Clara Mata

Flow characteristics of concentrated emulsions of very viscous oil in water

This article advances ideas and presents experiments on the flow characteristics of concentrated emulsions of Venezuelan bitumen in water plus surfactant. These emulsions are studied under a variety of flow conditions, namely, between rotating cylinders, in a colloid mill, and in pipes. The ideas advanced here concern the modeling of the highly viscous bitumen drops as solid spheres and their fracture under contact forces between neighboring drops, as in comminution, rather than break‐up by hydrodynamic forces. Further, we observe and discuss the local inversion of an emulsion due to local increases of the bitumen fraction induced by flow and the conditions that lead to slip flow, in which the drag is reduced by the formation of a lubricating layer of water at the wall. We believe that the results presented here unveil mechanisms that take place in the pumping and pipelining of oil‐in‐water emulsions and therefore contribute to the understanding of the dynamic stability of these systems.

Self-lubricated transport of bitumen froth

Bitumen froth is produced from the oil sands of Athabasca using the Clarks Hot Water Extraction process. When transported in a pipeline, water present in the froth is released in regions of high shear, namely at the pipe wall. This results in a lubricating layer of water that allows bitumen froth pumping at greatly reduced pressures and hence the potential for savings in pumping energy consumption. Experiments establishing the features of the self-lubrication phenomenon were carried out in a 25 mm diameter pipeloop at the University of Minnesota, and in a 0.6m diameter pilot pipeline at Syncrude, Canada. The pressure gradient of lubricated flows in 25 mm, 50 mm and 0.6m diameter pipes closely follow the empirical law of Blasius for turbulent pipe flow; the pressure gradient is proportional to the ratio of the 7/4th power of the velocity to the 5/4th power of the pipe diameter, but the constant of proportionality is about 10 to 20 times larger than that for water alone. We used Reichardts model for turbulent Couette flow with a friction velocity based on the shear stress acting on the pipe wall due to the imposed pressure gradient to predict the effective thickness of the lubricating layer of water. The agreement with direct measurements is satisfactory. Mechanisms for self-lubrication are also considered.

Delayed-die swell and sedimentation of elongated particles in wormlike micellar solutions

Abstract It has been proposed recently that the combined action of inertia and non-linear viscoelasticity may be the origin of very peculiar behaviors with dramatic changes of flow type. Two examples are the problem of delayed die swell and the orientation of elongated particles sedimenting in solutions of wormlike micelles. These solutions give rise to well defined viscoelastic properties which can be tuned precisely by changing the surfactant weight fraction. Our analysis does not rely on any constitutive equation and our results strongly support the interpretation that delayed die swell and sedimentation of long bodies in wormlike micellar solutions are ruled by a change of flow type from subcritical to supercritical.

Foam control using a fluidized bed of hydrophobic particles

Abstract Applications of foams and foaming are found in many industries such as the flotation of minerals, enhanced oil recovery, drilling in oil reservoirs, insulation, construction and refining processes such as vacuum distillation and delay-coker reactors. However, foaming and defoaming are not yet understood. Foams trap gas, and are not wanted in many applications. It has been found that foaming may be strongly suppressed by fluidizing hydrophilic particles in the bubbly mixture below the foam, in a cold slit bubble reactor. This suppression is achieved by increasing the wetted area of the solid’s surface (walls and particles), by bed expansion and by decreasing the gas hold-up by increasing the effective density of the liquid–solid mixture. Never before has a fluidized bed been used to study the antifoam action of hydrophobic particles. In this work, we fluidized hydrophobic and hydrophilic versions of two different sands in a slit bubble reactor. We found that the hydrophobic sands suppress the foam substantially better than their hydrophilic counterparts. We also observed that, when foam is not present in the reactor (i.e. at high liquid velocities), the gas hold-up in the bubbly mixture was higher for the hydrophobic version of one sand. This result may be explained in terms of attachment of the particles onto the air bubbles, which increases the residence time of the gas phase. On the other hand, the gas hold-up in the bubbly mixture for the hydrophobic version of the other sand was smaller. A possible explanation is that the bubble adhesion to a non-wettable particle, leads to a decrease in the apparent density of the particle, which in turn is responsible for a larger bed expansion and smaller gas hold-up compared with wettable particle systems. These results suggest that the degree of hydrophobicity matters. Hydrophobic particles appear to break, and not only suppress foam; and they may have a wider application.

Cell motion in a two-stream microfluidic channel

Microfluidic channels have been proposed as a method for removal of cryoprotective agents from cell suspensions [Fleming, Longmire, and Hubel, J. Biomech. Eng. 129, 703 (2007)]. The device tested consists of a rectangular cross section channel of 500 μm depth, 25 mm width, and 160 mm length, through which a cell suspension and wash stream flow in parallel. Cryoprotective agents diffuse from the cell stream to the wash stream and the wash stream is discarded. The washed cell stream is then ready for use. This device must be capable of removing 95% of the dimethyl sulfoxide (DMSO) from the cell stream with minimal cell losses. Our previous studies have demonstrated our ability to remove DMSO [Mata, Longmire, McKenna, Glass, and Hubel, Microfluid. Nanofluid. 5, 529 (2008)]. The next phase of the investigation involves characterizing the influence of flow conditions on cell motion through the device. To that end, Jurkat cells (lymphoblasts) in a 10% DMSO solution were flowed through the microfluidic channel in parallel with a wash stream composed of phosphate buffered saline solution (PBS). Average cell stream velocities were varied from 0.94 to 8.5 mm/s (Re 1.7 to 6.0). Cell viability at the outlet was high, indicating that cells are not damaged during their passage through the device. Gravitational settling caused an accumulation of cells near the bottom of the channel, where flow velocities are low. Cell settling leads results in an initial transient period for cell motion through the device. For the initial portion of cells flowing through the device, cells tend to accumulate in the device until a critical device population time is reached. Cell recovery (number of cells out of the device divided by the number of cells input to the device) is high (90–100%) after the device has been fully populated. For a single stage device with average cell stream velocities of ⩾6 mm/s, cell recovery was 90–100%. As more stages are added to the device, the population time for the device increases. Gravitational settling of cells also leads to a time-varying cell concentration from the input syringe to the inlet of the channel, as well as cell losses due to cells remaining in the horizontally-oriented syringe. Reorienting the syringes to a vertical position eliminates these losses. Cell motion within the channel can be modulated by the flow conditions used. For sufficiently high Reynolds numbers, the Segre-Silberberg effect [Segre and Silberberg, J. Fluid Mech. 14, 115 (1962)] can be used to move cells from the low velocity region of the cell stream to a higher velocity region thereby reducing the transient portion of processing the cells and improving overall recovery of cells through the device.

Comparison of Geometries for Diffusion-Based Extraction of Dimethyl Sulphoxide From a Cell Suspension

Microfluidics can be used in a variety of medical applications. In this study, a microfluidic device is being developed to remove cryoprotective agents from cells post thaw (1–150ml). Hematopoietic stem cells are typically cryopreserved with Dimethyl sulphoxide (DMSO), which is toxic upon infusion. Conventional methods of removing DMSO results in cells losses of 25–30%. The overall objective of this study is to characterize the influence of flow geometry on extraction of DMSO from a cell stream. For all the flow geometries analyzed, flow rate fraction, Peclet Number, and channel geometry had the greatest influence on extraction of DMSO from the cell stream. The range of flow rate fractions that can achieve the desired removal ranges between 0.10 and 0.30. Similarly, the range of Peclet numbers is 250–2500. Distinct differences in channel length could be observed between the different flow configurations studied. The flow rates and channel geometries studied suggest that clinical volumes of cell suspensions (1–100ml) can be processed using a multi-stage microfluidic device in short periods of time (<1hr).

Post Thaw Processing of Red Blood Cells Using Microfluidics

Transfusion of red blood cells (RBCs) has been an integral part of medicine since the middle of the last century. In particular, RBCs have been the most transfused blood product in battlefield trauma care[1]. Cryopreservation of red blood cells has been used in particular by the military for a variety of reasons. The ability to preserve cells facilitates transportation of blood from one location to another. Combat can result in sharp increases in blood use for the treatment of civilian and military and frozen blood supplies can be used to meet variations in blood demand. In the civilian blood services, cryopreserved blood is typically used only for rare blood types[2] or for transfusion of neonates[3].Copyright

Levitation of Core Flows

A simple model is proposed for a 2D horizontal core annular flow in which the effect of gravity due to the difference in the densities of the two fluids is the eccentricity of the core. We split the domain through the center of the core; we characterized each sub domain by means of local variables, for instance the holdup ratio. We found that the smallest global pressure drop is achieved if and only if the local holdup ratios are equal to the global holdup ratio (for a perfect core annular flow this only happens when the core is centered). We used this result in a direct simulation of spatially periodic 2D wavy core annular flows carried out under the assumption that the viscosity of the oil core is so large that it moves as a rigid solid which may nevertheless be deformed by pressure forces in the water. The waves that develop are asymmetric with steep slopes in the high pressure region at the front face of the wave crest and shallower slopes at the low pressure region at lee side of the crest, as Bai et al. describe (1996). However two new issues are confronted in our 2D simulation. First, the shape and length of the upper and lower waves are different. Second, the displaced 2D core can be thought to represent eccentricity. We conclude that a positive pressure force is required to levitate the core off the wall when the densities are not matched and that the difference in the upper and lower wave shapes restore the effect of the eccentricity by allowing the local holdup ratios to be equal to the global holdup ratio; thus the pressure drop is the smallest possible.