Automated Semiconductor Defect Recipe Creation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The manual creation of defect review process recipes for semiconductor inspection tools is time-consuming and prone to human error, requiring significant operator time and training, especially when dealing with multiple defect types and layers, which hampers productivity and increases the likelihood of errors in defect detection.
Innovation Solution
An automated method and system that determine the identity of a specimen, identify inspection results, and create a defect review process recipe without user input, using a look-up table and sensors to select appropriate parameters for the defect review tool, such as a scanning electron microscope, based on the specimen's layer type and inspection results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual recipe creation is used for defect review process, then operator control and flexibility are maintained, but recipe setup time increases significantly and human error likelihood increases
Solution Approach 1:
The defect review tool automatically creates its own recipe by determining specimen identity through sensors (e.g., bar code reader), retrieving inspection results, and selecting appropriate parameters without requiring manual input from operators. This self-service mechanism eliminates human error and significantly reduces recipe setup time.
Solution Approach 2:
The system performs preliminary actions by pre-storing inspection results and specimen information in databases, and pre-configuring parameter selection logic. When a specimen is loaded, the recipe is automatically generated based on previously stored information, avoiding time-consuming manual configuration and reducing errors.
2Productivity
If automated recipe creation is implemented, then recipe setup time decreases and productivity increases, but system complexity increases
Solution Approach 1:
The defect review tool integrates multiple functions into a single automated system: specimen identification through sensors, database querying for inspection results, parameter selection based on specimen type and defect characteristics, and automatic recipe generation. This multi-functionality increases productivity while managing complexity through integration rather than separate manual processes.
Solution Approach 2:
The system uses databases as intermediaries to store and retrieve specimen information and inspection results, and uses parameter selection logic as an intermediary to translate specimen characteristics into appropriate review parameters. These intermediaries automate the process while keeping the system structure manageable.
3Measurement precision
If multiple parameters are adjusted for different defect types and layers, then defect detection accuracy improves, but the difficulty of determining appropriate parameters increases
Solution Approach 1:
The system automatically changes review parameters based on specimen identity and inspection results. Different parameter sets are selected for different specimen types, layers, and defect characteristics, ensuring optimal detection accuracy without requiring manual determination of appropriate parameters for each case.
Solution Approach 2:
The system uses inspection results as feedback to determine appropriate review parameters. By analyzing the outcomes of initial inspection, the automated recipe creation process selects parameters that are most likely to detect and characterize the observed defects, improving detection accuracy while eliminating the need for expert manual parameter selection.
Data Source
AI summary
Methods and systems for creating a recipe for a defect review process are provided. One method includes determining an identity of a specimen on which the defect review process will be performed. The method also includes identifying inspection results for the specimen based on the identity. In addition, the method includes creating the recipe for the defect review process based on the inspection results. One system includes a sensor configured to generate output responsive to an identity of a specimen on which the defect review process will be performed. The system also includes a processor configured to determine the identity of the specimen using the output, to identify inspection results for the specimen based on the identity, and to create the recipe for the defect review process based on the inspection results.