Office of Justice Programs

Bloodstain pattern analysis is a technique used in crime-scene reconstruction to determine the point of origin of a blood droplet as well as the method of its creation, e.g., dripping, wiping, or low-to-highspeed impact caused by anything from blunt trauma to cast-off to gunshot wounds. The primary problem of interest in this analysis is to determine of the initial size, speed, and impact angle of a blood droplet that has struck a solid surface through an examination of the bloodstain pattern left on the surface. This research project addressed this problem using a detailed fluid dynamical study of the impact and spreading of a liquid droplet on planar surfaces of variable roughness, wettability, and absorbency oriented at various angles with respect to the velocity vector of the approaching droplet. In particular, this research used a coordinated program of laboratory experiments and numerical simulations to examine the influence of the many parameters associated with such droplet impacts. Given its importance to crime-scene reconstruction, the focus of this work is on how to use the shape of the final stain to determine the initial conditions of droplet impact. The main goals of this research program are as follows: a) quantify the effects of the droplet impact angle, the initial droplet size and speed, and the solid surface roughness and wettability on the pattern of the final observed stain, and b) analyze this data to provide simplified, but relevant phenomenological models of droplet impact, spreading, and splashing that can be directly used by practitioners in the field of forensic science. The experimental portion of this research used a specially designed droplet generator to create individual liquid droplets of a specified diameter and velocity. The liquid was a specially prepared mixture of water, glycerin, and alcohol that has the same density, viscosity, and surface tension as human blood. These liquid droplets were propelled against three different solid surfaces, i.e., glass, bathroom tile, and paper, held at a specified angle with respect to the vertical. The droplet impact speed was measured by a pair of photodiodes and the impact of the droplet against the surface was recorded with a high-speed video camera held normal to the impact surface. The video images were analyzed to obtain the maximum spreading width and length of the droplet as well as the number of irregularities or projections in the shape of the droplet (spines) that may be present. These data were then analyzed to determine any specific relationships with the main fluid dynamical parameters that govern this impact process, i.e., the Reynolds and Weber numbers. The development of a numerical simulation of the droplet impact problem was partially completed during this project. When completed, this will be a time-dependent, 3-D, multiphase flow simulation based on a wavelet adaptive grid that was designed from scratch using modern computing techniques optimized for fast parallel processing on a single graphic processing unit (GPU). This simulation will be capable of accurately predicting the motion of a liquid droplet as it impacts and spreads along a solid surface, a typical droplet impact event.