Semiconductor Nanoclusters: Ge vs. Si


  • structures containing interface clusters show a larger programming window
  •  programming already  possible at small programming voltages
  •  retention time is relatively short


  •  for smaller progr. voltages (E < 4 MVcm-1) memory effect relatively small
  •  prog. window at higher fields comparable to Ge based structures
  •  good retention characteristics
  •  microstructural properties are still under investigation

still open questions / ongoing investigations:

  •  Where exactly is the charge stored ? (bulk or interface clusters / traps)
  •  Can we adjust the interface cluster band ?
  •  How does the microstructure of the Si implanted layers look like ?