Photomask Defect Inspection Using Active Region Criteria
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing complexity of photomasks with smaller elements at higher densities in LSIs makes defect inspection and correction costly and inefficient, as existing methods struggle to optimally distinguish between defective and non-defective patterns, particularly intricate graphic patterns, leading to high inspection costs and yield issues.
Innovation Solution
A method that involves preparing circuit data, layout data, and mask-manufacturing data with attributes indicating active and non-active regions, using a connection-information table to set criteria for defect determination based on the type of patterns formed on the photomask, allowing for adaptive defect inspection and correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If defect detection sensitivity is increased to detect all defects in intricate photomask patterns, then measurement precision is improved, but productivity deteriorates due to increased inspection time and cost
Solution Approach 1:
The patent applies different defect determination criteria to different regions of the photomask based on their functional importance. Active regions (containing actual circuit patterns) use strict criteria where any deviation is considered defective, while non-active regions (containing dummy patterns for manufacturing control) use lenient criteria where certain deviations are tolerated. This local differentiation allows high detection sensitivity for critical areas without requiring equally stringent inspection of all areas, thereby maintaining measurement precision where needed while improving overall inspection efficiency.
2Manufacturing precision
If all detected defects are corrected using defect-correcting apparatus, then manufacturing precision is improved, but loss of time increases due to repeated inspection and correction cycles
Solution Approach 1:
The patent extracts and identifies dummy patterns in non-active regions and applies separate determination criteria to them. By extracting these non-critical patterns from the overall defect assessment, the system avoids flagging them as defects requiring correction. This selective extraction prevents unnecessary correction cycles for patterns that do not affect LSI functionality, thereby reducing time loss while maintaining manufacturing precision for actual circuit patterns in active regions.
Solution Approach 2:
The patent performs preliminary classification of photomask regions into active and non-active areas before defect determination. By预先 (in advance) identifying which regions contain functionally critical patterns and which contain dummy patterns, the system can apply appropriate criteria from the outset. This preliminary action prevents the generation of false defect reports that would otherwise trigger unnecessary correction and re-inspection cycles, thus reducing time loss while ensuring manufacturing precision for critical regions.
3Productivity
If photomasks with smaller elements at higher density are manufactured to achieve higher LSI integration density, then productivity is improved, but device complexity increases making defect inspection more difficult
Solution Approach 1:
The patent segments the photomask pattern data into distinct categories: active region patterns (actual circuits) and non-active region patterns (dummy patterns). This segmentation allows the inspection system to process and evaluate different pattern types separately using appropriate criteria. For high-density photomasks with intricate patterns, this segmentation simplifies the inspection complexity by focusing strict evaluation only on active regions while applying lenient criteria to non-active regions, thereby enabling efficient inspection of complex, high-density designs.
Solution Approach 2:
The patent applies different determination criteria to different spatial regions of the photomask based on their functional significance. Active regions containing actual circuit patterns receive strict local quality control with sensitive defect detection, while non-active regions containing dummy patterns receive lenient local quality control. This local quality differentiation manages the complexity of inspecting high-density photomasks by concentrating inspection resources on critical areas rather than uniformly processing all patterns, thus improving productivity while managing device complexity.
Data Source
AI summary
A method of determining defects in photomasks according to the present invention is designed to increase the yield of the manufacture of photomasks and to decrease the cost of inspecting the photomasks. In the method, circuit data 1 representing a circuit to be formed on a semiconductor substrate by photolithography is prepared, and layout data 2 is prepared from the circuit data 1. The layout data is converted to compensated layout data by performing RET. Further, mask-manufacturing data is developed from the compensated layout data. To form patterns on a semiconductor substrate by photolithography, attribute information is imparted to the mask-manufacturing data. The attribute information represents whether the patterns are adaptive to electrically active regions or electrically non-active region. In the mask-inspecting process 6, a criterion for determining whether the patterns formed on the photomasks have defects is changed in accordance with the attribute information.