DEVELOPING AN EMPIRICAL MODEL TO PREDICT THE SIZE OF VERY-FINE PARTICLES IN NATURAL WATER SUSPENSIONS VIA A LABORATORY SETTLING EXPERIMENT

Understanding the settling of suspended particles in water is important in geologic and engineering fields. Settling of suspended load has been studied with the aim of determining useful empirical equations for settling velocity. Studies have commonly taken the approach of modifying the Stokes settling equation for spheres for various conditions affecting Reynolds numbers, including grain angularity. Equations for settling velocity use particle size as an input variable. Most studies have modeled particles from about 10μm to 1mm, but commonly in the 100μm to 1mm size range. Variability in the shapes of natural suspended particles has been the biggest challenge in applying the models to natural systems.

The goal of the current investigation is to predict the particle size and concentration of very-fine suspended load via a routine laboratory settling experiment. This study differs from others in several ways. First, the equations solve for particle size as the dependent variable, with time as the independent variable. Second, the size range investigated is from ~ 5μm to ~ 250nm, addressing the seldom-investigated sub-micron region. Third, particles examined are natural suspended material settled from ground water samples. And fourth, models incorporate grain shape as observed by dynamic imaging particle analysis and scanning electron microscopy (SEM).

Particle size vs. settling-time data collected from suspensions in this study are being compared to several published model equations for settling of theoretical particles of a similar size range. Preliminary comparisons indicate that power functions and higher-order polynomial functions both under-estimate the settling velocity of the very-fine particles. However, both types of functions do approximate the slope of size/settling time closely over the size range present. Preliminary equations developed from the laboratory measurements of this study indicate that a second-order polynomial equation may be effective in predicting size of very-fine suspended particles. The relative effects of shape and angularity observed in the natural sediment grains of this study are being investigated and incorporated into model equations.