Advanced Micro Devices Corporation

Advance Micro Devices, Inc.

Maria Chow Chan

Michael K Templeton

Kathleen R Early

Hao Fang

Nicholas H Tripsas

Mark S Chang

Minh Van Ngo

Mark T Ramsbey

Suzette Keefe Pangrle

William George En

Olov B Karlsson

Lewis N Shen

King Wai Kelwin Ko

Angela Tai Hui

Christopher F Lyons

Yu Nan Sun

Tuan Duc Pham

Lu You

Jeffrey Allan Shields

Mike Templeton

In a two-step spacer fabrication process for a non-volatile memory device, a thin oxide layer is deposited on a wafer substrate leaving a gap in the core of the non-volatile memory device. Implantation and/or oxide-nitride-oxide removal can be accomplished through this gap. After implantation, a second spacer is deposited. After the second spacer deposition, a periphery spacer etch is performed. By the above method, a spacer is formed.

Dual spacer process for non-volatile memory devices

In a first aspect of the present invention, a flash memory array is disclosed. The flash memory array comprises a substrate comprising active regions, wherein the active regions are defined by a layer of nitride, the layer of nitride including a top surface. The flash memory array further comprises shallow trenches in the substrate, each of the shallow trenches including a layer of oxide, the layer of oxide having a top surface, wherein the top surface of the layer of oxide and the top surface of the layer of nitride are on substantially the same plane and channel areas wherein the occurrences of polyl stringers in the channel areas is substantially reduced. In a second aspect of the present invention, a method and system for fabricating a flash memory array is disclosed. The method comprises the steps of providing a layer of nitride over a substrate, forming trenches in the substrate and then growing a layer of oxide in the trenches. Finally, the layer of oxide is polished back. Through the use of the preferred embodiment of the present invention, a shallow trench isolation process is implemented as opposed to LOCOS process, thereby reducing the occurrence of polyl stringers in the channel area. Accordingly, the occurrence of unwanted electrical shorting paths between the adjacent regions is substantially reduced.

Flash memory array and a method and system of fabrication thereof

A method and system for providing an alignment mark for a thin layer in a semiconductor device is disclosed. The semiconductor device includes at least one alternative part having a first thickness greater than a second thickness of the thin layer. The method and system include providing the thin layer and providing the alignment mark for the thin layer in the at least one alternative part. The alignment mark has a depth that is greater than the second thickness of the thin layer. In one aspect, the method and system include providing a mask for the thin layer. The mask includes an alignment mark portion that covers the at least one alternative part and that is for providing the alignment mark. In this aspect, the method and system also include removing a portion of the at least one alternative part to provide the alignment mark in the at least one alternative part.

Method and system for providing a robust alignment mark at thin oxide layers

A method for plasma etching of silicon nitride using a mixture of trifluoromethane and oxygen in a ratio of approximately 8 to 1 to selectively etch silicon nitride in preference to silicon dioxide and photoresist, resulting in critical dimension gain.

Silicon nitride etch process with critical dimension gain

A method and system for providing a plurality of lines in a semiconductor memory device is disclosed. The method and system include providing a semiconductor substrate, providing a plurality of lines and providing an adjacent feature. The plurality of lines includes an adjacent line adjacent to the adjacent feature. The each of the plurality of lines has a line width that is substantially the same for each of the plurality of lines. The plurality of lines is preferably formed utilizing a mask to print a physical mask for the plurality of lines and the adjacent feature. The mask includes a mask assist feature between at least a first polygon for the adjacent line and at least a second polygon for the adjacent feature. The mask assist feature has a size that is sufficiently large to affect the width of the adjacent line and that is sufficiently small to prevent a corresponding feature from being printed on the physical mask. The method and system also preferably include removing a second portion of the layer of material exposed by the pattern of the physical mask to form the plurality of lines.

Method for improved control of lines adjacent to a select gate using a mask assist feature

A method for fabricating a first memory cell and a second memory cell electrically isolated from each other is provided. A first polysilicon (poly I) layer is formed on an oxide coated substrate. Then, a sacrificial oxide layer and nitride layer are formed for masking the poly I layer. At least a portion of the masking layer is etched to pattern the first memory cell and the second memory cell and an unmasked portion therebetween. The unmasked portion of the poly I layer is transformed into an insulator via thermal oxidation such that the insulator separates a floating gate of the first memory cell from a floating gate of the second memory cell. The insulator is etched so as to form a gap having gradually sloping sidewalls between a floating gate of the first memory cell and a floating gate of the second memory cell, the gap isolating the floating gate of the first memory cell from the floating gate of the second memory cell. Thereafter, an interpoly dielectric layer and a second polysilicon (poly II) layer are formed substantially free of abrupt changes in step height.

Memory cell structure for elimination of oxynitride (ONO) etch residue and polysilicon stringers

In a first aspect of the present invention, a semiconductor device is disclosed. The semiconductor device comprises at least two gate stacks, each gate stack having two sides and oxide spacers on each of the two sides of each of the at least two gate stacks, wherein at least one of the oxide spacers is triangular shaped. In a second aspect of the present invention, a method and system for processing a semiconductor device is disclosed. The method and system for processing a semiconductor comprise forming at least two gate stacks over a semiconductor substrate, depositing an oxide layer over the at least two gate stacks, and etching the oxide layer to form at least one oxide spacer in between the at least two gate stacks, wherein the at least one oxide spacer is triangular shape. Through the use the present invention, the voids that are created in the semiconductor device during conventional semiconductor processing are eliminated. This is accomplished by creating oxide spacers having a triangular shape when etching the oxide layer to form the oxide spacer. By creating a triangular shaped oxide spacer, subsequent layers of material can be deposited over the oxide spacer without creating voids in the semiconductor device. Accordingly, as a result of the use of the present invention, the oxide spacers are strengthened, which increases the reliability of the semiconductor device.

Method and system for processing a semiconductor device

In the manufacture of sub-0.35 micron semiconductors using deep ultraviolet lithography, a bilayer of silicon dioxide on top of silicon oxynitride is used as bottom anti-reflective coating and an etch hard mask for photolithographic resist. Since the silicon dioxide is optically transparent at the deep ultraviolet wavelengths being used (248 nm), its thickness in combination with a preselected reflective silicon oxynitride thickness satisfies the zero reflectivity goal and, at the same time, is adequately thick to serve as a hard mask for self-aligned etch and self-aligned-source etch.

Bilayer anti-reflective coating and etch hard mask

A method and system for fabricating a flash memory array comprising a core area and a periphery area is disclosed. The method and system comprises depositing a layer of poly2 over the core area and the periphery area, selectively etching the core area, and selectively etching the poly2 only in the periphery area wherein the occurrence of stringers is reduced. Through the use of the preferred embodiment of the present invention, the core and periphery areas are etched separately after the deposition of the poly2, thereby reducing the occurrence of stringers at the core/periphery interface. Accordingly, the occurrence of unwanted electrical shorting paths between the adjacent transistors is substantially reduced.

Method and system for fabricating a flash memory array

In the manufacture of sub 0.35 micron semiconductors using deep ultraviolet lithography, a bilayer of silicon dioxide on top of silicon oxynitride is used as bottom anti-reflective coating and an etch hard mask for photolithographic resist. Since the silicon dioxide is optically transparent at the deep ultraviolet wavelengths being used (248 nm), its thickness in combination with a preselected reflective silicon oxynitride thickness satisfies the zero reflectivity goal and, at the same time, is adequately thick to serve as a hard mask for self-aligned etch and self-aligned-source etch.

Elimination of oxynitride (ONO) etch residue and polysilicon stringers

A method is provided for removing an bottom anti-reflective coating (BARC) from a transistor gate following at least one etch back process associated with a spacer formation and/or subsequent resistor protect etching process or processes. The method eliminates the need to use HF acid in the stripping process by substantially reducing the thickness of the BARC during each of the etching back processes, such that, only a thin layer of BARC material remains that can be easily removed with phosphoric acid.

Methods for removal of an anti-reflective coating following a resist

A method of preventing poly stringers in the formation of a memory device includes the steps of forming at least one field oxide region in a substrate and forming a tunnel oxide over the substrate. A first polysilicon layer is then formed over the tunnel oxide and a poly mask having a mask profile is formed over the first polysilicon layer. The first polysilicon layer is then etched in portions exposed by the poly mask, thereby creating a first polysilicon layer etch profile, wherein the first polysilicon layer etch profile is substantially ideally anisotropic and independent of the mask profile. An insulating layer and a conductive layer is then formed over the etched first polysilicon layer and portions of the conductive layer are then etched to form word lines. The insulating layer is then etched in regions adjacent the word lines, thereby leaving a substantially vertical insulative fence along the first polysilicon layer etch profile. Lastly, the first polysilicon layer is again etched in regions adjacent the word lines, wherein the etching of the first polysilicon layer is substantially anisotropic and removes substantially all the polysilicon in the regions adjacent the word lines, resulting in the prevention of poly stringers.

Elimination of poly stringers with straight poly profile

A method for fabricating a first memory cell and a second memory cell electrically isolated from each other. A first polysilicon (poly I) layer is formed on an oxide coated substrate. A masking layer is deposited or grown on the poly I layer, and at least a portion of the masking layer is etched so as to pattern the first memory cell and the second memory cell and an unmasked portion therebetween. The unmasked portion of the poly I layer is transformed into an insulator via thermal oxidation such that the insulator electrically isolates a floating gate of the first memory cell from a floating gate of the second memory cell. An interpoly dielectric layer and a second polysilicon (poly II) layer is formed over the poly I layer and insulator substantially free of abrupt changes in step height.

A method for fabricating a first memory cell and a second memory cell having floating gates electrically isolated from each other. A first polysilicon (poly I) layer is formed on an oxide coated substrate, portions of the poly I layer to serve as future floating gates for the first and second memory cells. An interpoly dielectric layer is formed over the poly I layer. At least a portion of the interpoly dielectric layer is etched to expose at least a portion of the poly I layer so as to pattern the floating gates on either side of the exposed portion of the poly I layer. The exposed portion of the poly I layer is transformed into an insulator via thermal oxidation such that the insulator electrically isolates a floating gate of the first memory cell from a floating gate of the second memory cell. A second polysilicon (poly II) layer is formed substantially free of abrupt changes in step height.

Growing community of inventors

San Jose, CA, United States of America

Maria Chow Chan