VSB Lithography Calibration for Elementary Pattern Errors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for variable-shaped-beam electron-beam lithography fail to systematically correct errors associated with small elementary patterns, which affect the accuracy of pattern transfer and estimation of electron spread models.
Innovation Solution
A method involving the production of calibration patterns with geometric figures divided into smaller elementary patterns, measuring their actual dimensions, and applying a regression method to construct a mathematical model expressing dimensional variations or exposure dose errors as a function of pattern dimensions, allowing for data preparation and correction of layout patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If small elementary patterns are used to produce fine features, then pattern resolution is improved, but dimensional accuracy deteriorates due to uncorrected systematic errors
Solution Approach 1:
The patent applies parameter changes by systematically varying the dimensions of calibration patterns (different widths, lengths, and geometries) to establish empirical relationships between nominal and actual dimensions. This allows the creation of correction factors that account for systematic errors in small pattern fabrication, thereby improving dimensional accuracy without sacrificing resolution.
Solution Approach 2:
The patent implements feedback through an iterative calibration process where measured actual dimensions of calibration patterns are used to update correction models. These corrected models then feed back into the data preparation process, continuously improving the accuracy of elementary pattern dimension control based on empirical measurements rather than relying solely on theoretical models.
2Productivity
If conventional physical models are used for data preparation, then processing speed is maintained, but pattern transfer accuracy deteriorates for small elementary patterns
Solution Approach 1:
The patent applies preliminary action by pre-calculating correction factors through systematic calibration measurements before actual production. The empirical correction models are established in advance through measuring calibration patterns with known nominal dimensions, allowing rapid application of corrections during data preparation without compromising production speed or pattern transfer accuracy.
Solution Approach 2:
The patent creates a composite approach by combining conventional physical models (point spread function, resist model) with empirically derived correction factors. This hybrid model integrates the speed advantages of theoretical models with the accuracy benefits of experimental measurements, producing a composite data preparation method that maintains productivity while improving precision for small patterns.
3Length of moving object
If elementary pattern dimensions are reduced below critical dimension, then feature size control is improved, but systematic errors increase without correction
Solution Approach 1:
The patent substitutes mechanical/physical modeling approaches with an empirically-based correction system. Instead of relying solely on theoretical models of electron scattering and resist behavior, the system replaces these complex physical simulations with measured correction factors derived from actual calibration patterns, thereby achieving better dimensional consistency for sub-critical features.
Solution Approach 2:
The patent systematically changes parameters by measuring calibration patterns across a range of different dimensions (widths from 50nm to 200nm, various lengths and geometries) to establish how systematic errors vary with size. This parametric approach allows the creation of size-dependent correction factors that maintain dimensional consistency across different feature scales.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach significantly reduces errors in pattern transfer and improves the estimation of electron spread models, enabling more accurate data preparation and correction of layout patterns in electron-beam lithography.
Implementation Method 1
An electron beam is used to expose a resist deposited on the substrate according to the layout to be transferred. The exposed resist undergoes a chemical transformation that allows it to be selectively removed
Implementation Method 2
This is mainly due to scattering of the electrons in the resist and to backscattering thereof by the substrate
Data Source
AI summary
A method for calibrating elementary patterns in variable-shaped-beam electron-beam lithography, includes the following steps: producing, by variable-shaped-beam electron-beam lithography, a calibration pattern comprising geometric figures each having a nominal critical dimension, the figures being divided into elementary patterns of smaller dimensions than each the nominal critical dimension; measuring the actual critical dimension of each the geometric figure; and applying a regression method on the basis of the actual critical dimensions thus determined to construct a mathematical model expressing either a variation in dimensions of the elementary patterns, or an error in the exposure dose of the elementary patterns producing an equivalent effect to the variation in dimensions, as a function of the dimensions of the elementary patterns. Application to the preparation of data with a view to transferring a pattern to a substrate by variable-shaped-beam electron-beam lithography.