Non-Volatile SRAM Programming via Volatile Bit Lines

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing semiconductor memory devices face challenges in programming and erasing the non-volatile section without additional silicon overhead, particularly in efficiently utilizing the bit lines from the volatile portion to minimize space and enhance speed.

Innovation Solution

The method involves programming and erasing tri-gate structures in the non-volatile SRAM cells using bit lines, where specific voltage conditions and pulse applications are used to control recall and store transistors, allowing for programming and erasing of SONOS transistors without additional silicon overhead, enabling dual programming capabilities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If additional voltage supply nodes are added to the non-volatile section for programming and erasing, then programming and erasing functionality is improved, but silicon area increases

Engineering Contradiction:
Improveprogramming and erasing functionalityVSAvoidsilicon area
Core Design Contradiction:
Adaptability or versatilityVSArea of stationary object

Solution Approach 1:

The bit lines from the volatile SRAM section are made to serve dual purposes: they continue to function as data transmission lines for the volatile portion while simultaneously serving as voltage supply lines for programming and erasing the non-volatile SONOS section. This multi-functionality eliminates the need for separate voltage supply nodes, thereby maintaining programming/erasing capability without increasing silicon area.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent merges the volatile SRAM bit lines with the non-volatile programming/erasing voltage supply function. By combining these previously separate functions into a single shared infrastructure, the design achieves space efficiency while maintaining full functionality of both volatile and non-volatile operations.

Inventive Principle:
Principle #5Merging (Combining)

2Ease of operation

If separate voltage supply nodes are used for non-volatile programming and erasing, then programming control is improved, but device complexity increases

Engineering Contradiction:
Improveprogramming controlVSAvoiddevice complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The bit lines are designed to operate in multiple modes: data transmission mode for volatile SRAM operations and voltage supply mode for non-volatile programming/erasing. This unified approach reduces device complexity by eliminating separate control infrastructure while maintaining precise programming control through mode-dependent operation of the same physical lines.

Inventive Principle:
Principle #6Universality (Multi-functionality)

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 allows for efficient programming and erasing of non-volatile SRAM cells using bit lines, reducing silicon usage and enhancing memory efficiency, enabling dual programming and storage modes within the same tri-gate structure.

Implementation Method 1

programming the non-volatile section of a non-volatile Static Random Access Memory (SRAM) from the bit lines of the volatile section

Methodology Applied
Scientific EffectElectron tunneling: Electron Beam

Data Source

PatentUS7539054B2Method and apparatus to program and erase a non-volatile static random access memory from the bit lines
Publication Date: 2009.05.26 NVIDIA CORP
  • US7539054B2 patent drawing
  • US7539054B2 patent drawing
  • US7539054B2 patent drawing

AI summary

A system and method for programming and erasing a semiconductor memory is disclosed. More particularly, the present invention uses the bit lines of a volatile memory portion of semiconductor memory so as to program and erase the non-volatile portion of the semiconductor memory.