The influence of pipe diameter on flow regime transitions and on the modeling of stratified, slug, and annular flows is being studied. Two flow facilities are available. One has horizontal pipes with diameters of 1 in., 2 in., and 4 in. The other is a vertical system with pipes of 3/8 in., ¾ in., and 1½ in. Accomplishments include the development of a theory to predict the transition from stratified to slug flow, the development of an interpretation of entrainment measurements for annular flow in terms of the fundamental rate processes, the development of an understanding of the wave patterns in stratified flow, and the use of photographic methods to determine drop size in annular flow. A new phenomenon of hindered droplet turbulence has been discovered.

Laboratory and supercomputer experiments are being used to understand the structure of turbulence close to a wall. Highly organized flows are being identified that are responsible for the sustaining of wall turbulence. The effect of external influences, such as pressure gradients, drag-reducing agents, and imposed flow oscillations, on these structures is being studied. New photographic techniques (PIV) are being exploited to obtain simultaneous measurements at as many as 12,000 points.

Turbulent flow over wavy surfaces is being studied both in the laboratory and by a direct numerical simulation. A particular emphasis during the year is the separated region that exists for large-amplitude waves. This well-defined separation bubble is being studied to provide the physical understanding needed to compute and control separated flows. The flow in the separated region is highly three-dimensional and seldom resembles the pattern indicated by the time-averaged streamlines.

Gas absorption at an interface is of considerable interest in environmental and processing problems. It is controlled by flow fluctuations in the liquid in a region of about 200 microns thickness close to the interface. These fluctuations are greatly enhanced when waves are present. Experiments are being conducted to understand this process. A technique involving oxygen quenching of fluorescence is used to study the concentration field close to the interface. Optical methods are being developed to map out the instantaneous spatial variation of wave slope.

One of the more interesting phenomena that is observed in concentrated suspensions is the development of instabilities when particles of different size or density are present. As a result of this instability, suspensions that are initially homogeneous may develop large-scale inhomogeneities. In earlier work, we successfully predicted the stability mechanism and derived a stability criterion based on linear stability theory. A weakly nonlinear theory explained additional details of the motion. In our ongoing efforts, we are using numerical many-particle simulations to calculate the resistance functions required to apply the stability criterion to real multiphase systems.

In many engineering applications, we require a knowledge of the velocity field for flow past porous objects. Simplified models, such as Darcy's law, provide a reasonable description of the flow in the interior, but require empirical coefficients to match the boundary conditions with the outer flow. In our research effort, we are investigating the nature of the boundary conditions by solving for the exact microscopic flow field in model porous media. The use of Green's functions converts the governing partial differential equations into a set of singular integral equations over the surfaces of the particles.

The dispersed phase of an emulsion is deformed by the shear forces present when an emulsion flows through a tube. Upon exiting the tube, jets of emulsions are known to undergo large increases in diameter, probably due to relaxation of deformed droplets. By modeling this process, we have learned much about the behavior of viscous jets in a gravitational field and the multiple modes of instability possible when parallel jets interact. We are now combining these results with the complex rheology present in emulsified systems. Our goal is to provide a rational connection between emulsion properties and the observed jet diameter.

Stokesian dynamics is being used to show the perturbations possible when a colloidal dispersion contains a few particles with a radius different from an otherwise mon odisperse system. From the results we hope to explain the well known order/disorder transitions that occur when nominally monodisperse systems are subjected to increasing shear rates.

Application of several kilovolts per millimeter across a nonaqueous suspension can result in dramatic changes in suspension rheology. Electrical polarization forces give rise to a fibrinated structure within the suspension which is degraded by shear. We are investigating the origin of this structure, its strength and its response to shear by developing models for the behavior of suspensions of polarized particles. Idealized experimental systems are being used to clarify the role of particle and continuous phase material properties on the magnitude of the electrically induced enhancement of stress transfer seen in these systems. Particular attention is paid to particle and continuous phase conductivity and the role of the electrode/suspension interface.

Manipulating the tertiary structures of proteins is crucial to many biological technologies. X-ray diffraction is the technique of choice to gain such insight. However, proteins are notoriously difficult to crystallize and, as a result, data on tertiary structures remain limited. In this study, we investigate protein/protein interactions mediated by the solvent to learn better how to induce crystallization. Our work focuses on using continuous phase chemical potential as a variable in controlling the ordering process.