United Microelectronics Corp.

Silicon Motion, Inc.

Macronix International Co., Ltd.

Wen-Chung Chang

Chien-Hung Chen

Ya-Huei Huang

Feng-Ji Tsai

Shen-De Wang

Sung-Bin Lin

Cherng-En Sun

Ke-Feng Lin

Tzu-Ping Chen

Min-Hsuan Tsai

Ming-Yen Liu

Tseng Hsun Liu

Bing-Fei Wu

Chien-Cheng Lin

Shun-Li Lin

Yen-Ting Ho

Cheng-Chih Chien

Jou-Wei Fu

Yun-Chu Lin

Ching-Yi Lee

A method for fabricating a transistor includes providing a substrate, having a gate region and a first trench in the substrate at a first side of the gate region; forming a first gate insulating layer, disposed on a first portion of the gate region, opposite to the first trench; forming a second gate insulating layer, disposed on a second portion of the gate region and a first portion of the first trench abutting to the gate region, wherein the second gate insulating layer is thicker than the first gate insulating layer; forming a gate layer, disposed on the first and second gate insulating layers, having a downward protruding portion corresponding to the first trench; forming a first doped region in the substrate at least under the first trench; and forming a second doped region in the substrate at a second side of the gate region.

Method for fabricating transistor structure

A transistor structure includes a substrate, having a gate region and a first trench in the substate at a first side of the gate region. Further, a first gate insulating layer is disposed on a first portion of the gate region, opposite to the first trench. A second gate insulating layer is disposed on a second portion of the gate region and a first portion of the first trench abutting to the gate region, wherein the second gate insulating layer is thicker than the first gate insulating layer. A gate layer is disposed on the first and second gate insulating layers, having a downward protruding portion corresponding to the first trench. A first doped region is in the substrate at least under the first trench. A second doped region is in the substrate at a second side of the gate region.

Transistor structure and method for fabricating the same

A semiconductor device includes a semiconductor substrate, an isolation structure; a first gate dielectric layer and a first gate electrode. The isolation structure is formed in the semiconductor substrate to divide the semiconductor substrate at least into a first active region and a second active region. The first gate dielectric layer is disposed on the first active region, and has a plane top surface contacting to a sidewall of the isolation structure and forming an acute angle therewith. The first gate electrode stacked on the plane top surface.

Semiconductor device and method for fabricating the same

A trapping gate forming process includes the following. An oxide/nitride/oxide layer is formed on a substrate. A hard mask is formed to cover the oxide/nitride/oxide layer. The hard mask, the oxide/nitride/oxide layer and the substrate are patterned to form at least a trench in the hard mask, the oxide/nitride/oxide layer along a first direction. An isolation structure is formed in the trench. A first gate is formed across the oxide/nitride/oxide layer along a second direction orthogonal to the first direction. A flash cell formed by said process includes a substrate, a first gate and an oxide/nitride/oxide layer. The substrate contains at least an active area extending along a first direction. The first gate is disposed across the active area along a second direction orthogonal to the first direction, thereby intersecting an overlapping area. The oxide/nitride/oxide layer is disposed in the overlapping area between the first gate and the active area.

Trapping gate forming process and flash cell

A method for manufacturing a non-volatile memory with SONOS memory cells, which includes steps of: providing a substrate; forming a first gate oxide layer and a first gate conductive layer onto the substrate; forming a MOS transistor gate by executing a photolithography process on the first gate conductive layer, and then forming an ONO structure on the substrate; and forming a second gate conductive layer on the ONO substrate, and then forming a NVM transistor gate by executing a photolithography process on the second gate conductive layer.

Method of manufacturing non-volatile memory having SONOS memory cells

A method for manufacturing a non-volatile memory structure includes providing a substrate having a memory region and a logic region defined thereon, masking the logic region while forming at least a first gate in the memory region, forming an oxide-nitride-oxide (ONO) structure under the first gate, forming an oxide structure covering the ONO structure on the substrate, masking the memory region while forming a second gate in the logic region, and forming a first spacer on sidewalls of the first gate and a second spacer on sidewalls of the second gate simultaneously.

Non-volatile memory structure and manufacturing method thereof

A method for manufacturing a non-volatile memory structure includes providing a substrate having a gate structure, performing a first oxidation process to form a first SiO layer at least covering a bottom corner of the conductive layer, performing a first etching process to remove the first SiO layer and a portion of the dielectric layer to form a cavity, performing a second oxidation process to form a second SiO layer covering sidewalls of the cavity and a third SiO layer covering a surface of the substrate, forming a first SiN layer filling in the cavity and covering the gate structure on the substrate, and removing a portion of the first SiN layer to form a SiN structure including a foot portion filling in the cavity and an erection portion upwardly extended from the foot portion, and the erection portion covering sidewalls of the gate structure.

Non-volatile memory device with undercut ONO trapping structure and manufacturing method thereof

A method for forming a memory cell structure includes following steps. A substrate including at least a memory cell region defined thereon is provided, and a first gate stack is formed in the memory cell region. A first LDD implantation is performed to form a first LDD at one side of the first gate stack in the memory cell region, and the first LDD includes a first conductivity type. A second LDD implantation is performed to form a second LDD at one side of the first gate stack opposite to the first LDD in the memory cell region, and the second LDD includes the first conductivity type. The first LDD and the second LDD are different from each other.

Memory cell structure and method for forming the same

A method for programming a non-volatile memory cell is described. The memory cell includes a substrate, a gate over the substrate, a charge-trapping structure at least between the substrate and the gate, and first and second S/D regions in the substrate beside the gate. The method includes performing a channel-initiated secondary electron (CHISEL) injection process to inject electrons to the charge-trapping structure.

Method for programming non-volatile memory cell, non-volatile memory array and non-volatile memory apparatus

A manipulating method of a nonvolatile memory is provided and comprises following steps. The nonvolatile memory having a plurality of memory cell is provided. Two adjacent memory cells correspond to one bit and comprise a substrate, a first and another first doping regions, a second doping region, a charge trapping layer, a control gate, a first bit line, a source line and a second bit line different from the first bit line. A first and a second channel are formed. The charge trapping layer is disposed on the first and the second channels. The two adjacent memory cells are programmed by following steps. A first positive and negative voltages are applied to the control gate between the first and the second doping regions and the control gate between the second and the another first doping regions, respectively. A first voltage is applied to the source line.

Nonvolatile memory and manipulating method thereof

A memory apparatus and a method of evenly using the blocks of a flash memory are provided. The memory apparatus comprises a flash memory and a controller. The flash memory comprises a data region with a plurality of data blocks and a spare region with a plurality of spare blocks. The controller is configured to receive data corresponding to the first data block, select a spare block, program data into the spare block when the erase count corresponding to the spare block is less than the predetermined value or to select a second data block and program data stored in the second data block into the spare block when the erased count corresponding to the spare block reaches the predetermined value. As a result, the blocks of the flash memory are used evenly.

Memory apparatus and method of evenly using the blocks of a flash memory

A structure applied to a photolithographic process is provided. The structure includes at least a film layer, an optical isolation layer, an anti-reflection coating and a photoresist layer sequentially formed over a substrate. In the photolithographic process, the optical isolation layer stops light from penetrating down to the film layer. Since the optical isolation layer is set up underneath the photoresist layer, light emitted from a light source during photo-exposure is prevented from reflecting from the substrate surface after passing through the film layer. Thus, the critical dimensions of the photolithographic process are unaffected by any change in the thickness of the film layer.

Structure applied to a photolithographic process

A device for prolonging the lifetime of a nonvolatile memory is applied to the connection between a host electronic machine and a nonvolatile memory device. The device includes a RAM (Random Access Memory) buffer zone, a counter, and two sets of inverters, wherein the RAM buffer zone is employed to store a unit data train temporarily; the counter is used to count the total bits of logic '0'; the inverters are used to lessen the times for readying/writing a nonvolatile memory device by inverting the unit data train based on the count (for example, when the number of logic '&#x201c;s exceeds the number of logic &#x201c;&#x201c;s, and the state flag keeps track of whether an inversion has occurred or is needed.

Growing community of inventors

Hsinchu, Taiwan

Wen-Chung Chang