Lithography Control Using Basis Functions for Lot Variation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing lithographic processes face instability due to lot-to-lot variations in substrate model parameters, leading to inaccurate positioning of patterns on substrates, which affects the yield of semiconductor devices.

Innovation Solution

A method is introduced to define a substrate model using basis functions that represent process parameter fingerprints across substrates, allowing for the calculation of control parameters by filtering out noise and stabilizing model parameters based on similarity to process variations, thereby reducing the impact of lot-to-lot fluctuations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If substrate model parameters are determined using measurement data from previous lots, then the lithographic process can be controlled based on historical data, but the process becomes unstable due to lot-to-lot variations in the parameters

Engineering Contradiction:
Improveprocess stabilityVSAvoidparameter accuracy
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent applies preliminary action by measuring and characterizing substrate properties (such as wafer grid distortions, alignment mark positions, and topography) before the actual lithographic exposure process. These preliminary measurements are used to calculate substrate model parameters that predict and compensate for expected variations in subsequent processing steps, thereby improving process stability while maintaining parameter accuracy through advance characterization.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements feedback mechanisms by continuously measuring substrate properties during and between lithographic runs, then using these measurements to update and refine substrate model parameters. The system feeds back corrected parameters to the lithographic apparatus control system, allowing real-time adjustment of exposure conditions based on actual substrate behavior, which resolves the contradiction between using historical data and maintaining accuracy.

Inventive Principle:
Principle #23Feedback

2Manufacturing precision

If extensive measurement operations are performed prior to exposing the substrate, then the positioning accuracy of patterns improves, but the processing time and complexity increase

Engineering Contradiction:
Improvepattern placement accuracyVSAvoidmeasurement time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent performs critical measurements of substrate properties (alignment marks, wafer grid, topography) as preliminary actions before lithographic exposure. These measurements establish baseline data that enables accurate positioning corrections to be applied during the actual exposure process, achieving high manufacturing precision while minimizing time loss by completing measurements in advance rather than during exposure.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent employs dynamic measurement and correction strategies where the measurement process itself is optimized and adapted based on substrate characteristics. The system dynamically adjusts measurement parameters and selection of which substrates require full measurement versus rapid characterization, balancing measurement thoroughness with time constraints to achieve accurate pattern placement without excessive processing time.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS12366809B2Methods and apparatus for controlling a lithographic process
Publication Date: 2025.07.22 ASML NETHERLANDS BV
  • US12366809B2 patent drawing
  • US12366809B2 patent drawing
  • US12366809B2 patent drawing

AI summary

A method of determining a control parameter for a lithographic process is disclosed, the method includes: defining a substrate model for representing a process parameter fingerprint across a substrate, the substrate model being defined as a combination of basis functions including at least one basis function suitable for representing variation of the process parameter fingerprint between substrates and/or batches of substrates; receiving measurements of the process parameter across at least one substrate; calculating substrate model parameters using the measurements and the basis functions; and determining the control parameter based on the substrate model parameters and the similarity of the at least one basis function to a process parameter fingerprint variation between substrates and/or batches of substrates.