摘要:
A fabrication method for a memory device with a floating gate is provided. A substrate is provided. A channel doping step is performed on the substrate, wherein the actual threshold voltage of the subsequently formed memory device becomes greater than the preset threshold voltage. A stack gate and source/drain regions are then sequentially formed on the substrate to complete the formation of the memory device. The drain-turn-on leakage is prevented by an increase of the actual threshold voltage.
摘要:
A NROM cell for reducing for reducing the second-bit effect is described. The NORM cell of the present invention is formed with a substrate, a silicon oxide/silicon nitride/silicon oxide (ONO) layer disposed on the substrate, a gate disposed on the silicon oxide/silicon nitride/silicon oxide layer, source/drain regions configured in the substrate beside the gate, and a shallow pocket doped region configured between the source/drain regions and the ONO layer beside the gate. The depth of the shallow pocket doped region is sufficiently small to prevent interference to the current flow that travels to the source/drain regions.
摘要:
A fabrication method for a flash memory device with a split floating gate is described. The method provides a substrate, wherein an oxide layer and a patterned sacrificial layer are sequentially formed on the substrate. Ion implantation is then conducted to form source/drain regions with lightly doped source/drain regions in the substrate beside the sides of the patterned sacrificial layer using the patterned sacrificial layer as a mask. Isotropic etching is further conducted to remove a part of the patterned sacrificial layer, followed by forming two conductive spacers on the sidewalls of the patterned sacrificial layer. The patterned sacrificial layer and oxide layer that is exposed by the two conductive spacers are then removed to form two floating gates. Subsequently, a dielectric layer and a control gate are formed on the substrate.
摘要:
A method of fabricating a flash memory is disclosed. The method begins a stacked gate on the substrate. A shallow junction doping is performed on a substrate having a stacked gate already formed thereon, with the stacked gate serving as a mask, so as to form a shallow junction doped region in the substrate adjacent to both sides of the stacked gate. A mask layer is formed on the substrate to cover a top surface and sidewalls of the stacked gate, while exposing portions of the shallow junction doped region. With the mask layer serving as a mask, a deep junction doping is performed on the substrate to form a deep junction doped region in the substrate adjacent to both sides of the mask layer. After the mask layer is removed, a thermal process is performed to form a source/drain region having both the shallow junction doped region and deep junction doped region.
摘要:
A method of programming the memory cell comprises setting the memory cell to an initial state of a first gate threshold voltage, performing a processing sequence including: applying a voltage bias between the gate and the first junction region to cause electric hole to migrate towards and be retained in the trapping layer, and evaluating a read current generated in response to the voltage bias to determine whether a second gate threshold voltage is reached, wherein the second gate threshold voltage is lower than the first gate threshold voltage. The processing sequence is repeated a number of times by varying one or more time the voltage bias between the gate and the first junction region until the second gate threshold voltage is reached and the memory cell is in a program state.
摘要:
A method of operating a non-volatile memory cell, wherein the non-volatile memory cell includes a word line, a first bit line, and a second bit line, the method includes programming the memory cell that includes applying a high positive bias to the first bit line, applying a ground bias to the second bit line, and applying a high negative bias to the word line, wherein positively-charged holes tunnel through the dielectric layer into the trapping layer.
摘要:
A method of a read scheme for a non-volatile memory cell. The non-volatile memory cell includes a substrate, a source, a drain and a gate above a channel separated by a nonconducting charge trapping material sandwiched between first and second insulating layers. The method applies a first positive drain-to-source bias, a second positive source-to-substrate bias, and a third positive gate-to-source bias to read the source-side charges trapped in the trapping material near the source side.
摘要:
A virtual ground nonvolatile memory cell array is formed by a plurality of adjacent nonvolatile memory cells arranged in rows and columns so as to form an array. Each of the nonvolatile memory cells is formed by an N channel MOSFET with a trapping layer formed between two isolating layers. In the erase state, the trapping layer stores an amount of electrons. A method for programming the virtual ground nonvolatile memory cell array is also disclosed. The potentials applied to the bitlines and wordlines in the array are preset to program nonvolatile memory cells and not to disturb cells adjacent to the nonvolatile memory cell to be programmed.
摘要:
A non-volatile memory device is described, comprising a plurality of memory cells, a plurality of word lines, a plurality of drain lines, and a plurality of source lines, wherein two adjacent memory cells in a column constitute a cell pair, and all cell pairs are arranged in rows and columns. The two memory cells in each cell pair share a source region, and two adjacent cell pairs in a column share a drain region. The source regions and the gates of the memory cells in the same row are coupled to a source line and a word line, respectively, and the drain regions of the memory cells in the same column are coupled to a drain line.
摘要:
A non-volatile memory is described, which comprises a plurality of memory cells, a plurality of word lines, a plurality of drain lines and a plurality of source lines. Two adjacent memory cells in the same row share a source and are grouped into a cell pair, and all of the cell pairs are arranged in rows and columns, wherein two cell pairs in the same row share a drain. The sources of the memory cells in the same row are connected to a source line, and the drains of the memory cells in the same row are connected to a drain line. The gates of the memory cells in the same column are coupled to a word line.