Splitgate Ferroelectric Transistor

We propose to develop a new nonvolatile Ferroelectric Random Access Memory (FeRAM) device with a novel 1T gate cell structure for very high-density memories. This gate design will allow a ferroelectric memory to rival stacked-gate FLASH cell in density butwithout the drawbacks of floating gate technology, e.g. early wear-out and slow write times. For this 1T split-gate cell we will use a SMI newly discovered cerium ferroelectric thin film. This film retains its ferroelectric properties when depositieddirectly on silicon and will allow for MOS channel control based directly on the ferroelectric polarization state. The split-gate architechture allows for a relatively simple, complete high-density 1T cell structure for maximum density. The split-gateferroelectric transistor structure described in this proposal offers a much simpler fabrication process for a 1T memory cell versus a stacked gate FLASH cell (current commercial standard for high-density memory). The high speed switching and high enduranceproperties of the ferroelectric will also expand the market of nonvolatile memories with the potential to replace not only FLASH, but also DRAM and SRAM.