Document Type


Date of Award

Spring 2018

Degree Name

Doctor of Philosophy in Chemical Engineering - (Ph.D.)


Chemical, Biological and Pharmaceutical Engineering

First Advisor

Edward L. Dreyzin

Second Advisor

Robert Benedict Barat

Third Advisor

Mirko Schoenitz

Fourth Advisor

Roman S. Voronov

Fifth Advisor

Demitrios Stamatis


This work investigates combustion of nanocomposite thermite powders prepared by arrested reactive milling (ARM). The focus is on how ARM as a top-down approach to nano-thermite building generating fully-dense nanocomposite particles with dimensions of 1-100 µm affects the rates and mechanism of their combustion. A variety of thermites are milled using both aluminum and zirconium as fuels combined with several oxidizers (WoO3, MoO3, CuO, Fe2O3, and Bi2O3). The powders are ignited using both an electrostatic discharge (ESD) and a CO2 laser beam.

A range of parameters vary in the first set of experiments in order to broadly understand the underlying combustion mechanisms of nanocomposite thermite powders. Only the aluminum thermites are considered in these experiments and had their particle sizes, preparation method (milled, mixed, or electrosprayed), and milling times adjusted in order to see their effects on combustion. Additionally the ESD ignition experiments vary the environment between air, argon, and vacuum, as well as varying the ignition voltages from 5 up to 20 kV at a constant capacitance of 2000 pF. The ignited particles are monitored using a photomultiplier tube (PMT) equipped with an interference filter. It is observed that the reaction rates of the ESD-initiated powders are unaffected by their particle size but are affected by their scale of mixing between their fuel and oxidizer within the particles themselves. The different preparation methods play a significant role in determining the powders performance. Mixed nano-powders agglomerated quite easily, which hinder their combustion performance. The electrosprayed powders perform well in all environments, and the milled powders perform best in oxidizer-free environments (when no reoxidation of the oxidizer could occur).

A set of experiments employing ESD ignition focus on the effects of powder load on its combustion properties. The experiments utilize a similar PMT setup with an additional 32-channel PMT coupled with a spectrometer to record optical emission in the range of 373-641 nm. It is discovered that when a monolayer of the powder was ignited, only single particles are ejected from the substrate and burned very rapidly. A thicker layer of powder (0.5 mm) struck by ESD produce an aerosol cloud, which ignite with a delay and burn substantially longer. It is theorized that the difference was due to different heating rates between the two experiments. In monolayer experiments, all ignited particles are ignited directly by ESD. Only a small fraction of particles in the thicker layered powder is heated directly by ESD; most particles are heated slower due to heat transfer from the initially ignited powder. More in depth experiments on the heating rate are conducted utilizing the fast heating of the thermites powders by ESD at ca. 109 K/s along with an experiment, in which the same thermite particles are heated and ignited by laser with the heating rate of ca. 106 K/s. It is discovered that laser-ignited particles combusted slower due to a loss of their nanostructure, while ESD-ignited particles maintained their nanostructure and burned much more quickly.

Utilizing the results from all the experiments, and combining them with combustion information previously obtained for Al and its ignition, with reaction controlled by polymorphic phase transformations in alumina (amorphous, gamma, and alpha), a model is developed enabling one to describe quantitatively the very high burn rates observed for the nanothermite particles rapidly heated by ESD. The model considers nanostructure accounting for the inclusion size distribution obtained from SEM images of actual milled particles, along with other considerations including heat loses, phase transformations, density changes, and particle size. The model is able to match combustion times and temperatures with those recorded from the earlier ESD combustion experiments.