Datapath Noise Controller for Lithography Dose Uniformity

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Lithographic apparatuses using arrays of individually controllable elements face challenges in maintaining dose uniformity and critical dimension uniformity due to rounding errors from quantization of data streams, which affect the precision of pattern transfer onto substrates.

Innovation Solution

Incorporating a noise controller to introduce a predetermined amount of noise into the setpoint data stream before digitization, which randomizes rounding errors and improves dose and critical dimension uniformity without increasing quantization resolution or datapath bandwidth.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If quantization is applied to data streams defining setpoints for array elements, then data can be processed and transmitted, but rounding errors occur that reduce dose uniformity and critical dimension uniformity

Engineering Contradiction:

Improvedata processing capabilityVSAvoiddose uniformity

Core Design Contradiction:

ProductivityVSManufacturing precision

Solution Approach 1:

The patent converts the harmful quantization rounding errors into beneficial randomized noise by adding dither signal before quantization. This transforms systematic dose uniformity degradation into randomized errors that average out, improving overall dose uniformity while maintaining data processing capability.

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

Solution Approach 2:

The patent changes the statistical parameters of the error distribution by adding random noise before quantization. This modifies the error characteristics from systematic bias to randomized distribution with zero mean, allowing errors to cancel out through averaging across multiple elements.

Inventive Principle:

Principle #35Parameter changes

2Manufacturing precision

If quantization resolution is increased to reduce rounding errors, then dose uniformity improves, but datapath bandwidth requirements increase

Engineering Contradiction:

Improvedose uniformityVSAvoiddata bandwidth

Core Design Contradiction:

Manufacturing precisionVSQuantity of substance

Solution Approach 1:

The patent changes the statistical properties of quantization errors through dither addition, allowing coarser quantization (fewer bits) to achieve the same effective precision. This reduces data bandwidth requirements while maintaining dose uniformity through randomized error distribution that averages out.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS7508491B2Lithographic apparatus and device manufacturing method utilized to reduce quantization influence of datapath SLM interface to dose uniformity

Publication Date: 2009.03.24 ASML NETHERLANDS BV

AI summary

A lithography method and system comprise an array of individually controllable elements, a first datapath section, a second datapath section, and a quantization device. The array of individually controllable elements modulates a beam of radiation. The first datapath section converts data defining a requested dose pattern into a stream of setpoint data defining a desired sequence of setpoints for the array of individually controllable elements. The second datapath section stores and reproduces the stream of setpoint data for the array of individually controllable elements. The quantization device digitizes the stream of setpoint data before it is passed from the first datapath section to the second datapath section. The first datapath section further comprises a noise controller configured to incorporate noise of a predetermined magnitude into the stream of setpoint data before it is digitized by the quantization device.