# US 2022/0285364 A1 · Methods and apparatus for hierarchical bitline for three-dimensional dynamic random-access memory

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2022/0285364 A1
- Document ID: US20220285364A1
- Status: Published application
- Kind: A1
- Application number: 17/680,993
- Filed: 2022-02-25
- Published: 2022-09-08
- Earliest claimed: 2021-03-05
- Assignees: Applied Materials, Inc.
- Applicants: Applied Materials, Inc.
- Inventors: Fredrick David FISHBURN
- Web page: https://patentgrep.com/patent/US20220285364A1

## CPC

- H10B 12/30: {DRAM devices comprising one-transistor - one-capacitor [1T-1C] memory cells}
- H10B 12/482: {Bit lines}
- H10B 12/05: {Making the transistor}
- H10B 12/488: {Word lines}
- H10B: ELECTRONIC MEMORY DEVICES

## Abstract

Methods for forming 3D DRAM leverage L-pad formations to increase memory density. Methods may include etching a substrate to form two Si walls oriented parallel to each other and forming a space therebetween, depositing a plurality of alternating Si layers and SiGe layers using epitaxial growth processes to form horizontal deposition layers on the space between the two Si walls and vertical deposition layers on sidewalls of the two Si walls, depositing a CMP stop layer on the substrate, planarizing the substrate to the CMP stop layer, removing a portion of a top of the two Si walls and forming an L-pad formation, deep etching a pattern of holes into the space between the two Si walls in horizontal portions of the plurality of alternating Si layers and SiGe layers, and forming vertical wordline structures from the pattern of holes in the horizontal portions.

## Claims

### Claim 1

A method of forming structures for three-dimensional (3D) dynamic random-access memory (DRAM), comprising:
etching a substrate to form at least two silicon walls oriented parallel to each other and forming a space therebetween, wherein the at least two silicon walls have a first height;
depositing a plurality of alternating crystalline silicon layers and crystalline silicon germanium layers using epitaxial growth processes to form horizontal deposition layers on the space between the at least two silicon walls and vertical deposition layers on sidewalls of the at least two silicon walls;
depositing a chemical mechanical (CMP) stop layer on the substrate; and
planarizing the substrate to the CMP stop layer, removing a portion of a top of the at least two silicon walls and forming an L-pad formation, wherein the at least two silicon walls have a second height after planarizing that is less than the first height.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
deep etching a pattern of holes into the space between the at least two silicon walls in horizontal portions of the plurality of alternating crystalline silicon layers and crystalline silicon germanium layers; and
forming vertical wordline structures from the pattern of holes in the horizontal portions.

### Claim 3 (depends on 2)

The method of claim 2, further comprising:
etching at least two trenches perpendicular to one of the at least two silicon walls, the at least two trenches electrically isolating a global bitline access area formed in the L-pad formation, the global bitline access area configured to provide access to a portion of the plurality of alternating crystalline silicon layers and crystalline silicon germanium layers used to form the vertical wordline structures; and
replacing the crystalline silicon germanium layers of the global bitline access area with dielectric material and replacing the crystalline silicon layers of the global bitline access area with metal material to form metal bitlines.

### Claim 4 (depends on 3)

The method of claim 3, further comprising:
forming at least one global bitline perpendicular to the global bitline access area and distal to the one of the at least two silicon walls and adjacent to the vertical wordline structures, each one of the at least one global bitline electrically connected to one of at least one of the metal bitlines of the global bitline access area.

### Claim 5 (depends on 4)

The method of claim 4, further comprising:
forming at least one local bitline switch from at least one vertical wordline structure, the at least one local bitline switch configured to electrically connect and disconnect the at least one global bitline to local bitline interfacing with the vertical wordline structures.

### Claim 6 (depends on 4)

The method of claim 4, further comprising:
forming a bitline equilibrate adjacent to the vertical wordline structures but distal to the at least one global bitline; and
forming at least one bitline equilibrate switch from at least one vertical wordline structure, the at least one bitline equilibrate switch configured to electrically connect and disconnect local bitlines interfacing with the vertical wordline structures.

### Claim 7 (depends on 2)

The method of claim 2, further comprising:
selectively etching vertical portions of the crystalline silicon germanium layers of the L-pad formation to remove a portion of the crystalline silicon germanium layers between vertical portions of the crystalline silicon layers;
selectively etching vertical portions of the crystalline silicon layers to reduce a thickness of the vertical portions of the crystalline silicon layers to form fins for at least one fin field-effect transistor;
forming shallow trench isolation areas in the vertical portions of the crystalline silicon layers and the crystalline silicon germanium layers for the fin field-effect transistor; and
forming nitride plugs in the shallow trench isolation areas for the fin field-effect transistor.

### Claim 8 (depends on 1)

The method of claim 1, wherein a first thickness of the at least two silicon walls is approximately 2 microns to approximately 3 microns.

### Claim 9 (depends on 1)

The method of claim 1, wherein the first height of the at least two silicon walls is approximately 2 microns to approximately 20 microns.

### Claim 10 (depends on 9)

The method of claim 9, wherein the first height of the at least two silicon walls is approximately 10 microns.

### Claim 11 (depends on 1)

The method of claim 1, wherein the space between the at least two silicon walls is approximately twice a length of a local bitline.

### Claim 12 (depends on 11)

The method of claim 11, wherein the space between the at least two silicon walls is approximately 60 microns.

### Claim 13

A method of forming structures for three-dimensional (3D) dynamic random-access memory (DRAM) comprising:
etching a substrate to form at least two silicon walls oriented parallel to each other and forming a space therebetween, wherein the at least two silicon walls have a first height;
conformally depositing a plurality of alternating crystalline silicon layers and crystalline silicon germanium layers using epitaxial growth processes to form horizontal deposition layers on the space between the at least two silicon walls and vertical deposition layers on sidewalls of the at least two silicon walls;
depositing a chemical mechanical (CMP) stop layer on the substrate;
planarizing the substrate to the CMP stop layer, removing a portion of a top of the at least two silicon walls and forming an L-pad formation, wherein the at least two silicon walls have a second height after planarizing that is less than the first height;
selectively etching vertical portions of the crystalline silicon germanium layers of the L-pad formation to remove a portion of the crystalline silicon germanium layers between vertical portions of the crystalline silicon layers;
selectively etching vertical portions of the crystalline silicon layers to reduce a thickness of the vertical portions of the crystalline silicon layers to form fins for at least one fin field-effect transistor;
forming shallow trench isolation areas in the vertical portions of the crystalline silicon layers and the crystalline silicon germanium layers for the fin field-effect transistor; and
forming nitride plugs in the shallow trench isolation areas for the fin field-effect transistor.

### Claim 14 (depends on 13)

The method of claim 13, wherein the thickness of the fins is approximately 10 nm to approximately 12 nm.

### Claim 15 (depends on 13)

The method of claim 13, further comprising:
depositing a first oxide layer on the crystalline silicon layers and the crystalline silicon germanium layers after selectively etching vertical portions of the crystalline silicon layers;
densifying the first oxide layer;
forming the shallow trench isolation areas by etching a least one portion of the first oxide layer to expose at least one end of a vertical crystalline silicon layer and etching the vertical crystalline silicon layer and adjacent vertical silicon germanium layers; and
depositing a second oxide layer in the shallow trench isolation areas to partially fill the shallow trench isolation areas.

### Claim 16 (depends on 15)

The method of claim 15, wherein a combined thickness of the first oxide layer and the second oxide layer in the shallow isolation areas is approximately 6 nm.

### Claim 17

A non-transitory, computer readable medium having instructions stored thereon that, when executed, cause a method of forming structures for three-dimensional (3D) dynamic random-access memory (DRAM) to be performed, the method comprising:
etching a substrate to form at least two silicon walls oriented parallel to each other and forming a space therebetween, wherein the at least two silicon walls have a first height;
depositing a plurality of alternating crystalline silicon layers and crystalline silicon germanium layers using epitaxial growth processes to form horizontal deposition layers on the space between the at least two silicon walls and vertical deposition layers on sidewalls of the at least two silicon walls;
depositing a chemical mechanical (CMP) stop layer on the substrate; and
planarizing the substrate to the CMP stop layer, removing a portion of a top of the at least two silicon walls and forming an L-pad formation, wherein the at least two silicon walls have a second height after planarizing that is less than the first height.

### Claim 18 (depends on 17)

The non-transitory, computer readable medium of claim 17, the method further comprising:
deep etching a pattern of holes into the space between the at least two silicon walls in horizontal portions of the plurality of alternating crystalline silicon layers and crystalline silicon germanium layers; and
forming vertical wordline structures from the pattern of holes in the horizontal portions.

### Claim 19 (depends on 18)

The non-transitory, computer readable medium of claim 18, the method further comprising:
etching at least two trenches perpendicular to one of the at least two silicon walls, the at least two trenches electrically isolating a global bitline access area formed in the L-pad formation, the global bitline access area configured to provide access to a portion of the plurality of alternating crystalline silicon layers and crystalline silicon germanium layers used to form the vertical wordline structures;
replacing the crystalline silicon germanium layers of the global bitline access area with dielectric material and replacing the crystalline silicon layers of the global bitline access area with metal material to form metal bitlines;
forming at least one global bitline perpendicular to the global bitline access area and distal to the one of the at least two silicon walls and adjacent to the vertical wordline structures, each one of the at least one global bitline electrically connected to one of at least one of the metal bitlines of the global bitline access area;
forming at least one local bitline switch from at least one vertical wordline structure, the at least one local bitline switch configured to electrically connect and disconnect the at least one global bitline to local bitline interfacing with the vertical wordline structures;
forming a bitline equilibrate adjacent to the vertical wordline structures but distal to the at least one global bitline; and
forming at least one bitline equilibrate switch from at least one vertical wordline structure, the at least one bitline equilibrate switch configured to electrically connect and disconnect local bitlines interfacing with the vertical wordline structures.

### Claim 20 (depends on 17)

The non-transitory, computer readable medium of claim 17, the method further comprising:
selectively etching vertical portions of the crystalline silicon germanium layers of the L-pad formation to remove a portion of the crystalline silicon germanium layers between vertical portions of the crystalline silicon layers;
selectively etching vertical portions of the crystalline silicon layers to reduce a thickness of the vertical portions of the crystalline silicon layers to form fins for at least one fin field-effect transistor;
forming shallow trench isolation areas in the vertical portions of the crystalline silicon layers and the crystalline silicon germanium layers for the fin field-effect transistor; and
forming nitride plugs in the shallow trench isolation areas for the fin field-effect transistor.
