transport of reactant and substrate to and from the enzyme. Little is understood of the behavior of enzymes within such micelles or of how their kinetic properties differ from enzymes in solution. The investigators create micelles using non-polar surfactants, so that transport of polar organic substrates can be enhanced by the addition of carrier molecules in the organic phase, such as quaternary alkyl amines. There is current interest in describing the behavior of water in reverse micelles from a phase equilibrium thermodynamic viewpoint. Of even greater importance is the development of a molecular-thermodynamic model to relate the properties of the aqueous miniphase to the external conditions, so that a predictive model can be employed for bioreactor design.

This grant is for purchase of a supercritical fluid chromatograph (SFC). It will provide versatile analytical capabilities for an ongoing NSF supported projects on novel bioreactor configurations and on development of novel enzyme- catalyzed reactions to produce new materials which cannot be manufactured by conventional chemical methods. SFC offers significant advantages over other methods for chemical analysis such as gas-liquid chromatography and high- performance liquid chromatography. SFC will enable the investigators to obtain essential quantitative data efficiently and rapidly and will also enable them to make quantitative measurements at very low concentrations that cannot be made with conventional equipment.

The proposed research is concerned with separating a mixture of proteins in aqueous solution through selective precipitation, i. e., by formation of a phase rich in the target protein in equilibrium with a second phase where the concentration of the target protein is very low. Precipitation is achieved by salts or nonionic polymers or both. To obtain a quantitative understanding of protein- precipitation equilibria, a molecular-thermodynamic model based on an extended Guggenheim/McMillan-Mayer osmotic viral expansion is presented. The goal of this research is to establish an engineering-oriented correlation for rational design of protein-precipitation processes. A long-range benefit of this research is that it is likely to provide useful knowledge for better understanding of protein crystallization.

Solvent-induced crystallization is a novel method for separating an inorganic solute from a concentrated aqueous solution which can provide an attractive low-energy alternative to conventional crystallization methods. Precipitation is initiated when an aqueous phase is extracted into the organic phase or the solvent dissolves in the aqueous phase. The resultant mother liquor, containing the solvent, water, and, residual solute, is then regenerated by temperature adjustment into two phase, a water phase for recycle, and a solvent phase. The latter is further dried by re-equilibrating with a saturated aqueous phase thereby completing the regeneration cycle. In addition, some solutes separate by virtue of their differing solubilities into immiscible phases which can be exploited to drive certain metathesis chemistries when one of the products is removed by extraction via the crystallization step. This work proposes to exploit these techniques by obtaining new experimental data for a variety of representative salt/solvent/water systems and to develop a molecular- thermodynamic model to correlate the data both new and previously published. The results of this work will provide the framework for a family of economically attractive, industrial-scale separation processes.

Hydrogels have significant medical and pharmaceutical applications, particularly for contact-lens materials, drug-delivery vehicles, and artificial organs. This work is concerned with establishing a quantitative understanding of the properties of hydrogels in water which may contain biologically important solutes. Emphasis is directed at fundamental experimental and theoretical studies of hydrogel swelling and solute- partitioning behavior for guiding the design of novel materials that may find application in medicine and pharmacy. Novel hydrogels will be synthesized by copolymerizing 2-hydroxyethyl methacrylate (HEMA) with selected specialty comonomers; the added comonomers have been chosen to impart specific swelling and solute- partitioning behavior to the hydrogels. Swelling properties of these hydrogels will be measured as a function of the hydrogel structure (cross- link density, monomer concentration, comonomer concentration) and solution conditions (temperature, ionic strength, pH, solute concentration). Solute partitioning will be measured for a series of solutes at varying hydrogel and solution conditions. Model solutes have been chosen to cover a range of molecular weight and chemical constitution. Attention will be given to model solutes whose partitioning behavior is of interest in biomedical applications of hydrogels. A molecular-thermodynamic model will be established for relating hydrogel swelling and solute-partitioning behavior to hydrogel, solute, and solution properties. Preliminary results suggest that conventional models for hydrogel elasticity cannot describe swelling behavior as a function of monomer concentration. Mechanical measurements of hydrogel-network elasticity will be performed to provide data that will aid in applying new elasticity theories to hydrogel systems. To understand how microstructure affects hydrogel performance in solution, transmission electron microscopy will be used to observe hydrogel microstructure as a function of hydrogel chemistry and composition. Structure-property relationships inferred from microstructure observations will aid in model development. This research will provide fundamental physico-chemical information of the properties of hydrogels, and on the interactions of these hydrogels with aqueous solutions of medically important solutes. This information will aid in the design and development of novel hydrogel materials for applications in medicine and pharmacy.

Experimental and theoretical bio-processing studies are directed toward the development of a molecular-thermodyamic model for separationg a target protein from a mixture of proteins in aqueuous solution. Protein separation is achieved by precipitation induced by salts, nonionic polymers, or both. The resulting system consists of two liquid phases in equilbrium, with one enriched in the target protein. A molecular-thermosynamic model is outlined for describing protein-precipitation phase equilibria. This model is derived from an understanding of the intermolecular forces in the precipitating solutions; it uses the mean-spherical- approximation to define the reference system as a mixture of charged hard spheres. Pertubations to this reference system result from the dispersion forces, dipole-dipole forces and perhaps, specific association forces. Osmotic pressures calculated from the model will be compared to osmotic-pressure data obtained from membrane osmometry and to osmotic second virial coefficients measured by low-angle laser-light scattering. Experimental phase-equilibrium (precipitation) measurements will provide data against which the molecular- thermodynamic model can be compared. Protein aggregate size distribution data from light scattering will be used to augment the theory when significant aggregation occurs. The goal of this reech is to establish an engineering-oriented correlation for the rational design of protein-precipitation processes. The long-range benefit of this research is that it is likely to provide useful knowledge for a better understanding of protein crystallization bioprocesses.

John Prausnitz CTS-9411366 Thermodynamic properties of polymer/mixed-solvent systems are necessary for a variety of applications including surface acoustic-wave vapor sensors; recovery of organic vapors from waste-air streams using a polymeric membrane; pervaporation, and other polymeric membrane-separation processes; polymer devolatilization; vapor-phase photografting; and for optimum formulation of paints and coatings. For rational process and product design, we require new experimental data and new correlation based on molecular thermodynamics. Vapor-liquid equilibria for several polymer/mixed-solvent systems will be measured , both for dilute polymer solutions (using a sampled equilibrium cell ) and concentrated solutions (using a gravimetric polymer swelling technique). Also, the theta temperature for various polymer/mixed-solvent systems will be measured and correlated with a procedure that involves measuring (by scanning) the cloud-point temperatures for a range of polymer concentrations and extrapolating to unity volume fraction polymer. The goal of the study is to provide experimental data and correlations for thermodynamic properties of these systems for a wide range of applications in acoustic-wave vapor sensors, VOC membrane separation from air streams, vapor-phase photo-grafting, and paint formulation. The successful completion of the project would significantly enhance the knowledge and technology base in these areas, enabling better design of existing processes and development of new processes and products. The experimental techniques to be used are modifications to existing techniques developed by the P.I.'s and others, to account for mixed solvents. The correlational work will be with the P.I. 's previously developed Perturbed Hard-Sphere-Chain equation of state. The theta temperature is the spinodal temperature in the limit of infinite polymer chain length and infinite dilution of the polymer in the solvent. A program for this calculation has been develop ed by the P.I. and co-workers. For the experiments, a tentative list of systems is given, and these are keyed to the specific technological application. A typical error analysis for the gravimetric technique is given.

9530793 Blanch Recovery and purification of proteins is of vital importance in biotechnology. Design and optimization of separation and purification processes requires an understanding of the behavior of proteins in complex aqueous solutions, where salts, polymers or other solutes may be dilute or concentrated. The objective of this proposal is to develop a separation processes. We are concerned in particular with protein precipitation by salts and with protein crystallization; however, protein solution behavior is also important in stability and formulation of therapeutic proteins, and in understanding intermolecular reactions. This proposal is for continued support of a two-year NSF project OThermodynamics of Protein Precipitation in Aqueous SolutionsO. The approach is to describe protein solution behavior on a molecular level, in terms of a two-body potential of mean force (PMF), which describes the overall interactions between two protein molecules. The PMF is determined by the physicochemical properties of the protein, the nature of the solution and temperature. From the PMF, thermodynamic solution properties may be obtained for predicting protein-precipitation phase-equilibrium properties. It is planned to employ liquid-state integral-equation theory to develop new expressions for the compressibility and Helmholtz energy of protein solutions; from these expressions phase equilibria are obtained. This modeling approach is coupled with an experimental program to determine the phase equilibria of a number of model proteins with particular attention given to selective precipitation of a target protein from a mixture. In addition, specific protein interactions will be experimentally quantified. Low-angle laser-light scattering and membrane osmometry will provide information on intermolecular forces and aggregation. Specific protein-protein interactions will be examined by dynamic light scattering. Electrostatic and hydrophobic contributions to the PMF will be probed by sy stematically changing amino-acid residues on the model proteins. Specific ion-protein interactions will be quantified using differential refractometry Cl NMR and protein titrations. *** A further component of the proposed research is to examine solution conditions which favor protein crystallization. The hypothesis that a crystallizing solvent results in weak attractive interactions between proteins, while strongly attractive conditions favor amorphous precipitation, will be examined by determining the osmotic second virial coefficients for a number of proteins in crystallizing solvents. It is planned to investigate critically an idea supported by preliminary data, viz. that crystallization is most likely when the osmotic second virial coefficient of a protein is slightly, but not strongly, negative.

ABSTRACT
CTS-9911744
J. Prausnitz
U. of California @ Berkeley

The natural-gas industry in the US is a multibillion-dollar per year industry. Small increases in efficiency in natural-gas production can bring very large financial advantages and appreciable conservation of a finite natural resource.

Efficient production from natural-gas wells requires a quantitative understanding of mixed-hydrocarbon vapor-liquid equilibria over a fair range of temperature and a wide range of composition and pressure, including the critical region. Because experimental data are available for limited variety of synthetic an natural gas mixtures, there is a good phenomenological picture of phase behavior; but, for calculating these equilibria, we do not have a satisfactory molecular-thermodynamics method that is reliable in all regions of practical interest, that is, near-to and far-from the (vapor-liquid) critical region. This research is directed at producing such a method using recent results from applied statistical mechanics: Wertheim's theory for (short) chain fluids as developed by Chapman, Radocz and others, coupled with renormalized-group (RG) theory of White recently developed and extended top mixtures by Lue and Prausnitz.

The essence of RG theory is to include in a "classical" equation of state corrections for density fluctuations that are not important at conditions remote from critical but become very important near the critical point. White's method (as extended by Lue and Prausnitz) does that in a relatively simple way, thereby much improving the ability to calculate phase equilibria near to and far from the critical region.

The ultimate goal of this work is to present a theoretically based, engineering-oriented computerized procedure for direct use by design engineers in the fossil-fuel industries. This research will produce new engineering-science knowledge. It will extend, modify and synthesized recent advances in statistical mechanics toward significant application.