Wafer Bonding Recipe Tuning for Distortion Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional wafer bonding processes face challenges in controlling post-bond distortion due to inter-lot variations in wafer shape and processing conditions, leading to poor inter-lot distortion control and increased costs from direct measurement of post-bond distortion.
Innovation Solution
A method utilizing wafer metrology data to dynamically tune the wafer bonding recipe through a model that predicts and adjusts process conditions to meet post-bond distortion thresholds, employing feedforward and feedback operations to optimize the bonding process based on pre- and post-bond wafer measurements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional fusion bonding process is used, then wafer bonding is achieved, but post-bond distortion control is poor due to inter-lot variations
Solution Approach 1:
The patent changes process parameters (bonding temperature, pressure, time) based on measured wafer characteristics (flatness, shape) to compensate for inter-lot variations. The model adjusts bonding parameters dynamically according to the specific wafer being processed, transforming a fixed process into an adaptive one that maintains consistent post-bond distortion control across different wafer lots.
Solution Approach 2:
The patent implements a feedback loop where wafer characteristics are measured before bonding, the results are input to a model that predicts post-bond distortion, and the model outputs adjusted bonding parameters. This closed-loop system continuously refines the bonding process based on actual wafer properties, improving distortion control while adapting to inter-lot variations.
2Measurement precision
If direct measurement of post-bond distortion is performed, then distortion data is obtained, but measurement costs increase
Solution Approach 1:
The patent creates a virtual model (digital twin) of the bonding process that simulates and predicts post-bond distortion based on pre-bond wafer measurements. Instead of physically measuring every post-bond wafer, the model generates a copy/prediction of the expected distortion, allowing quality control without the cost of extensive physical measurements while maintaining measurement precision through accurate modeling.
Solution Approach 2:
The patent performs measurements and predictions before the bonding process completes. By measuring wafer characteristics pre-bond and using the model to predict post-bond distortion in advance, the system obtains distortion data without requiring costly post-bond measurements, effectively performing the measurement action beforehand.
3Ease of manufacture
If wafer bonding recipe is fixed, then process simplicity is maintained, but inter-lot distortion control deteriorates
Solution Approach 1:
The patent transforms the static, fixed bonding recipe into a dynamic system that automatically adjusts parameters based on input wafer characteristics. The model takes pre-bond wafer measurements and dynamically generates optimized bonding parameters for each lot, maintaining ease of manufacture through automated adjustment while significantly improving inter-lot distortion control.
Solution Approach 2:
The patent systematically changes bonding parameters (temperature, pressure, time) based on measured wafer properties. The model determines optimal parameter sets for different wafer types, allowing the process to adapt to inter-lot variations while maintaining a relatively simple overall workflow through automated parameter selection.
Data Source
AI summary
A method includes having a first wafer bonding recipe and a model of a wafer bonding process, the model comprising an input indicative of a physical parameter of a first wafer to be bonded to a second wafer and configured to output a wafer bonding recipe based on the physical parameter of the first wafer; obtaining measurements of the first wafer to obtain the physical parameter of the first wafer; generating, by the model, the first wafer bonding recipe based on the physical parameter of the first wafer; and bonding the first wafer to the second wafer in accordance with the first wafer bonding recipe to produce a first post-bond wafer.