DOI

https://doi.org/10.25772/RF9D-ZX42

Defense Date

2005

Document Type

Dissertation

Degree Name

Doctor of Philosophy

Department

Mechanical Engineering

First Advisor

Dr. Timothy M. Cameron

Second Advisor

Dr. James McLeskey

Abstract

In this work, an experimental method is developed to study the effects of particle size, flow rate, pulsation, particle/substrate material, and temperature on the short-term reentrainment of submicron particles. The particles tested are in the size range of 10-900 nm and are deposited by wetting the inside of capillary tubes with a liquid suspension. The tubes are then dried in a desiccator. The particles are reentrained under turbulent dry air flow conditions and a condensation particle counter is used to measure the number of entrained particles.There has been very limited work done with nanoscale particles in general and no previous experimental work has reported about this particular parameter set. In order to interpret the data, a bimodal lognormal probability density for the ratio of adhesion force to removal forces is suggested. The majority of particles is attached to the surface by strong forces and cannot be entrained. However, a small fraction of particles, called loose particles, is attached to the surface by much smaller forces. Based on experimental data, an analytical equation for the fraction of loose particles in terms of a dimensionless force is developed. This dimensionless force is a function of particle size and gas flow rate. The temporal variations of fraction of deposited particles are calculated by incorporating the fraction of loose particles with the model of Wen and Kasper (1989).The experimental data confirmed the theoretical expectation that entrainment strongly depends on particle size and decreases as the size of the particle decreases. Both higher flow rates and pulsation of the flow increase the entrainment. Pulsation causes the distribution of forces to broaden. It is shown that the effect of particle/substrate material on entrainment can be predicted by the compound Hamaker constant provided that the morphology and the roughness of the system remain the same. Otherwise, the effect of roughness or morphology may override the effect of Hamaker constant.

Rights

© The Author

Is Part Of

VCU University Archives

Is Part Of

VCU Theses and Dissertations

Date of Submission

June 2008

Included in

Engineering Commons

Share

COinS