Aggregation experiments were conducted inside a rotating levitation drum. The dust grains are deagglomerated by means of a pyrotechnical dispersion device and are supported against Earth's gravity by the aerodynamic drag forces of the explosion gas which rotates with the drum . A particle component with number density n has collision times 1
T = -, n ■ a ■ vr that depend on the individual particle's cross section o and the relative velocity vr between grains. Adopting the numerical results obtained by Niibold and Glassmeier , the collision times for nonmagnetic grains are by several orders of magnitude longer than for magnetic particles. Therefore, magnetic dust aggregation should proceed rapidly, producing huge elongated aggregate structures.
This could indeed be seen in our experiments: on time scales of the order of seconds only, we grew long chain-like aggregates and more complex, web-like structures (Figure 2). The biggest aggregates were approximately 0.5 cm long and could easily be seen with the naked eye. The structural differences relative to growth experiments with magnetically neutral silicate grains conducted by Heim  are easily noticeable. Since those were formed inside the same experimental setup using nonmagnetic silica (Si02) spheres, we conclude that the structural differences are due to the magnetic interactions between the ferrite particles.
The aggregates were collected for further study using optical and electron microscopy. The microscopic images also show the more complex web-like structures (Figure 2; right image) described by Nuth et al. . The latter contain chain-like aggregates similar to that in Figure 2 as 'spokes' around central clusters. We carried out a first analysis in order to establish the fractal dimensions of these aggreagtes using the density-density correlation Junction. The simple chain-like structures have D = 1, whereas D = 1.2 ± 0.1 for the more complex aggregates.
The microphysics of the coagulation process will be subject to accompanying theoretical and numerical studies . The highly irregular shape of our ferrite grains makes the comparison with numerical models, that usually rely on spherically shaped particles, rather difficult. Repeating these experiments with regularly shaped dust samples is one of our future goals. Furthermore, a microgravity experiment on the same subject will take part in this year's ESA student parabolic flight campaign (www.tu-bs.de/ institute/geophysik/forschung/projekte/hotzenplotz/index.html). As can be shown using Equation 1, only magnetic grains have time to coagulate in a parabolic flight aggregation experiment. On the other hand, the neutral dust component might be trapped by magnetic aggregates of the type shown in Figure 2 (right image). Using samples of magnetic and nonmagnetic dust, the 'fishing net' hypothesis will thus receive special attention.
Was this article helpful?