Ge(001) before and after gold evaporation and annealing
Kernel of the 4-probe microscope (left), crushed tip on Si surface (middle), two STM tips over gold structures on Ge(001) surface

Electron- and photon- induced of alkali halides

Alkali halides are perfect model materials to investigate interactions of ionizing radiation with solid state especially effects leading to defect creation and surface decomposition. Surprisingly, despite the fact that a huge amount of work has been devoted to those phenomena over decades, still new facts come to light changing our understanding of process of erosion of crystals irradiated by electrons or photons.

In the crystal irradiated by electrons or photons of energy exceeding bandgap a multitude of phenomena occur.

In general, the energy is distributed in the form of excitons and lattice vibrations. Excitons possess kinetic energy (they are 'hot'). They can either lose it due to interaction with phonons and consequently be selftrapped, or reach surface and decompose there, leading to emission of atoms from the crystal with kinetic energy exceeding thermal energy expected from Maxwell distribution for given temperature.

Exciton selftrapping leads in some extent to creation of pairs of Frenkel defects (H-center - an halogen molecule in interstitial position, F-center - a vacancy in halogen sublattice with an electron bound in), which in turn diffusing in the crystal lead to emission of neutral atoms from the surface with kinetic energy corresponding to Maxwell distribution for crystal temperature. Very mobile H-centres are responsible for emission of halogen atoms, whereas F-centers are responsible for emission of alkali metal atoms. For more details see M. Szymonski Dan...., J.J. Kolodziej, M. Szymonski PRB

However, defect interaction with different sites on the surface leads to further complication. An H-center reaching the surface forms an ad-atom on the surface. An atom is bound very weakly to the ionic lattice and can be evaporated easily. The case of F-centers is more complicated. An excited F*-center is mobile in the crystal, whereas a ground-state F-center is not. Moreover, ground-state F-center does not possess enough energy to induce an atomic emission. As for an excited form, it is possible not from the perfect crystal surface, however, but from a low-coordinate site (and edge or kink site). The F*-centers which fail to get to low-coordinated sites are reflected back to the bulk. Since excitation have a finite lifetime, defects deexcite to their ground, immobile form. The defects accumulate in the proximity of the surface forming a network of traps for migrating H-centers. In that way, changing density of surface low-coordinated sites can control emission fluxes of both crystal components.

DFM investigation show that the erosion of the crystal occurs in layer-by-layer mode. That is realized by creation, growth and coalescence of rectangular pits
 
of monolayer depth in the topmost layer of the crystal. Analyses of DFM images show that density of edges present on the surface is changing periodically - it is maximal when approximately half of the monolayer is removed, whereas it is minimal when integer number of layers have been removed.
 
Certainly, in reality the process is not perfect and after certain time of irradiation three-dimensional surface topography is created and periodicity of low-coordinated sites is lost.

All above information are sufficient to explain an unexpected oscillatory behavior of desorption yields of both constituents in first stages of crystal desorption as measured by quadrupole mass spectrometer.


Measurements of desorption yields in early stages of the process (until few layers are removed) show that apart from the first few seconds signal of both components (halogen and metal atoms) exhibit oscillatory behavior corresponding to periodic changes of surface topography.
 
Both signals are almost in phase, however halogen yield is slightly late. That means that in the surface proximity certain population of accumulated defects is created since Frenkel defects are created in pairs and can recombine in mutual recombination. 
 
As a result, the process can be described as a subtle interplay between concentrations of defects in the bulk directed by density of low-coordinated sites on the surface. 
    

    To read more see:
  • B. Such, P. Czuba, P. Piatkowski, M. Szymonski, AFM Studies of Electron-Stimulated Desorption Process of KBr(001) Surface, Surface Science 451 (2000) 203-207.
  • B. Such, J. Kolodziej, P. Czuba, P. Piatkowski, P. Struski, F. Krok, M. Szymonski, Surface Topography Dependent Desorption of Alkali Halides, Phys. Rev. Lett. 85 (2000) 2621 -2624.
  • B. Such, J Kolodziej, F. Krok, P. Struski, P. Piatkowski, M. Szymonski, Surface topography dependent desorption of sodium chloride, Radiat. Eff. and . Def. in Solids 156 (2001) 69.
  • M. Szymonski, J. Kolodziej, B. Such, P. Piatkowski, P. Struski, P. Czuba, F. Krok, Nano-Scale Modification of IonicSurfaces Induced by Electronic Transitions, Prog. Surf. Sci. 67 (2001) 123.
  • R. Bennewitz, S. Schar, V. Barwich, O. Pfeiffer, E. Meyer, F. Krok, B. Such, J. Kolodzej, M. Szymonski, Atomic-Resolution Images of Radiation Damage in KBr, Surf. Sci. Lett. 474 (2001) L197-L202.
  • J. Kolodziej, B. Such, P. Czuba, F. Krok, P. Piatkowski, P. Struski, M. Szymonski, R. Bennewitz, S. Schar, E. Meyer, Frenkel Defect Interactions at Surfaces of Irradiated Alkali Halides Studied by Noncontact Atomic-Force Microscopy, Surf. Sci. 482-485 (2001) 903.
  • M. Szymonski, P. Struski, A. Siegel, J.J. Kolodziej, B. Such, P.Piatkowski, P. Czuba, F. Krok, Ionic Crystal Decomposition with Light, Acta Physica Polonica B 33 (2002) 2237.
  • M. Szymonski, J. Kolodziej, P. Struski, B. Such, P. Czuba, P. Piatkowski, Nanoscale Studies of Electron-Stimulated Desorption of Alkali Halides, Electron Technology 33 (2000) 369-374.
  • B. Such, J.J. Kolodziej, F. Krok, K. Meisel, P. Czuba, P. Piatkowski, P. Struski, M. Szymonski, Modyfikacja powierzchni w procesie elektronowo stymulowanej desorpcji, Elektronika, materiały II Kongresu PTP, str. 48.
joomla template 1.6