Direct Evidence of Electron Tunneling in the Ionization of Sputtered Atoms

Abstract

Direct evidence has been obtained for electron tunneling between sputtered Cs + ions and the solid surfaces from which they originate. Strong neutralization of the Cs + ions is observed whenever the tunneling channel is opened by changing the surface work function so that a crossing of the Fermi level by the Cs 6s level occurs. A tunneling model which takes the spatial dependences of the position and the width of the atomic level into account explains the data successfully. PACS numbers: 79.20.Nc, 34.50.Hc, 73.30.+y During the ion-beam bombardment of solid surfaces, an appreciable number of sputtered atoms come out ionized. These ionized species form the basis of secondary-ion mass spectrometry for materials analysis. Mechanisms proposed recently for such inelastic processes involve either electronic excitations by atomic collisions 1 or atom-surface resonant electron tunneling. 2 " 5 There is a conceptual difference between these two approaches. In the electron-tunneling model, the discrete electronic level of the sputtered atom interacts with the continuum of electronic states of the solid. The central feature of this continuum of electronic states is the Fermi level which separates occupied and unoccupied states. On the other hand, the Fermi level plays no explicit role in collisional models. In this paper, we report the first direct observation of the importance of the position of the Fermi level in the ionization of sputtered atoms.We chose to study the sputtering of Cs + in this experiment. It is known that at small coverages, Cs atoms are chemisorbed as Cs + on metal surfaces, with the empty 6s level lying above the metal Fermi level. 6 Hence if the work function cp of the surface is larger than the ionization potential / of Cs (3.9 eV), the 6s level of the sputtered Cs atom will always face empty states of the metal when the Cs atom escapes and little neutralization by electron tunneling can occur. By use of an adatom electric dipole layer to adjust the relative position between the Fermi level and the vacuum level (as measured by a change in cp), the 6s level can be deliberately forced to "cross" the Fermi level (when/xp) as the Cs atom escapes, making electron tunneling energetically possible.We have used three substrates with different work functions to test this concept: Au film {cp = 5.27eV), w-Si(lll) (<p=4.57eV), and Al film {cp = 4.3 eV). 7 The experiment was performed in an ultrahigh-vacuum system with lx 10" 10 Torr base pressure. Auger analysis and low-energy electron diffraction were used for in situ surface characterization. The clean (7x7) Si(lll) surface was produced by repeated Ohmic heating of the Si(lll) substrate to 1100 °C for 30-sec intervals. Clean Al and Au films were deposited in situ on the clean Si (111) substrate by thermal evaporation. Li atoms were used to lower the work function of the surfaces because the small mass and ionic radius of Li cause the least perturbation to the momentum transfer of the sputtering process. Cs and Li atoms were deposited by thermal evaporation ...

Work function dependent neutralization of low-energy noble gas ions

“…1͒. The constant C depends on the details of the charge exchange model used, 24,59 but its value is not relevant for the discussion here. The velocity is defined as vϭv f Ќ because the neutralization probability is determined on the outgoing trajectory.…”

“…This lower work function compared to the energy of the first excited level is caused by the level shift of the ions near the surface. 56,59,60 The shift of the first excited level a few angstrom from the surface where the charge state is frozen is of the order of 1 or 2 eV. 13,14 For RN to take place, the Fermi level has to be near or above the energy of the first excited level at the freezing distance.…”

Surface analysis of thermionic dispenser cathodes

“…This is consistent with figure 5.5, where the work function below which the resonant neutralisation is possible for the different ions is approximately 1.5 eV lower than expected based on the energy of the level at infinite distance from the surface (see table 5.1). This lower work function compared to the energy of the first excited level is caused by the level shift of the ions near the surface [69,72,73]. The level shift of the first excited levels is of the order of 1-2 eV a few Ångstrom from the surface where the charge state is frozen [24,25].…”

Behavior of the reemitted slow positron on hydrogen-adsorbed Ni (111) surface

“…Reemitted process is consistent with two models. The first is a resonant-electron-transfer model derived from the theory for ion neutralization [12]. The second model relates the positron emission rate to the density of final states of the emitted positron [13].…”