Regional Pattern Density Determination for Semiconductor Fabrication

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for determining long-range density in semiconductor design fabrication processes are resource-intensive and do not allow for interactive, near real-time manipulations, making remedial changes during the fabrication of advanced microprocessors, such as those involving rapid thermal anneal or polysilicon critical dimension control, impractically time-consuming.

Innovation Solution

A method and system that determine localized regional pattern density by calculating pattern density values for each cell, averaging these values within a selected region, and storing the results, allowing for rapid determination of regional pattern density across multiple cells, significantly reducing computational time.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional density calculation methods are used for long-range density checks, then measurement precision is maintained, but productivity deteriorates significantly due to computational time requirements exceeding several days

Engineering Contradiction:
Improvedensity calculation precisionVSAvoidcomputational speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent divides the chip into multiple cells and further segments the calculation into localized regions around each cell. Instead of calculating density for the entire chip at once, the method calculates localized regional density for individual cells and combines results, reducing the computational burden while maintaining measurement precision

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from calculating density across the entire two-dimensional chip area to calculating localized density in smaller regional zones around each cell. This dimensional breakdown allows parallel processing of multiple regions, significantly improving computational speed while preserving accuracy through the aggregation of localized measurements

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

2Reliability

If conventional density calculation methods are used, then comprehensive density coverage is achieved, but loss of time increases making remedial changes impossible during fabrication

Engineering Contradiction:
Improvedensity check completenessVSAvoidcalculation time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent performs preliminary localized density calculations for each cell region before final aggregation. By pre-calculating regional densities and storing them, the system enables rapid retrieval and combination of results, reducing total calculation time while maintaining comprehensive density assessment

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements a continuous process where localized density calculations for multiple cells are performed in sequence or parallel, with results accumulated and combined. This continuous approach ensures comprehensive coverage of the entire chip while maintaining efficient throughput, enabling real-time feedback during fabrication

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentUS7703053B2Regional pattern density determination method and system
Publication Date: 2010.04.20 X CORP
  • US7703053B2 patent drawing
  • US7703053B2 patent drawing
  • US7703053B2 patent drawing

AI summary

A method and system of determining a localized measure of regional pattern density in a fabrication process of a chip are disclosed. In one embodiment, the method includes determining pattern density values for each cell of a plurality of cells of interest; averaging the pattern density values for each cell within a first selected region about a target cell to determine the localized measure of regional pattern density for the target cell; storing the localized measure of regional pattern density for the target cell; and repeating the averaging and the storing for each of the plurality of cells. The simplification of data allows for a localized measure of regional pattern density determination in much less time than conventional techniques.