The interaction of energetic ͑up to 200 eV/atom͒ size-selected Co n clusters with HOPG is studied both experimentally and theoretically. Etching of the radiation damaged areas introduced by cluster impacts provides a measure of the depth to which the collision cascades are developed and allows a comparison of these data with the molecular dynamics simulations. Good agreement between the experimental results and modeling is obtained. It is shown that the projected range of the cluster constituents can be linearly scaled with the projected momentum ͑the cluster momentum divided by surface impact area͒. With decrease in cluster energies to ca. 10 eV/atom the transition from implantation to pinning is suggested. It is found that even after quite energetic impacts residual clusters remain intact in the shallow graphite layer. These clusters can catalyze reaction of atmospheric oxygen with damaged graphite areas under the thermal heating that leads to the formation of narrow ͑5-15 nm͒ random in shape surface channels ͑trenches͒ in the top few graphene layers. Thus, small imbedded Co nanoparticles can be used as a processing tool for graphene.

The formation of nanostructures on rutile (TiO2) surfaces formed after the implantation of kiloelectronvolt-energy cluster ions and megaelectronvolt- to gigaelectronvolt-energy multiply charged heavy ions (Iq+, Taq+ and Uq+) is studied. Despite the differences in stopping and energy transfer mechanisms between the kiloelectronvolt-energy cluster ions and megaelectronvolt-energy monoatomic ions, their impacts lead to a similar type of surface damage, namely craters. For the cluster ion implantation the craters are caused by the multiple-collision effect (dominated by nuclear stopping) and the high density of energy and momentum transferred to the target, while for the case of megaelectronvolt multiply charged ions the craters are probably formed due to the Coulomb explosion and fast energy transfer caused by the electronic stopping. At ion energies in the gigaelectronvolt range, nanosize protrusions, so-called hillocks, are observed on the surface. It is suggested that electronic stopping leads to the formation of continuous tracks and the transferred energy is high enough to melt the material along the whole projectile path. Elastic rebound of the tension between the molten and solid state phases leads to liquid flow, expansion and quenching of the melt, thus forming the hillocks. Atomic force microscopy measurements carried out under different environmental conditions (temperature and atmosphere) suggest that the damaged material at the nanosize impact spots has very different water affinity properties (higher hydrophilicity or water adsorption) compared with the non-irradiated rutile surface.

This work describes construction of a source and optimisation of its parameters for production of cluster ion beams using material ablation by the second harmonic of a Nd:YAG laser (532 nm). The influence of different source parameters such as carrier gas pressure, laser power, delay time between gas, and laser pulses as well as nozzle configuration on the cluster formation are studied. For the current experiments the laser ablation cluster source was optimized for production of Con+ cluster ions. Clusters with n up to 150 atoms are registered by a time-of-flight mass spectrometer. Deposition of size-selected Co50+ clusters with kinetic energies in the interval of 250-4850 eV/cluster on highly ordered pyrolytic graphite is studied. At the highest impact energies the clusters are implanted. Craters and well-like structures can be seen by scanning tunneling microscopy at impact spots. A decrease in cluster kinetic energy leads to formation of bumplike structures which probably represent damaged graphite areas with incorporated Co atoms. Further decrease in the cluster impact energy to the level of 450-250 eV/cluster creates condition for so-called cluster pinning when the cluster constituents are intact but the energy transferred to the graphite is still enough to produce radiation defects to which the cluster is bound.

No abstract