We present laboratory experiments on the formation of macroscopic dust aggregates. The centimeter-sized highly porous bodies are produced by random ballistic deposition from individual micrometer-sized dust particles. We find packing densities between 0.07 and 0.15 for uncompressed samples, dependent on the shape and size distribution of the constituent dust grains. Impacts into these bodies are simulated by uniaxial compression experiments. We find that the maximum compression, equivalent to the highest protoplanetary impact velocities of $50 m s À1 , increases the packing density to 0.20-0.33. Tensile strength measurements with our laboratory samples yield values in the range 200-1100 Pa for slightly compressed samples. We review packing densities and tensile strengths found for primitive solar system bodies, e.g., for comets, primitive meteorites, and meteoroids. We find a consistency between packing densities and tensile strengths of our laboratory samples with those from cometary origin.

We present experimental results on the mechanical properties of macroscopic agglomerates formed by ballistic hit-and-stick deposition. The agglomerates, produced with a new experimental method, consist of monodisperse SiO2 spheres with 1.5 microm diameter and have a volume filling factor of phi=0.15, matching very closely the theoretical value for random ballistic deposition. They are mechanically stable against unidirectional compression of up to 500 Pa. For pressures above that value, the volume filling factor increases to a maximum of phi=0.33 for pressures above 10(5) Pa. The tensile strength of slightly compressed samples (phi=0.2) is 1000 Pa.

The outcome of the first stage of planetary formation, which is characterized by ballistic agglomeration of preplanetary dust grains due to Brownian motion in the free molecular flow regime of the solar nebula, is still somewhat speculative. We performed a microgravity experiment flown onboard the space shuttle in which we simulated, for the first time, the onset of free preplanetary dust accumulation and revealed the structures and growth rates of the first dust agglomerates in the young solar system. We find that a thermally aggregating swarm of dust particles evolves very rapidly and forms unexpected open-structured agglomerates. PACS numbers: 96.35.Cp, 61.43.Hv, 81.10.Mx It is now widely accepted that planets form from the nebula of gas and dust that comprises nascent solar systems. Inelastic, adhesive collisions between these dust particles eventually form kilometer-sized bodies, called planetesimals, which then collide under the influence of their mutual gravity to form planets [1][2][3][4][5][6]. After condensation of the micron-sized dust grains in the cooling gas, these initially collide with each other due to thermal (Brownian) motion, and, by adhesion due to van der Waals forces, form aggregates. The agglomeration rate of freshly condensed [7,8] preplanetary dust grains is determined by three factors: the collision cross section, the collision velocity, and the sticking probability of the dust particles, which are mutually interdependent. Laboratory experiments with micron-sized solid particles and dust agglomerates thereof have shown that, for moderate collision velocities y c # 1 m s 21 , the sticking probability is always unity [9][10][11]. The collision cross section and the collision velocity strongly depend on the morphology of the interacting preplanetary dust aggregates. Open-structured, fluffy particles generally have a larger cross section than compact grains, but couple also much better to the gas motion, so that relative velocities between fluffy agglomerates are suppressed. The gas-grain interaction is best described by the dust particles' response time to the gas motion, t f . In the free molecular flow regime, t f~m s a , where m and s a are the mass and the geometrical cross section (i.e., the projected area) of the dust aggregate. Aggregation models [1,12] for the Brownian motion-driven dust growth predict a scenario in which dust clusters of similar mass predominantly contribute to the agglomeration process. This leads to the evolution of a quasimonodisperse distribution of aggregate masses and to a relation between aggregate mass and size s ofwith an exponent ("fractal dimension") in the range of D f ഠ 1.8 2.1. For quasimonodisperse systems, the mean aggregate mass grows by a power law in time, m͑t͒~t z , and is related to the mass dependence of the collision cross section s c~m m and the collision velocity y c~m n of the aggregating particles through m 1 n z21 z (e.g., see the review in [13]). Here, z, m, and n are the respective exponents of the assumed power law functions for ...

We derive the diffusion coefficient D of dust under the conditions of a protostellar disk. A reliable estimate of this quantity is essential for the transport, especially the sedimentation, of dust particles in the disk environment. In contrast to earlier treatments, the diffusion coefficient is derived on the basis of a mean field theory that allows a reliable determination of D. In the course of this derivation, we used two different turbulence models. We applied the diffusion coefficient in order to calculate how dust grains sediment and how a dust subdisk develops. Starting from the physical conditions in the subdisk, we discuss gravitational instabilities of this structure.

Observed protoplanetary disks consist of a large amount of micrometer-sized particles. Dullemond & Dominik (2005) pointed out for the first time the difficulty in explaining the strong mid-IR excess of classical T-Tauri stars without any dust-retention mechanisms. Because high relative velocities in between micrometer-sized and macroscopic particles exist in protoplanetary disks, we present experimental results on the erosion of macroscopic agglomerates consisting of micrometer-sized spherical particles via the impact of micrometer-sized particles. We find that after an initial phase, in which an impacting particle erodes up to 10 particles of an agglomerate, the impacting particles compress the agglomerate's surface, which partly passivates the agglomerates against erosion. Due to this effect the erosion halts within our error bars for impact velocities up to ∼30 m s −1 . For larger velocities, the erosion is reduced by an order of magnitude. This outcome is explained and confirmed by a numerical model. In a next step we build an analytical disk model and implement the experimentally found erosive effect. The model shows that erosion is a strong source of micrometer-sized particles in a protoplanetary disk. Finally we use the stationary solution of this model to explain the amount of micrometer-sized particles in observational infrared data of Furlan et al. (2006).

We performed micro-gravity collision experiments in our laboratory drop-tower using 5-cm-sized dust agglomerates with volume filling factors of 0.3 and 0.4, respectively. This work is an extension of our previous experiments reported in Beitz et al. (2011) to aggregates of more than one order of magnitude higher masses. The dust aggregates consisted of micrometer-sized silica particles and were macroscopically homogeneous. We measured the coefficient of restitution for collision velocities ranging from 1 cm s −1 to 0.5 m s −1 , and determined the fragmentation velocity. For low velocities, the coefficient of restitution decreases with increasing impact velocity, in contrast to findings by Beitz et al. (2011). At higher velocities, the value of the coefficient of restitution becomes constant, before the aggregates break at the onset of fragmentation. We interpret the qualitative change in the coefficient of restitution as the transition from a solid-body-dominated to a granular-medium-dominated behavior. We complement our experiments by molecular dynamics simulations of porous aggregates and obtain a reasonable match to the experimental data. We discuss the importance of our experiments for protoplanetary disks, debris disks, and planetary rings. The work is an extensional study to previous work of our group and gives a new insight in the velocity dependency of the coefficient of restitution due to improved measurements, better statistics and a theoretical approach.

In a protoplanetary disk, dust aggregates in the µm to mm size range possess mean collision velocities of 10 to 60 m s −1 with respect to dm-to m-size bodies. We performed laboratory collision experiments to explore this parameter regime and found a size-and velocity-dependent threshold between erosion and growth. By using a local Monte Carlo coagulation calculation and complementary a simple semi-analytical timescale approach, we show that erosion considerably limits particle growth in protoplanetary disks and leads to a steady-state dust-size distribution from µm to dm sized particles.

All atmosphere-less planetary bodies are covered with a dust layer, the so-called regolith, which determines the optical, mechanical and thermal properties of their surface. These properties depend on the regolith material, the size distribution of the particles it consists of, and the porosity to which these particles are packed. We performed experiments in parabolic flights to determine the gravity dependency of the packing density of regolith for solid-particle sizes of 60 µm and 1 mm as well as for 100-250 µm-sized agglomerates of 1.5 µm-sized solid grains. We utilized g-levels between 0.7 m s −2 and 18 m s −2 and completed our measurements with experiments under normal gravity conditions. Based on previous experimental and theoretical literature and supported by our new experiments, we developed an analytical model to calculate the regolith stratification of celestial rocky and icy bodies and estimated the mechanical yields of the regolith under the weight of an astronaut and a spacecraft resting on these objects.