The growth of microorganisms in fermentations where oil had been maintained as the continuous phase was examined to determine whether advantage could be gained from the increased solubility of oxygen in hydrocarbon. Although cell concentrations were highest in the aqueous phase of oil‐continuous systems, due to the large oil fraction, productivities achieved per unit fermenter volume were generally equivalent to those obtained from water‐continuous systems. With the oil‐continuous emulsions, the power requirement for aeration and mixing was less, and phase reversal resulted in a threefold concentration of cells in the aqueous medium, thereby facilitating their recovery.

Polymeric thickeners typically used as mobility control agents in oil recovery processes suffer from a variety of deficiencies, such as shear instabllity and brine intolerance. Block polymers containing both hydrophobic and hydrophilic segments have been shown to form high molecular weight, micellar type structures resulting in highly efficient, shear stable thickeners. Anlonically polymerized diblock polymers of styrene and 2-vinyipyrldinium halides were used to study the effects of block structure on thickening ability and shear and brine stabili. I t was found that the low molecular weight poly(styrene-b-alkylpyridinium halide) polymers when dissolved in water behave as high molecular weight species. These materials appear to exist as an aggregated polystyrene core stabilized by a highly solvated polar region. These materials are highly effective, shear stable thickeners even at very low concentrations. They do, however, suffer from the typical brine on polyelectrolytes which results in coil collapse and a drastic reduction in viscosity.

No abstract

The comparison to the exact result is not so good for this model, especially at longer t (see Figure 6). The weali-coupling result does reproduce the general rate of decay, but does not reproduce the full set of oscillations present in the exact solution. I n Figure 7 the function $(A) for this model is presented. Xote that g(X)would correspond to the absorption line shape for a transition to the 10) state, assuming that the states 1 ;) carry no oscillator strength.The third model considered was a "square wave" SDF with two matrix elements with magnitude five times the rest, symmetrically disposed around 6 = 0. I n this case the "weak-coupling" result fails to duplicate the strong oscillations found in the exact solution (see Figure 8). As shown in Figure 9, strong oscillations show up in both # ( T ) and k (~) . This model represents a case of a noiidissipative interaction between a set of states. As discussed in the main body of this paper, one way to proceed is to obtain new zeroth-order states which are mixtures of IO) and the two nondissipative states, and use u weak-coupling formulation in terms of these new zeroth-order states. This is essentially the same procedure as discussed by Douglas96 with regard to "anomalously long lifetimes" of small molecules, Note that in Figure 8 this effect is operative, as the rate of decay of 1 (01 t)l is less than ~( t )~~. I n all these calculations the exact solution will show a recursion (ie., 1 (o/t)j 2 will increase) at T r = 2 r / x . This behavior is not reproduced by u(Qz2, although #(T) will show a similar recursion. I n general cr(t)zz will tend to underestimate all oscillations in the exact time-dependent solution. It is probable that any model SDF that results in even slight oscillations in a(t)zz may not be properly treated in the weak-coupling limit.A statistical treatment of micellar solutions is given. It is based upon a Flory analysis of the ideal associated solution model of Prigogine and coworkers. In part -4, which is mostly applicable to aggregating nonaqueous solutions, a new definition of the critical micelle concentration (cmc) is proposed; it is defined as the inflection point of the osmotic coefficient us. concentration curve. The dependence of the average degree of aggregation on the solute concentration is derived from the minimization of the Helmholtz free energy of the solution. The results indicate that the theory can account for all the essential features (including the so-called "anomalies") of micellar solutions. A comparison between theory and experiment shows the agreement is satisfactory. In part B, solvation phenomena are considered. It is found that empirical formulas previously employed for the correlation of the properties of solvating systems may be obtained by a straightforward application of the Flory formalism.

interactions (3), protein-enzyme (1|.), and antibody-antigen (5) systems, and to salt-like protein complexes with naturaXXy occurring polyelectroXytes, such as 1 :_Xeie acids (6) and poiysaccharides (7). UntiX now most of the workers concentrated their efforts on the appXication of anaXyticaX and eXectrophoretic techniques to the study of these compXexes, but recent use of x-rays (8), diffusion (9), osmotic pressure (1 0), and streaming birefrin gence (XX) has yieXded very promising information. It was to be expected that the fieid of protein inter actions wouXd be extended to interactions with synthetic (3) L* G.