Shot Clusters for Semiconductor Lithography Mask Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As semiconductor feature sizes decrease, existing lithographic techniques face challenges in accurately creating desired shapes due to wavelength limitations of electromagnetic radiation, leading to harmful interactions between masks and light, which result in suboptimal exposed shapes.
Innovation Solution
A computer-implemented method for shot optimization is used to generate clusters of shots that approximate and modify mask shapes to closely match desired fabricated shapes, leveraging inverse lithography technology and optical proximity correction to iteratively refine the shot configuration, allowing for the creation of masks with a minimum number of shots while optimizing lithography performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If the wavelength of electromagnetic radiation is reduced to create smaller feature sizes, then the resolution of fabricated shapes is improved, but harmful interactions between the mask and light increase
Solution Approach 1:
The patent introduces shot clusters as an intermediary representation between the desired fabricated shapes and the actual mask patterns. These shot clusters act as a mediator that transforms the design intent into a form suitable for mask writing, reducing direct harmful interactions by using a discrete shot-based approach rather than continuous mask patterns
Solution Approach 2:
The patent changes the parameter representation from continuous mask shapes to discrete shot clusters with specific densities. By parameterizing the mask in terms of shot density and cluster distributions, the system can optimize the balance between resolution and harmful interactions through controlled parameter adjustments
2Manufacturing precision
If complex mask shapes are created to achieve desired fabricated shapes at sub-resolution levels, then the manufacturing precision is improved, but the device complexity increases
Solution Approach 1:
The patent segments complex mask shapes into discrete shot clusters that can be independently controlled and optimized. Instead of treating the mask as a continuous complex shape, it divides it into manageable shot units with specific densities, simplifying the overall device complexity while maintaining manufacturing precision
Solution Approach 2:
The patent inverts the traditional approach by not directly creating complex mask shapes from desired fabricated shapes. Instead, it starts with shot clusters and iteratively optimizes them to achieve the desired shapes, reversing the conventional design flow to reduce complexity
3Manufacturing precision
If more shots are used in mask writing to achieve better shape approximation, then the manufacturing precision is improved, but the productivity decreases
Solution Approach 1:
The patent applies partial action by using shot clusters with optimized density rather than maximizing the number of shots. The shot density is carefully controlled to provide sufficient shape approximation accuracy without the excessive action of using maximum possible shots, thus balancing precision and productivity
Solution Approach 2:
The patent implements feedback through iterative optimization of shot clusters. The system evaluates the approximation quality and adjusts shot densities and positions accordingly, using feedback from simulation results to refine the mask design and achieve optimal balance between precision and productivity
Data Source
AI summary
A design layout for a semiconductor chip includes information on shapes desired to be fabricated. Clusters of photolithographic exposure “shots” are generated and subject to a measure of shot density to approximate a mask shape that generates the desired fabricated shapes when exposed during wafer fabrication. A simulation is run on the clusters of shots to estimate the resulting fabrication shapes that the clusters of shots create. The clusters of shots are modified to align the estimated fabrication shapes more closely with desired fabrication shapes. The process of simulating and modifying the shots is iterative, repeating until the estimated fabrication shapes are within a desired error difference of the planned fabrication shape.