Our goal is to analyse how dust, ice and icy dust grains stick together (aggregate) in a microgravity environment that best simulates the protostellar regions where these particles combine to form planetesimals, cometary nuclei and asteroids. The data is expected to provide insight into current aggregation theories, as well as to contribute knowledge on the initial conditions and clumping properties leading to the creation of planetary systems.

Large gas and dust clouds in the interstellar medium indicate active star-forming regions--the Orion nebula is a famous example. As the clouds collapse, new-born (proto) stars form at their centres. One of the most important observational discoveries of the last decade is that most new-born (proto) stars are surrounded by a flattened disk, from which stars, planets and small bodies form (Figure 1). Since 1995 this includes the detection of over 100 extra-solar planets.

Figure 1: Properties of the protoplanetary disk at varying radii and superimposed on an actual image. Our experiment could provide immediate insight into the prevailing conditions contributing to the where and how of planet formation. (Photo Credit: Arturo Gomez of CTIO/NOAO, the Hubble Heritage Team and NASA).

At present, the formation of planets is not fully understood. Recent studies (Blum et al, 2002; Love et al, 2004) have shown that small, micron-sized dust particles can aggregate into millimetre-sized agglomerates within seconds, emulating what is believed to be the first stages of planet formation, as predicted by current theory. However, little experimental data exists pertaining to the aggregation of ice or ice-covered dust particles (Ehrenfreund et al, 2003). Since ices are ubiquitous in protostellar regions, the agglomeration properties of icy particles are necessary for understanding the formation of (proto) planets. the observed behaviour of the particles, based on the composition of the colliding grains, has direct relevance to protostellar regions at varying radial and azimuthal distances. Within the protoplanetary disk, ices are likely to evaporate in the vicinity of the protostar due to thermal desorption, but remain prevalent and influential in the outer regions.

Naturally, most experiments are on the surface of the Earth where 1g gravitational effects, including sedimentation and convection, overshadow the dominant forces normally governing protostellar matter (Blum, 2000). However, in interstellar space, microscopically small dust grains initially collide gently due to Brownian motion, differential drift motions and gas turbulence, and agglomerate due to adhesive surface forces, like van der Waals forces and hydrogen bonding (Fraser et al, 2003). Investigation of these effects is only possible under reduced and/or microgravity conditions because particle sedimentation does not occur during the weightless parabola of the flight, and therefore the same cloud of particles can be observed suspended in a chamber for a longer period of time, making it much more likely that we will observe aggregation, and agglomeration. In gravity-limited experiments, i.e. ground based laboratory work, particle samples are simply unable to aggregate before falling to the bottom of the experiment chamber. Extending the drop time by blowing a nitrogen or argon gas upwards in an Earth-bound test chamber is successful for short time periods, but introduces additional factors to the study. This particle suspension method in 1g is still a viable and important pre-flight test phase to our proposed research, but it cannot altogether address the particulars of the environment and forces that we wish to study.

In addition, new experiments performed just last year (Love et al, 2004) aboard the International Space Station used readily available substances, including table salt and hand-held video cameras, to perform informal aggregation studies in an extended zero gravity setting for the first time. Their results again confirm visible and repeatable clumping activity within a matter of seconds in a high-density particle concentration at standard room pressure and temperatures. These initial experiments lacked the composition and vacuum properties of interstellar space-which we propose to simulate in a chamber designed specifically for a more comprehensive study-but support the pursuit of additional, exacting studies in microgravity environments.

Current understanding of extraterrestrial ices relies entirely on comparisons between data accrued in Earth-based laboratory studies and observations of remote ices by spectrograph telescopes and planetary missions. Therefore, open questions remain as to whether ice studies (and ice production) on Earth are also good analogues for ices found (and produced) in a variety of space environments (Frase r et al., 2003).

The two major processes that we will be evaluating in our study are aggregation and agglomeration. Aggregation is a collection of particles that are loosely bound. On the other hand, agglomerates are more tightly bound, due to a surface reconstruction, often through melting or annealing on impact (Figure 2, next page). The two types differ depending on: 1) the nature of the particle contact and bond strength, and 2) the total surface area of the combined structure as compared to the total surface area of the pre-collision particle system. These effects will be visible in a digital frame-by-frame data analysis, as previous studies have already shown (Love et al, 2004). We hope for conclusive results by measuring particle growth, particle impacts (i.e. velocity and sticking probability), among other outcomes.

Figure 2: Predicted aggregate and agglomerate particle structures. Structure recognition provides clues to the physical processes involved, growth time scales and probable theoretical models.

To correlate our results we will be using the dust samples as a control sample by comparison with previous studies. In addition, we will be attempting an experiment that has not been done before: we will be creating and studying ice covered dust grains (formed in flight), which are believed to most accurately represent the particles present in the planet forming regions of a protoplanetary disk. This is phase 3. If our experiment runs successfully, our data and results will represent the most accurate simulation of particle clumping in the circumstellar region, as performed to date.