Nonvolatile Si quantum memory with self-aligned doubly-stacked dots

Synthesis and size differentiation of Ge nanocrystals in amorphous SiO2

“…If the separation between the different sized nanocrystal bands and size distributions can be controlled and reduced to a few nanometers, such a double band formation can be especially important for flash memory applications. In a recent study, Ohba et al [23] has studied the affect of double stacking of two different Si nanocrystal sizes on the retention properties of flash memory devices. They used Si nanocrystals of 5 and 10 nm in diameters, in two separate layers of SiO 2 matrix.…”

Silicon Nanocrystal Flash Memory

“…Accordingly, the retention time of Sample A is estimated to be 4580 h by three orders of magnitude longer than that of Sample B. For a comparison, an increase by two orders of magnitude was observed in a doubly stacked nc-Si dot memory [27].…”

“…Multiple memory nodes have been exampled to improve charge retention time with consuming negligible programming/erasing time [27,28].…”

“…Since the charge leakage between the upper dots and the channel can be suppressed by the energy barrier due to the Coulomb blockade in the lower dot, the charge retention time is improved exponentially by reducing the lower dot size. Meanwhile, the programming/erasing time could be maintained if the tunneling oxide thickness is simultaneously adjusted [27]. A multiple stacked floating gate was demonstrated by Zacharias et al [34].…”

“…The Curves I and II correspond to discharging of Samples A and B, respectively. In general, the retention time is defined as the elapsed time that allows the memory window to become zero [27,37]. Considering that the electron injection rate is usually smaller than the loss rate, one may suppose that the memory window may be zero, when the remaining electrons are 40%.…”