Center for Astrophysics and Supercomputing

We investigate the stress distribution at the base of a conical sandpile using both analytic calculations and a three dimensional discrete element code. In particular, we study how a minimum in the normal stress can occur under the highest part of the sandpile. It is found that piles composed of particles with the same size do not show a minimum in the normal stress. A stress minimum is only observed when the piles are composed of particles with different sizes, where the particles are size segregated in an ordered, symmetric, circular fashion, around the central axis of the sandpile. If a pile is composed of particles with different sizes, where the particles are randomly distributed throughout the pile, then no stress dip is observed. These results suggest that the stress dip is due to ordered, force contacts between equiheight particles which direct stress to the outer parts of the pile.

Metal disks of different size and density were placed at the bottom of a bed of monodisperse granular material. The system was vibrated sinusoidally in the vertical direction. It was observed that, if the angular acceleration of the shaking was slightly greater than that of gravity, the metal disks rose to the top of the bed. This result has been known for over sixty years, but a basic understanding of the mechanism responsible for the rise of the disks is still a subject of debate. Our experiments and theoretical model show that the ascent speed of the disk is proportional to the square root of the disk density, approximately proportional to the disk size, and is a function of the disk's depth in the bed. We also investigated the speed of ascent of the disk as a function of the shaking frequency, f s . We found that the effective friction or drag coefficient, β, between the disk and the granular bed, is proportional to a functional form of the frequency: β∝ (f s −f c )−4, where f c is the critical shaking frequency for the disk to start moving through the bed. We discuss how such a dependency may arise.

This paper examines the usefulness of a discrete element method simulation in studying solids mixing in gas-#uidized beds. At steady state, a part of the solids were tagged with a dark colour and the rest were tagged with a light colour. The state of mixing of bed solids was studied by analysing the variation of the proportions of the marked particles with time and position in the bed. This was achieved by scanning the bed using a sampling box. The variance of mixture composition based on the samples was incorporated into a mixing index which describes the degree of mixing of the particles at a particular time. This method enables assessment of the overall mixing behaviour in terms of the rate of mixing (through estimation of the time required for the mixing index to increase from zero to a certain value), together with the degree of mixing at the mixing equilibrium. Illustrative examples show that gas velocity and particle properties are important parameters in#uencing solids mixing in bubbling #uidized beds.

Efficient regulation of blood flow is critically important to the normal function of many organs, especially the brain. To investigate the circulation of blood in complex, multi-branching vascular networks, a computer model consisting of a virtual fractal model of the vasculature and a mathematical model describing the transport of blood has been developed. Although limited by some constraints, in particular, the use of simplistic, uniformly distributed model for cerebral vasculature and the omission of anastomosis, the proposed computer model was found to provide insights into blood circulation in the cerebral vascular branching network plus the physiological and pathological factors which may affect its functionality. The numerical study conducted on a model of the middle cerebral artery region signified the important effects of vessel compliance, blood viscosity variation as a function of the blood hematocrit, and flow velocity profile on the distributions of flow and pressure in the vascular network.

One of the principal reasons for failure of endovascular aneurysm repair (EVAR) is the occurrence of endoleaks, which regardless of size or type can transmit systemic pressure to the aneurysm sac. There is little debate that type I endoleaks (poor proximal or distal sealing) are associated with continued risk of aneurysm rupture and require treatment. Similarly, with type III endoleak, there is agreement that the defect in the device needs to be addressed; however, what to do with type II endoleaks and their effect on long-term outcome are not so clear. Aneurysm sac change is a primary parameter for determining the presence of an endoleak and assessing its impact. While diameter measurement has been the most commonly used method for determining sac changes, volume measurement has now been proven superior for monitoring structural changes in the 3-dimensional sac.

Using the concepts of fractal scaling and constrained constructive optimization (CCO), a branching tree model, which has physiologically meaningful geometric properties, can be constructed . A vascular branching tree model created in this way, although statistically correct in representing the vascular physiology, still does not possess a physiological correct arrangement of the major arteries. A distancefunction based technique for "staged growth" of vascular models has been developed in this work to address this issue. Time-dependent constraints based on a signed-distance level set function have been added, so that the tree models will first be grown near the designated surface(s) and, then, gradually allowed to penetrate into the enclosed volume. The proposed technique has been applied to construct a model of the human cerebral vasculature, which is characterized by the above-mentioned distribution of the arteries.

One of the principal reasons for failure of endovascular aneurysm repair (EVAR) is the occurrence of endoleaks, which regardless of size or type can transmit systemic pressure to the aneurysm sac. There is little debate that type I endoleaks (poor proximal or distal sealing) are associated with continued risk of aneurysm rupture and require treatment. Similarly, with type III endoleak, there is agreement that the defect in the device needs to be addressed; however, what to do with type II endoleaks and their effect on long-term outcome are not so clear. Aneurysm sac change is a primary parameter for determining the presence of an endoleak and assessing its impact. While diameter measurement has been the most commonly used method for determining sac changes, volume measurement has now been proven superior for monitoring structural changes in the 3-dimensional sac. Determining the source of an endoleak and the direction of flow are necessary for proper classification; however, while computed tomographic angiography has high sensitivity and specificity for detecting endoleaks, it is limited in its ability to show the direction of flow. Contrast-enhanced duplex ultrasound, on the other hand, is better able to quantify flow and characterize endoleaks. Flow is evidence of pressure, and increasing intrasac pressure increases wall tension, thus inducing progressive aneurysm expansion until rupture. Hence, determining intrasac pressure is becoming a vital component of endoleak assessment. All endoleaks can create systemic pressure inside the aneurysm sac, and there are a variety of intrasac pressure transducers being evaluated to assess this effect. A clinical pathway for patients with suspected type II endoleaks is based on a combination of imaging and pressure measurements. Imaging alone requires at least two interval examinations to determine the trend, while pressure measurements give immediate reassurance or an indication to intervene. Although still under development, pressure measurement is destined for general use and will provide a scientific basis for the management of type II endoleaks.