CMP Simulation Lookup Table for Layout Grid Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current CMP simulation methods are cumbersome, require large data collection, involve significant computation, and result in long prediction times, adversely affecting production efficiency in integrated circuit manufacturing.
Innovation Solution
A simulation method using a matrix table of line width logarithm-density, where each intersection corresponds to CMP results, with data interpolation and weighting factors to enhance simulation speed and accuracy, allowing for faster prediction of surface patterns post-CMP process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional CMP simulation method with comprehensive data collection and model debugging is used, then prediction accuracy is improved, but simulation time and computational resources are significantly increased
Solution Approach 1:
The patent pre-calculates and stores CMP simulation results for various line width and density combinations in a lookup table before actual production use. This preliminary action eliminates the need for time-consuming model debugging and data collection during production simulations, achieving both high accuracy and fast prediction speeds
Solution Approach 2:
The patent creates a simplified copy of the complex CMP simulation model by storing pre-computed results in a lookup table. Instead of running the full computational model during production, the system copies and retrieves pre-calculated data, dramatically reducing simulation time while maintaining prediction accuracy
2Measurement precision
If comprehensive data from all process steps is collected for model debugging, then simulation accuracy is improved, but data collection complexity and processing time are increased
Solution Approach 1:
The patent performs data collection and model debugging in advance during the lookup table construction phase. All necessary process data is collected and processed beforehand, eliminating the complexity of data collection during production simulations while ensuring accurate predictions
Solution Approach 2:
The patent extracts only the essential parameters (line width and density) needed for CMP simulation from the complex process data, storing them in a simplified lookup table structure. This extraction eliminates the need to handle comprehensive process data during production, reducing complexity while maintaining accuracy
3Measurement precision
If detailed model prediction calculations are performed for each grid, then prediction accuracy is improved, but computational load and production period are increased
Solution Approach 1:
The patent pre-computes prediction results for all possible line width and density combinations and stores them in a lookup table. During production, the system simply retrieves pre-calculated results based on actual grid parameters, achieving both high accuracy and fast processing speeds
Solution Approach 2:
The patent replaces complex real-time calculations with pre-computed copied data from the lookup table. The system copies relevant prediction results from the pre-built table based on matching line width and density values, eliminating the need for repeated computational operations
Data Source
AI summary
A simulation method of CMP process comprises: building a CMP model, and forming a matrix table of line width logarithm-density according to the CMP model, and making each intersection of the matrix table correspond to each CMP result under the corresponding line width and density; dividing a layout into a plurality of grids, and converting the equivalent line width and density of each grid into the coordinate of line width logarithm-density in the matrix table; fitting and calculating preliminary CMP simulation results of each grid according to the coordinate of each grid in the matrix table and the CMP simulation results of its adjacent intersections of the matrix table; fitting and computing final CMP simulation results of each grid according to a related weighting factor which considers the impact of adjacent grids for the current grid on the layout; outputting the final CMP simulation results of the whole layout.