Trench Isolation for Latch-Up Suppression in Bulk CMOS

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Bulk CMOS devices are susceptible to latch-up due to the close proximity of N-channel and P-channel transistors, leading to regenerative feedback and potential catastrophic failure, especially in radiation environments, with existing solutions like epitaxial substrates and guard ring diffusions being costly or complex.

Innovation Solution

The introduction of a semiconductor structure with doped wells and a dielectric-filled trench between them, where the trench's geometry and sidewall spacing are modified to increase carrier recombination velocity and disrupt the continuity of non-monocrystalline semiconductor material, creating an isolation region that limits parasitic transistor action.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If guard ring diffusions are used to suppress latch-up, then latch-up immunity is improved, but manufacturing cost increases and active area is reduced

Engineering Contradiction:
Improvelatch-up immunityVSAvoidmanufacturing cost
Core Design Contradiction:
ReliabilityVSEase of manufacture

Solution Approach 1:

The patent replaces expensive guard ring diffusions with a simpler trench isolation structure that uses standard dielectric materials and conventional fabrication processes, significantly reducing manufacturing cost while maintaining latch-up suppression functionality

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The patent extracts the essential function of carrier collection from the complex guard ring diffusion structure and implements it through a simplified trench isolation approach, removing unnecessary structural complexity while preserving the latch-up immunity function

Inventive Principle:
Principle #2Taking out (Extraction)

2Reliability

If guard ring diffusions are used to suppress latch-up, then latch-up immunity is improved, but active area is reduced

Engineering Contradiction:
Improvelatch-up immunityVSAvoidactive area
Core Design Contradiction:
ReliabilityVSArea of moving object

Solution Approach 1:

The trench isolation structure occupies minimal space compared to extensive guard ring diffusions, preserving maximum active area for functional circuit elements while providing effective latch-up suppression through the dielectric-filled trench

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The patent applies latch-up suppression locally at critical interfaces between N-well and P-substrate using targeted trench isolation, rather than using extensive guard rings across the entire device area, thus maximizing active area utilization

Inventive Principle:
Principle #3Local quality

3Reliability

If epitaxial substrates are used to increase latch-up immunity, then latch-up immunity is improved, but production cost and design complexity increase

Engineering Contradiction:
Improvelatch-up immunityVSAvoiddesign complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts the latch-up immunity function from complex epitaxial substrate requirements and implements it through a simpler structure using standard bulk CMOS substrates with trench isolation, reducing both production cost and design complexity

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the approach from modifying substrate properties (epitaxial layers) to modifying device structure (trench isolation geometry), allowing standard substrates to achieve enhanced latch-up immunity without increasing design complexity

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 effectively reduces latch-up susceptibility in bulk CMOS devices by impeding minority carrier flow and reducing bipolar gain, thereby enhancing device immunity to latch-up without increasing production costs or complexity.

Implementation Method 1

the trench's geometry and sidewall spacing are modified to increase carrier recombination velocity and disrupt the continuity of non-monocrystalline semiconductor material

Methodology Applied
Scientific EffectCarrier recombination:

Data Source

PatentUS7791145B2Semiconductor structures for latch-up suppression and methods of forming such semiconductor structures
Publication Date: 2010.09.07 AURIGA INNOVATIONS INC
  • US7791145B2 patent drawing
  • US7791145B2 patent drawing
  • US7791145B2 patent drawing

AI summary

Semiconductor structures and methods for suppressing latch-up in bulk CMOS devices. The semiconductor structure comprises a shaped-modified isolation region that is formed in a trench generally between two doped wells of the substrate in which the bulk CMOS devices are fabricated. The shaped-modified isolation region may comprise a widened dielectric-filled portion of the trench, which may optionally include a nearby damage region, or a narrowed dielectric-filled portion of the trench that partitions a damage region between the two doped wells. Latch-up may also be suppressed by providing a lattice-mismatched layer between the trench base and the dielectric filler in the trench.