On the role of thermal processes in sputtering and composition changes due to ions or laser pulses

Abstract

We here discuss some aspects of thermal processes due to ion-beam or laser-pulse interaction with solid surfaces. Considering first normal vaporization, i.e. particle emission in terms of the Hertz–Knudsen equation, we will show that it may have a limited importance with ions but has a very great importance with laser pulses. The latter remains true in spite of the fact that laser interactions are nearly always initiated by electronic events. Proceeding to normal boiling due to heterogeneous nuclei, the basic argument of Fucke and Seydel applies: the density of heterogeneous nuclei is too low (given the volume of the heated region) by many orders of magnitude to permit particle emission which is supplementary to that due to normal vaporization. Even boiling due to nucleation at the target surface can be questioned for the simple reason that a typical nucleus (50–100 nm) has dimensions similar to the ion range or the laser heating depth. Such nuclei therefore serve only to enhance normal vaporization. But phase explosion, also termed explosive boiling, is another problem. It is sufficient that the temperature approach the thermodynamic critical temperature, T tc, and nuclei are formed homogeneously at an enormous rate. Here, however, one must not forget that the problem enters of the time scale of the nucleation. If the values proposed by Martynyuk in 1977–1983 (M.M. Martynyuk, Phys. of Combustion and Explosions 13 (1977) 178; M.M. Martynyuk, Radio Engin. and Electronic Phys. 25 (1980) 100; M.M. Martynyuk, Russ. J. Phys. Chem. 57 (1983) 494) (1–100 ns) are still acceptable, then it follows that neither ion impact nor ps laser pulses can ever cause phase explosion, whereas the commonly used excimer laser pulses (∼30 ns ) will easily lead to it. The remaining thermal process, subsurface heating, is predicted to occur due to the target surface being fixed at a temperature characterizing particle emission. We regard this argument as wrong, if only because the temperature in question does not exist.