Backside Isolation Regions for Uniform Gate-All-Around IC Fabrication

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for forming isolation regions between semiconductor devices in ICs cause within-die variations, leading to yield issues and performance deterioration due to non-uniform semiconductor structures and gate etch out, particularly when cutting semiconductor structures before gate formation.

Innovation Solution

Forming backside isolation regions by removing semiconductor regions between gates from the backside of the IC device, which avoids non-uniform semiconductor structures and reduces within-die variations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If semiconductor structures are cut in isolation regions before gate formation, then isolation regions can be formed between devices, but non-uniform semiconductor structures and gate etch out occur leading to within-die variations

Engineering Contradiction:
Improveisolation region formation precisionVSAvoidwithin-die variation consistency
Core Design Contradiction:
Manufacturing precisionVSReliability

Solution Approach 1:

The patent inverts the conventional sequence by forming isolation regions after gate formation rather than before. Specifically, gates are formed across continuous semiconductor structures first, then isolation regions are created by removing semiconductor material in non-functional areas after the gates are in place. This reversal eliminates gate etch out issues and non-uniform semiconductor structures that plague the conventional approach, as the gates are already formed and protected when isolation regions are subsequently defined.

Inventive Principle:
Principle #13The other way round (Inversion)

Solution Approach 2:

The patent applies preliminary action by forming gates across the entire wafer including potential isolation regions before creating the isolation structures themselves. This ensures that gate patterns are established on uniform semiconductor structures first, and then isolation regions are defined afterward by selective removal of semiconductor material. The preliminary gate formation prevents within-die variations because all gates experience the same formation conditions on uniform substrate before any isolation structures are introduced.

Inventive Principle:
Principle #10Preliminary action

2Ease of manufacture

If conventional isolation region formation methods are used, then isolation between devices is achieved, but yield issues and performance deterioration occur due to within-die variations

Engineering Contradiction:
Improveisolation region fabricationVSAvoiddevice yield
Core Design Contradiction:
Ease of manufactureVSProductivity

Solution Approach 1:

By inverting the process sequence to form gates before isolation regions, the patent eliminates the harmful effects of gate etch out and non-uniform semiconductor structures that currently reduce yield. The conventional approach cuts semiconductor structures before gate formation, causing within-die variations that deteriorate device performance. The inverted approach forms gates on uniform structures first, then creates isolation regions afterward, thereby maintaining high yield while still achieving proper device isolation.

Inventive Principle:
Principle #13The other way round (Inversion)

3Device complexity

If semiconductor structures are cut before gate formation, then isolation regions can be defined, but non-uniform semiconductor structures result affecting downstream processes

Engineering Contradiction:
Improveisolation structure definitionVSAvoidsemiconductor structure uniformity
Core Design Contradiction:
Device complexityVSManufacturing precision

Solution Approach 1:

The patent resolves the uniformity issue by inverting the process sequence: gates are formed first on continuous, uniform semiconductor structures across the entire wafer, then isolation regions are created by selective removal of semiconductor material afterward. This ensures that all semiconductor structures maintain uniformity during the critical gate formation step, and subsequent isolation region definition does not compromise this uniformity since the gates are already in place and the semiconductor structures are not cut before gating.

Inventive Principle:
Principle #13The other way round (Inversion)

Data Source

PatentUS20250212470A1Integrated circuit device with backside isolaton region
Publication Date: 2025.06.26 INTEL CORP
  • US20250212470A1 patent drawing
  • US20250212470A1 patent drawing
  • US20250212470A1 patent drawing

AI summary

An IC device may have active regions and one or more isolation regions. The IC device includes gates that are in parallel. One or more semiconductor structures (e.g., fins, nanoribbons, etc.) may extend across each gate in the IC device. Some of the gates are in the active regions. The other gates are in the isolation region. A gate in an active region may be between semiconductor regions, which may function as the source region and drain region of a transistor. A gate in an isolation region may be between insulator regions. The insulator regions may be formed from the backside of the IC device. For instance, semiconductor regions may be formed in both the active regions and the isolation regions. The semiconductor regions in the regions designated to be isolation regions may be removed from the backside and filled with one or more electrical insulators.