Browse Prior Art Database

Vacuum Sealed Silicon Rich Oxide EEPROM Cell

IP.com Disclosure Number: IPCOM000122212D
Original Publication Date: 1991-Nov-01
Included in the Prior Art Database: 2005-Apr-04
Document File: 4 page(s) / 147K

Publishing Venue

IBM

Related People

Acovic, A: AUTHOR [+3]

Abstract

Disclosed is an EEPROM cell using electron transport in a vacuum between silicon-rich-oxide injectors on the control and the floating gates. Since vacuum is used instead of SiO2, the endurance and retention of the cell is very high, making it a candidate for a true non-volatile RAM cell. Use of SRO injectors and of vacuum allows the lowering of the programming voltages.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 50% of the total text.

Vacuum Sealed Silicon Rich Oxide EEPROM Cell

      Disclosed is an EEPROM cell using electron transport in a
vacuum between silicon-rich-oxide injectors on the control and the
floating gates. Since vacuum is used instead of SiO2, the endurance
and retention of the cell is very high, making it a candidate for a
true non-volatile RAM cell.  Use of SRO injectors and of vacuum
allows the lowering of the programming voltages.

      In EEPROM cells, electrons are transported to and from the
floating gate through a thin layer of silicon dioxide [1]. However,
trap- ping in this layer reduces slowly but steadily the difference
between threshold voltage of the programmed and erased state of the
cell, making it unusable after less than 106 write/erase cycles.
Furthermore, some of the carriers may leak from the floating gate
through the SiO2 layer, reducing the data retention time.  Finally,
in all EEPROM cells, it is desirable to increase the coupling of the
floating gate either to the substrate or to the control gate to
decrease the programming voltages.

      The disclosed cell is very similar to the standard EEPROM cell
using silicon-rich-oxide (SRO) injectors [1], with the important
difference that vacuum is used as the dielectric between the floating
gate and the control gate (Fig. 1).  This has numerous advantages:
      Since electron transport occurs only in vacuum, and since there
is no trapping in vacuum, there is no modification of the injection
efficiency during the lifetime of the cell.  The charges trapped in
the silicon injectors can leak away, and hence do not constitute a
permanent damage. Therefore, this cell has, in principle, unlimited
endurance, and can be used as a true non- volatile RAM cell.
      The use of silicon-rich-oxide on the top of the floating gate
and on the bottom of the control gate, allows the reduction of the
voltage necessary to inject electrons into the vacuum [2] (the
barrier for injection into vacuum is about 1eV higher than for
injection into SiO2), and makes the cell compatible with modern
MOSFETs circuits.
      The coupling of the floating gate to the silicon substrate is
greatly increased, since the floating gate is coupled to the control
gate by a thin vacuum layer whose permittivity is four times lower
than the SiO2 permittivity.  Hence, the field between the floating
gate and the control gate is enhanced, so that lower programming
voltage can be used. In addition, since the coupling of the floating
gate to the channel of the MOSFET is excellent, the amount of charge
necessary to produce a given change of threshold voltage is reduced,
decreasing the programming/erase time, and increasing reliability
even more. Leakage of charges trapped in the floating gate to the
control gate is greatly reduced by a perfect insulator (vacuum) on
five out of six sides of the floating gate.
      "Normal" SiO2 is used as the gate oxide on the top of the
channel of the MOSFET, so tha...