Acer Inc.

Taiwan Semiconductor Manufacturing Comp. Ltd.

Industrial Technology Research Institute

National Sun Yat-sen University of Kaohsiung

Fu-Yen Jian

Ting-Chang Chang

Yong-En Syu

Te-Chih Chen

Shih-Ching Chen

Ying-Lang Wang

Shih-Chieh Chang

Kuan-Chang Chang

Min-Chen Chen

Wen-Hung Lo

Po-Chun Yang

Yu-Shih Lin

Chia-Sheng Lin

Chih-Hao Dai

Hung-Wei Li

Methods for a resistive random access memory (RRAM) device are disclosed. A bottom electrode is formed over a substrate. A top electrode is formed over the bottom electrode. A resistive switching layer is formed interposed between the top electrode and the bottom electrode. The resistive switching is made of a composite of a metal, Si, and O, formed by oxidation of a metal silicide of a metal, co-deposition of the metal and silicon in oxygen ambiance, co-deposition of a metal oxide of the metal and silicon, or co-deposition of a metal oxide of the metal and silicon oxide. There may be an additional tunnel barrier layer between the top electrode and the bottom electrode. The top electrode and the bottom electrode may comprise multiple sub-layers.

Methods for resistive random access memory (RRAM)

Methods and apparatuses for a resistive random access memory (RRAM) device are disclosed. The RRAM device comprises a bottom electrode, a resistive switching layer disposed on the bottom electrode, and a top electrode disposed on the resistive switching layer. The resistive switching layer is made of a composite of a metal, Si, and O. There may be an additional tunnel barrier layer between the top electrode and the bottom electrode. The top electrode and the bottom electrode may comprise multiple sub-layers.

Methods and apparatus for resistive random access memory (RRAM)

A resistive random access memory (RRAM), a controlling method for the RRAM, and a manufacturing method therefor are provided. The RRAM includes a first electrode layer; a resistance switching layer disposed on the first electrode layer; a diffusion metal layer disposed on the resistance switching layer; and a second electrode layer disposed on the diffusion metal layer, wherein at least one extension electrode is disposed in the resistance switching layer.

Resistive random access memory, controlling method and manufacturing method therefor

A method for performing a programming operation to a first memory bit and a second memory bit of a device is described. The method includes applying a pulse train voltage to a metal gate of the device and grounding a substrate of the device. By floating/grounding a drain of the device and/or by floating/grounding the source of the device, the first memory and the second memory bit are programmed. The pulse train voltage includes 10 to 1000 pulses. One pulse includes a peak voltage and a base voltage. The peak voltage ranges from 0.5 V to 10 V. A duration of the peak voltage ranges from 1 nanosecond to 1 millisecond. The base voltage is 0 V. A duration of the base voltage ranges from 1 nanosecond to 1 millisecond.

MOSFET having memory characteristics

A resistance random access memory element includes a first electrode, an insulating layer, a diffusing metal layer, and a second electrode superimposed in sequence. The insulating layer includes a plurality of pointed electrodes. A method for making a resistance random access memory element includes growing and forming an insulating layer on a surface of a first electrode. A diffusing metal layer is formed on a surface of the insulating layer. A second electrode is mounted on a surface of the diffusing metal layer. A negative pole and a positive pole of a driving voltage are connected with the first and second electrodes, respectively. The diffusing metal in the diffusing metal layer is oxidized into metal ions by the driving voltage. The metal ions are driven into the insulating layer and form a plurality of pointed electrodes after reduction.

Resistance random access memory element and method for making the same

There is a method for enabling a SONOS transistor to be used as both a switch and a memory. FN tunneling is carried out through the source or drain of the transistor, so as to further change the state of electrons stored in an upper charge storage layer adjacent to the drain or source, and the variation in gate-induced drain leakage is used to recognize the memory state of the drain and source. A stable threshold voltage of the transistor is always maintained during this operation. The present invention enables one single transistor having dual features of switch and memory, while being provided with a two-bit memory effect, thus providing a higher memory density in comparison with a general transistor.

Method for enabling a SONOS transistor to be used as both a switch and a memory

One embodiment of the present invention provides an operation method of a memory device. The memory device includes a source, a drain, and a channel region between the source and the drain, a gate dielectric with a charge storage layer on the channel region, and a gate on the gate dielectric, wherein the source, the drain and the channel region are located in a substrate. The operation method includes the following steps: applying a reverse bias between the gate and the drain of the memory device to generate band-to-band hot holes in the substrate near the drain; injecting the band-to-band hot holes to a drain side of the charge storage layer; and performing a program/erase operation upon the memory device. The band-to-band hot holes in the drain side of the charge storage layer are not completely vanished by the program/erase operation.

Operation method of memory device

A method for operating a memory device is provided. In accordance with the method, the charges are stored in a source storage region, a drain storage region, and a channel storage region of a charge storage layer which respectively correspond to a source, a drain, and a channel of a SONOS transistor, thereby achieving 3-bit information storage in one cell. The channel storage region is programmed and erased by FN tunneling. Both of the source storage region and the drain storage region are programmed by channel hot electrons and erased by source-side or drain-side FN tunneling. The present invention can store three-bit data per cell, such that the storage density of the memory device can be substantially increased.

The present invention relates to a nonvolatile memory apparatus and a method of using a thin film transistor (TFT) as a nonvolatile memory by storing carriers in a body of the TFT, which operates a general TFT as a memory cell of a nonvolatile memory by manipulating the electrical characteristics of the TFT in order to integrate with other electrical components formed by TFTs, such as logic circuit or TFT-LCD pixel transistor, on the LCD panel without additional semiconductor manufacturing processes.

Nonvolatile memory apparatus and method of using thin film transistor as nonvolatile memory

The present invention relates to an erasing method for nonvolatile memory, which uses forward bias between the source/drain region and body contact to inject majority carriers into the body, and then accelerates the majority carriers by an electric field between the body and the gate to energize the majority carriers to overcome the oxide barrier and to erase the nonvolatile memory.

Erasing method for nonvolatile memory

A method for programming a nonvolatile memory includes applying at least a voltage to a source or a drain, so as to inject carriers of the source or drain into a substrate; applying a third voltage to a gate or the substrate, so that the carriers which are in the substrate having enough energy can surmount an oxide layer to reach a charge storage device.

Growing community of inventors

Kaohsiung, Taiwan

Fu-Yen Jian