OTP Fuse PN Junction Structure for Stable Programming Current

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current electronic fuses (e-fuses) in semiconductor ICs face challenges in controlling the dimensions of silicide fuse links, leading to variability in programming current, making it difficult to form e-fuses with desired programming characteristics.

Innovation Solution

A one-time programmable (OTP) fuse structure is developed, featuring a trench isolation, a gate metal layer, and a PN junction formed by highly doped n-type and p-type polysilicon layers, with a fuse link defined in the gate metal layer, allowing for better dimension control and reduced programming current variability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If silicide fuse links are used in conventional e-fuses, then the fuse structure is simple and manufacturing is straightforward, but the dimensions (active width and thickness) cannot be precisely controlled, leading to high variability in programming current

Engineering Contradiction:
Improvefuse link dimension controlVSAvoidfuse structure complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The invention transitions from controlling fuse link dimensions in the planar FEOL layer to controlling dimensions in the vertical back-end-of-line (BEOL) layer. By forming the fuse link in an interlayer dielectric layer above the substrate, the patent utilizes the vertical dimension to achieve precise dimensional control through standard photolithography and etching processes, bypassing the limitations of FEOL layer control.

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

Solution Approach 2:

The patent introduces an interlayer dielectric layer as an intermediary medium to form the fuse link. This dielectric layer serves as a host matrix that allows precise definition of fuse link dimensions through standard semiconductor fabrication processes, while also providing electrical isolation and mechanical support. The fuse link is formed as a conductive structure embedded within this dielectric intermediary.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Use of energy by moving object

If narrower and thinner fuse links are used, then less programming current is required, but the dimensions become difficult to control with conventional silicide processes

Engineering Contradiction:
Improveprogramming currentVSAvoidfuse link dimension control
Core Design Contradiction:
Use of energy by moving objectVSManufacturing precision

Solution Approach 1:

The patent achieves precise control of narrow and thin fuse link dimensions by moving the formation process to the BEOL layer, where standard photolithography and etching techniques can precisely define sub-micron dimensions. The vertical stacking allows for well-controlled thin fuse links with consistent cross-sectional areas, enabling low programming current while maintaining manufacturing precision.

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

Solution Approach 2:

The invention changes the material and structural parameters of the fuse link by using conductive structures (such as metal interconnects or doped polysilicon) within an interlayer dielectric, rather than relying on silicide thickness and width in the FEOL layer. This parameter change enables precise control of fuse link geometry and consequently the programming current characteristics.

Inventive Principle:
Principle #35Parameter changes

3Ease of manufacture

If variable fuse link dimensions are accepted, then manufacturing is easier, but the programming current becomes unpredictable and production margins are reduced

Engineering Contradiction:
Improvefuse fabricationVSAvoidprogramming current consistency
Core Design Contradiction:
Ease of manufactureVSReliability

Solution Approach 1:

By forming fuse links in the vertical BEOL layer using standard interconnect fabrication processes, the patent achieves both ease of manufacture and high reliability. These processes are well-established in semiconductor manufacturing, providing excellent dimensional control and repeatability across production batches, thereby ensuring consistent programming current while maintaining manufacturing simplicity.

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

Solution Approach 2:

The patent changes the fabrication approach from silicide deposition and annealing in the FEOL layer to standard photolithography, etching, and material deposition in the BEOL layer. This parameter change leverages mature manufacturing processes that offer tight process control and high repeatability, ensuring consistent fuse link dimensions and predictable programming current across production.

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

The OTP fuse structure enables precise programming and improved e-fuse efficiency with reduced variability in programming current, enhancing production margins and programming efficiency compared to conventional silicide-based e-fuses.

Implementation Method 1

a first terminal including a highly doped n-type polysilicon layer over the trench isolation; a second terminal including a highly doped p-type polysilicon layer over the trench isolation, wherein the highly doped n-type polysilicon layer contacts the highly doped p-type polysilicon layer, creating a PN junction

Methodology Applied
Scientific EffectPN junction: Diode

Implementation Method 2

E-fuses can be irreversibly programmed by passing an electric current through the fuse link to cause electromigration of the fuse link material. Once sufficient material has migrated, the fuse link is programmed (i.e., higher resistance/blown), stopping/resisting current from passing through the e-fuse.

Methodology Applied
Scientific EffectElectromigration:

Data Source

PatentEP4383328A1One-time programmable fuse using PN junction over gate metal layer, and related method
Publication Date: 2024.06.12 GLOBALFOUNDRIES US INC
  • EP4383328A1 patent drawingFigure 1~2
  • EP4383328A1 patent drawingFigure 3~4
  • EP4383328A1 patent drawingFigure 5

AI summary

A one-time programmable (OTP) fuse includes a trench isolation; a gate metal layer over the trench isolation; and a PN junction over the gate metal layer. More particularly, the OTP fuse may include a first terminal including a highly doped n-type polysilicon layer over the trench isolation, and a second terminal including a highly doped p-type polysilicon layer over the trench isolation. The highly doped n-type polysilicon layer contacts the highly doped p-type polysilicon layer, creating a PN junction and a fuse link defined in a portion of the gate metal layer between the trench isolation and the PN junction. The gate metal layer has a uniform thickness that allows better dimension control of the fuse link to reduce fuse programming current variability.