Brown University --
Joint Materials/Solid Mechanics Seminar Series
Department of Materials Science & Engineering, Knoxville, TN; and Oak Ridge National Laboratory, Materials Science and Technology Division, Oak Ridge, TN | |
Abstract: The indentation size effect has most often been studied using sharp, geometrically self-similar indenters like the Vickers square-based pyramid used in microhardness testing and the Berkovich triangular-based pyramid employed in nanoindentation testing. For these indenters, the size effect is manifested as an increase in hardness with decreasing depth of penetration, with the effect becoming apparent only at depths of less than a few microns. Recent studies have revealed, however, that the nature of the size effect depends critically on the geometry of the indenter. For example, the size effect for spherical indenters is manifested not through the depth of the penetration but rather through the size of the indenter itself, i.e., significant increases in hardness are observed as the radius of the indenter is decreased. Moreover, if the prevailing theories for the origin of the effect are correct, then there should be a significant influence of the tip angle for indentation with triangular pyramidal indenters. In this presentation, the influence of the indenter geometry on the size effect is examined by experiments conducted in a variety of materials with a variety of spherical and pyramidal indenters in order to critically evaluate prevailing mechanistic models. The presentation will also address the numerous difficulties encountered in obtaining meaningful experimental results.
* Research at the SHaRE User Center at the Oak Ridge National Laboratory was sponsored by the Division of Materials Sciences and Engineering, U.S. Department of Energy, under Contract DE-AC05-00OR22725 with UT-Battelle, LLC.
Brown University Center for Statistical Sciences Seminar
Professor Department of Mathematics/Statistics, Public Health and Law, Boston University | |
Refreshments at 3:45 pm |
Abstract: Many effective drugs, biologics and devices exist. Because of this, it is often considered unethical to undertake placebo controlled clinical trials to evaluate new treatments. Rather, randomized active control non-inferiority trials have become the norm. These trials, however, are accompanied by serious issues including selecting an active comparator, identifying previously run active comparator placebo controlled trials, deciding upon the form of comparison (e.g., differences or ratios), setting a non-inferiority margin, designing a new study where consistency with past trials hold, finding study sites where the trial can be performed, selecting the appropriate analysis set (Intent-to-treat or per protocol) and determining sample size. These issues have proven to be more problematic than naively anticipated. Further with the accumulation and advancement of these studies and the attempt to expand their use, new problems arose and others can be anticipated to arise. Some common problems are the realization that consistency with previous active comparator placebo trials does not hold, that the multiple available ad hoc (non-theory driven) tests present confusion rather than help and that the outcome data for the active comparator often do not match the anticipated results. Newer problems are the attempts of switching from non-inferiority to superiority testing and superiority to non-inferiority testing, the use of interim analyses, multiple treatments and dealing with recurrent events as the trial outcomes. In this talk we present a brief overview of the above with emphasis on the latter items. We discuss approaches for dealing with the issues. Real world examples are used for illustrations.
Sponsored by the Charles K. Colson Lectureship and Publication Fund
Co-Sponsored by the Bruce M. Bigelow Class of 1955 Lecture Series
Center for Fluid Mechanics
And
The Fluids, Thermal and Chemical Processes Group
Of
The Division Of Engineering
Seminar Series
Laboratoire de Génie Chimique Toulouse, France | |
Abstract: Dynamics of solid/liquid suspensions under shear flow. The dynamics of macroscopically homogenous sheared suspensions of neutrally buoyant, non-Brownian spheres is investigated in the limit of very small Reynolds and Stokes numbers using the Force Coupling Method.
Multi-body interactions are achieved by direct solution of Stokes flow equations coupled with the lagrangian tracking of particles in a fully periodic cubic domain. Statistical quantities (translation and rotation velocity fluctuation tensors, particle self-diffusion) were compared to the reference work of Drazer et. al.
The self-diffusion tensor was found to be anisotropic. Our results compared favourably with former numerical and experimental studies. The diffusion scales with the square of the volumetric concentration and is both related to velocity fluctuation enhancement and diffusion time increase. The pair probability density function computed while the particles move under shear flow has been compared with a purely random suspension. The maximum value is strongly enhanced by hydrodynamic interactions. Similar analysis is carried out on bidisperse suspensions for various concentrations of both species.
Department of Mathematics Colloquium
Scientific Computing Seminar
<--- 2006 Index