3D Ferroelectric Memory Stacking for DRAM Density Limits

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current DRAM technology faces limitations in storage density, power consumption, and scalability, as it is nearing the end of its lateral scaling roadmap and cannot be easily stacked into 3D architectures, requiring a more efficient and denser memory solution.

Innovation Solution

The development of three-dimensional ferroelectric memory structures using ferroelectric memory cells with a ferroelectric layer, such as hafnium oxide, which enables non-volatile storage and rapid switching speeds, allowing for the creation of dense, low-latency memory structures like FeFETs and FTJs in various architectures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If conventional DRAM technology is used, then fast data transfer performance is achieved, but storage density is limited and power consumption is relatively high

Engineering Contradiction:
Improvestorage densityVSAvoidpower consumption
Core Design Contradiction:
Quantity of substanceVSUse of energy by moving object

Solution Approach 1:

The patent transitions from two-dimensional lateral scaling to three-dimensional vertical stacking by arranging memory cells in a 3D configuration with bit lines extending in first directions and word lines in second directions perpendicular to the bit lines. This dimensional change enables significantly higher storage density while maintaining fast access speeds and reducing power consumption per bit.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Quantity of substance

If lateral scaling of DRAM is continued, then storage capacity increases, but scalability is reached and further improvement becomes difficult

Engineering Contradiction:
Improvestorage capacityVSAvoidscalability
Core Design Contradiction:
Quantity of substanceVSAdaptability or versatility

Solution Approach 1:

The invention employs three-dimensional stacking architecture where memory cells are arranged vertically with multiple layers of bit lines and word lines intersecting at right angles. This 3D configuration provides a scalable path for increasing storage capacity beyond the limitations of lateral scaling, allowing continuous adaptation and expansion.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Quantity of substance

If DRAM is stacked into 3D architectures, then storage density improves, but manufacturing complexity increases

Engineering Contradiction:
Improvestorage densityVSAvoidmanufacturing complexity
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent divides the memory structure into repeating modular units consisting of alternating layers of first and second conductive materials, with bit lines and word lines formed as separate interleaved stacks. This segmentation into standardized modules simplifies the manufacturing process by enabling repetitive fabrication steps and reducing overall device complexity despite the 3D architecture.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The invention utilizes alternating layers of conductive materials with different properties (first conductive material layers for bit lines, second conductive material layers for word lines) to simplify the formation process. By changing material parameters and leveraging the distinct electrical properties of different conductive layers, the patent reduces manufacturing complexity while achieving high storage density.

Inventive Principle:
Principle #35Parameter changes

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach provides a high-density, low-power, and fast memory solution that surpasses conventional DRAM performance, enabling scalable and efficient non-volatile storage with reduced power consumption and increased storage capacity.

Implementation Method 1

each ferroelectric memory cell comprising a ferroelectric layer adapted to provide non-volatile storage of data

Methodology Applied
Scientific EffectFerroelectricity:

Implementation Method 2

a tunnel barrier layer contactingly engaging the ferroelectric layer, and opposing first and second conductive layers between which the ferroelectric layer and the tunnel barrier are disposed

Methodology Applied
Scientific EffectQuantum tunneling:

Data Source

PatentUS20210272983A1Three-dimensional ferroelectric memory
Publication Date: 2021.09.02 SEAGATE TECH LLC
  • US20210272983A1 patent drawing
  • US20210272983A1 patent drawing
  • US20210272983A1 patent drawing

AI summary

A memory device has ferroelectric memory cells arranged into a three-dimensional (3D) structure. Each ferroelectric memory cell has a ferroelectric layer adapted to provide non-volatile storage of data. In some cases, each ferroelectric memory cell is arranged as a ferroelectric field effect transistor (FeFET) comprising a source region, a drain region, and a control gate region, the control gate region comprising the ferroelectric layer. In other cases, each ferroelectric memory cell is arranged as a ferroelectric tunnel junction (FTJ) comprising opposing conductive electrode layers between which the ferroelectric layer and a tunnel junction layer are contactingly disposed. The ferroelectric layer may be formed of HfO2, ZrO2, Hf1-xZxO2, etc. The tunnel barrier layer may be formed of Al2O3, MgO, SrTiO3, etc. The memory can be used as a substitute for DRAM, a main memory in a data storage device, a data cache, etc.