Quantifying Dynamic Processes in Reactive Materials: An Extended Burn Tube Test

A common method for measuring the reactivity of rapidly deflagrating materials has been to loosely pack a desired mixture into an approx. 10 cm long x 3 mm diameter acrylic tube, ignite the material on one end, and report the observed self-propagating flame velocity through the material. While this method can yield qualitative information, linking this to quantitative intrinsic properties, such as particle burn time, has remained challenging. In this work, we significantly redesign the traditional burn tube experiment. Between 25 and 250 mg of nanocomposite aluminum/ copper oxide (Al/CuO) thermite is loosely packed into the capped end of a 1.8 m long tube, and the remainder of the tube is left unfilled. The material is ignited using a hot wire, resulting in a steadily-propagating luminous front, which extends part, or all, of the way down the length of the tube. We suggest the behavior is a result of reactive entrainment, which occurs when the reaction time scale becomes longer than the characteristic momentum relaxation time scale. When this criterion is met, and when there are significant pressure gradients present or produced during the reaction, material will be entrained by the gases before and/or during the reaction; a behavior very different from conventional thermites, which use larger particle sizes. The effect of sample mass and tube diameter on propagation velocity is investigated, and we find a linear scaling with mass and a power law scaling with tube radius (r(-1.3)) between 1.56 and 4.76 mm radii. For several conditions, we observe the reaction complete; defined by the distance where the propagation velocity decreases to a fixed fraction of its steady value. The ratio of quench distance to propagation velocity was found to approach a constant value of 3.29 +/- 0.70 ms, which we suggest is the burn time of the Material. For the range of masses studied in this work, the burn time was found to be independent of mass, implying that it is an intrinsic particle burn time. We expect other quantitative information can be deduced from this experiment, so long as the chosen sample mass and tube dimensions allow one to observe the full extent of reaction.