Metals have a fundamental role in human society and have both Earth and Space based applications. However, despite their extensive use, their behaviour under certain conditions remains relatively unknown. In normal and microgravity environments that have a high oxygen concentration and pressure, metals have been seen to sustain combustion. The relative flammability and burn rate of metals during combustion differs with both the type of metal and size of the test sample. Previous and current studies have indicated which metals sustain combustion under specified conditions. Despite this advancement in the field of metals combustion, relatively little research has focused on how the configuration of a metal test piece alters the reported combustion data.
A preliminary experimental investigation into the effects that sample configuration has on the combustion of iron was performed. This involved the testing of four sample configurations. These were a small square rod (side length of 2.5mm), large square rod (side length of 4mm), rectangular rod (5.5mm x 3mm) and a hexagonal rod (diameter of 6mm). All test pieces were made from mild steel with a 95% minimum iron content. Tests were performed in normal gravity at a test pressure of 100psi. Each sample configuration was tested three times to obtain a statistically relevant sample size. All combustion tests were filmed using a video camera and recorder. Ultrasonic data was also collected to aid in the determination of the regression rate of the melting interface (RRMI). Ignition and final pressures for all tests were also recorded. The oxygen used was industrial oxygen, with a minimum of 95% purity and was supplied by Linde Gas.
Results for the combustion tests performed showed that sample size does effect the burn rate of iron rods. Larger samples tend to burn more slowly at a given test pressure. From the experimental investigation it was found that the small square rods, which had the smallest cross-sectional area, (6.25mm2) burned at the fastest rate (4.04mm/s) followed by the rectangular rod, (2.51mm/s) and the large square rod had the slowest RRMI of 2.33mm/s. Cross-sectional area of the large square and rectangular rods were very similar with values computed as 16 and 16.5mm2 respectively. Considering the cross-sectional area alone, and applying the current knowledge that samples with smaller cross-section, tend to propagate the burning region faster, it was expected that the large square rod would have burned faster than the rectangular rod. This assertion was not substantiated by the experiments performed in this study. It is suggested that the difference in the length to width ratio, hence configuration of the rectangular rod, caused them to burn at a faster rate. Estimations of the volume of the molten metal ball were calculated with the largest volume occurring for the rectangular rods (46.01mm3), followed by the large square rods (40.55mm3) with the smallest molten metal ball volumes associated with the small square rod (27.55mm2). No sample of the hexagonal rod ignited at the test pressure. It was concluded that as the sample was much larger than the other three, the threshold pressure was higher, with the test pressure falling below this limit. Due to restrictions on the test apparatus no further investigation into the hexagonal rods was performed.
The study performed investigated the effects that sample configuration had on the combustion process of iron. This however was only a preliminary study and more research in this area should be performed. Through experimental investigation of the configurational effect on the combustion of metals, an increase in the fundamental knowledge of metal combustion will be possible. This will improve the safety of current oxygen systems where metal components, susceptible to combustion are essential. Improved safety and knowledge of space based structures will also be possible with an advancement of knowledge in this field.