Molecular-dynamics simulations of cluster–surface collisions: Emission of large fragments

Abstract

Large-scale classical molecular-dynamics simulations of (H2O)n (n=1032,4094) collisions with graphite have been carried out. The clusters have an initial internal temperature of 180 K and collide with an incident velocity in the normal direction between 200 and 1000 m/s. The 1032-clusters are trapped on the surface and completely disintegrate by evaporation. The 4094-clusters are found to partly survive the surface impact provided that the surface is sufficiently hot. These clusters are trapped on the surface for up to 50 ps before leaving the surface under strong evaporation of small fragments. The time spent on the surface is too short for full equilibration to occur, which limits the fragmentation of the clusters. The size of the emitted fragment is roughly 30% of the incident cluster size. The cluster emission mechanism is found to be very sensitive to the rate of the surface-induced heating and thus to the surface temperature. The incident cluster velocity is less critical for the outcome of the collision process but influences the time spent on the surface. The trends seen in the simulations agree well with recent experimental data for collisions of large water clusters with graphite [Chem. Phys. Lett. 329, 200 (2000)].

On the detection of mesospheric meteoric smoke particles embedded in noctilucent cloud particles with rocket-borne dust probes

“…Their velocity regime is similar to ours, however, their experiments utilize collision surfaces with temperatures up to 1400 K, which is above what is relevant for our applications. Nevertheless, studies of lower temperature collisions (see, e.g., Tomsic et al, 70 Andersson et al, 72 and Markovic et al 87 ) show that the preservation of initial kinetic energy is very similar for low and high temperature collisions; however, in the latter case, the final energy spectrum tends to include a significant amount of low energy particles. In addition to this, Tomsic et al 71 found that surfaces with temperatures ∼300 K produce relatively narrow velocity distributions of the scattered fragments, so our uncertainty in initial fragment energy of about 30% will cover much of the variance.…”

“…They should therefore also not contribute much to the current at BP. [69][70][71] If large ice particles, of say 50 nm, rebound and carry with them an initial charge of around −4e, they could only contribute a current of 4%-8% of the maximum current measured at BP. This follows from the results of Havnes et al, 2 Havnes and Naesheim, 3 Kassa et al 55 that an impacting 50 nm particle can typically produce 50-100 charged fragments.…”

“…The collision dynamics of nanoparticles is not very well known, and our assumptions about nanoscale ice particles are based on studies by Tomsic, 68 Tomsic, Marković, and Pettersson, 69 Tomsic et al, 70,71 and Andersson et al 72 of impacting pure ice particles. For nanoscale ice particles, their experiments show that O(100) eV-particles may conserve up to 70% of their initial energy when colliding with gold or graphite coated surfaces at an incident angle of 70 • .…”

“…As noted by Kuuluvainen et al, 74 the transfer of charge to nanoparticles impacting on surfaces as well as their bouncing properties is not well understood. In studies of pure ice particles, [68][69][70][71][72] it is found that surface impacts of smaller particles generally lead to that they stick to the surface and evaporate while larger particles bounce off and appear to be more or less intact. Such a scenario cannot explain the large production of charged fragments which has been observed in several rocket experiments, 1,3,55 where the secondary charge production in a triboelectric mechanism corresponds to that a particle of radius 50 nm should produce between 50 and 100 charged fragments, while a smaller particle of radius say 10 nm produces around five charged fragments.…”

Molecular dynamics simulations of the micro-solvation of ions and molecules during cluster–surface collisions

“…We employ an SPC-based [17], rigid water model that previously has been used for MD simulations of H 2 O cluster-surface impact [15,[18][19][20]. The Na þ ion interacts with its charge of 1e electrostatically with the point charges of 0.41e and )0.82e located at the positions of the hydrogen and oxygen atoms in the SPCwater, respectively.…”