Learning-Based Layout Rules for IC Latch-Up Mitigation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Integrated circuits, particularly CMOS devices, are susceptible to latch-up failures due to parasitic devices triggered by temperature increases, which can lead to functional disruptions or destruction, especially in harsh environmental conditions, and die shrinks exacerbate this issue by reducing latch-up robustness.

Innovation Solution

A learning-based system that obtains latch-up data from integrated circuits operating under various temperature conditions and trains on layout rules to generate new spacing rules between the core and input/output portions, incorporating heavily doped regions to mitigate latch-up without altering device dimensions, thus enabling efficient die size reduction while maintaining robustness.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If die shrinkage is implemented to increase the number of potential die per wafer, then productivity increases, but latch-up robustness deteriorates

Engineering Contradiction:
Improvenumber of potential die per waferVSAvoidlatch-up robustness
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent applies different spacing rules to different regions of the integrated circuit layout. Specifically, it maintains larger spacing between the core portion and input/output portion where latch-up susceptibility is highest, while allowing tighter spacing in other regions. This localized approach to spacing optimization enables die shrinkage while preserving latch-up robustness in critical areas.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent implements spacing rules that proactively prevent latch-up conditions before they can occur. By establishing minimum spacing requirements between the core portion and input/output portion during the design phase, the system prevents the formation of parasitic structures that could lead to latch-up, rather than attempting to mitigate latch-up after it occurs.

Inventive Principle:
Principle #10Preliminary action

2Adaptability or versatility

If temperature increases to enable operation in harsh environmental conditions, then adaptability improves, but latch-up susceptibility increases

Engineering Contradiction:
Improveoperational temperature rangeVSAvoidlatch-up susceptibility
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent applies spacing rules that counteract the temperature-induced reduction in diode forward bias voltage. By maintaining adequate spacing between the core portion and input/output portion, the design preemptively prevents the parasitic bipolar transistor structure from being triggered into an on state, even when temperature increases reduce the latch-up trigger current.

Inventive Principle:
Principle #9Preliminary anti-action

3Reliability

If spacing between core portion and input/output portion is increased to prevent latch-up, then latch-up robustness improves, but die area increases

Engineering Contradiction:
Improvelatch-up robustnessVSAvoiddie area
Core Design Contradiction:
ReliabilityVSArea of stationary object

Solution Approach 1:

The patent applies spacing rules that optimize the balance between latch-up robustness and die area. By targeting the specific region between the core portion and input/output portion for spacing control, the design achieves adequate latch-up protection only where needed, rather than uniformly increasing spacing across the entire die. This localized approach minimizes the impact on overall die area while maintaining robustness in critical regions.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS11853683B2Learning-based analyzer for mitigating latch-up in integrated circuits
Publication Date: 2023.12.26 SILICON SPACE TECH D B A VORAGO TECH
  • US11853683B2 patent drawing
  • US11853683B2 patent drawing
  • US11853683B2 patent drawing

AI summary

Systems and methods related to learning-based analyzers (both supervised and unsupervised) for mitigating latch-up in integrated circuits are provided. An example method includes obtaining latch-up data concerning at least one integrated circuit configured to operate under a range of temperature conditions, where the at least one integrated circuit comprises a core portion including at least a plurality of devices each having one or more structural features formed using a lithographic process, and an input/output portion. The method further includes training the learning-based system based on training data derived from the latch-up data and a first layout rule concerning a first spacing between the core portion and the input/output portion. The method further includes using the learning-based system generating a second layout rule concerning the first spacing between the core portion and the input/output portion, where the second layout rule is different from the first layout rule.