Beta-Ga2O3 Epitaxial Layer Prediction for MOCVD Pre-Reaction Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing MOCVD process for growing beta-Ga2O3 epitaxial layers is plagued by pre-reactions at high temperatures, leading to three-dimensional growth and deteriorated quality due to the formation of new nucleation points, and the reliance on subjective human experience for process regulation often misses the optimal window, especially for beta-Ga2O3 epitaxy.
Innovation Solution
A prediction method using a trained epitaxial layer prediction model that incorporates machine learning to adjust growth parameters such as O/Ga ratio, chamber pressure, and growth time to optimize the process window, ensuring high-quality two-dimensional growth by MOCVD.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If high reaction temperature and strong oxidation are used to grow beta-Ga2O3 epitaxial layer by MOCVD, then the growth rate is improved, but pre-reaction occurs leading to three-dimensional growth and deteriorated quality
Solution Approach 1:
The patent applies parameter changes by adjusting the O2 flow rate and O/Ga ratio to optimize the oxidation process. By controlling these parameters, the method achieves high-quality two-dimensional growth without pre-reaction, resolving the contradiction between growth rate and quality. The trained prediction model identifies optimal parameter combinations that maintain high productivity while preventing harmful pre-reactions.
Solution Approach 2:
The patent implements feedback through a trained prediction model that uses growth parameters and predicted data (including FWHM difference) to guide process optimization. The model provides feedback on the likely outcome of parameter settings, allowing operators to adjust conditions to achieve desired two-dimensional growth quality while maintaining high growth rates.
2Ease of operation
If traditional subjective human experience is used to regulate growth parameters, then operational simplicity is maintained, but optimal process window is missed leading to pre-reaction and quality deterioration
Solution Approach 1:
The patent applies self-service by implementing an automated prediction model that independently analyzes growth parameters and provides optimization recommendations without requiring subjective human judgment. The system serves itself by using trained algorithms to identify optimal process windows, eliminating the limitations of human experience while maintaining ease of operation through automated guidance.
Solution Approach 2:
The patent replaces the mechanical system of human subjective judgment with an automated prediction model based on machine learning. This substitution eliminates the inconsistency and limitations of human experience in regulating growth parameters, providing objective, data-driven optimization while keeping the system easy to operate through automated recommendations.
3Ease of manufacture
If pre-reaction is allowed to occur at high temperature, then nucleation points are formed promoting three-dimensional growth, but epitaxial layer quality is seriously deteriorated
Solution Approach 1:
The patent applies preliminary anti-action by using the prediction model to identify and prevent pre-reaction conditions before they occur. The model analyzes growth parameters and warns against conditions that would lead to pre-reaction, allowing operators to adjust parameters in advance to achieve controlled two-dimensional growth without the harmful effects of premature nucleation.
Solution Approach 2:
The patent converts the potential harm of pre-reaction into benefit by using the prediction model to identify the precise conditions where oxidation should occur. Instead of allowing uncontrolled pre-reaction, the system uses controlled oxidation at optimized parameters to promote beneficial two-dimensional growth, transforming what would be a harmful process into a useful one.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method achieves high-quality two-dimensional step-flow growth of beta-Ga2O3 epitaxial layers by adjusting growth parameters based on predicted data, reducing full width at half maximum (FWHM) differences and improving surface roughness, thereby enhancing the overall quality of the epitaxial layer.
Implementation Method 1
when beta-Ga 2 O 3 is epitaxially grown by Metal-organic Chemical Vapor Deposition (MOCVD)
Implementation Method 2
the oxidation of O 2 is stronger, which leads to the pre-reaction in high temperature environment
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
A prediction method and a device for epitaxial layer of beta-GazOs grown by MOCVD are provided. The method includes acquiring preset growth parameters, and inputting the preset growth parameters into a trained epitaxial layer prediction model, the preset growth parameters includes an O/Ga ratio, chamber pressure, temperature, and growth time; outputting predicted data of the beta-GazOs epitaxial layer corresponding to the preset growth parameters by the epitaxial layer prediction model, the predicted data at least includes full width at half maximum difference. In the present disclosure, In the present disclosure, the machine learning model is used to predict the epitaxial layer, and the preset growth parameters are adjusted according to the predicted data to select the growth parameters suitable for the process window of beta-GazOs epitaxial growth, and then the generated parameters are input into MOCVD for epitaxy. By combining machine learning with MOCVD process, high-quality two-dimensional step-flow growth with TMGa and O2 as reactants on unintentional oblique substrate is realized, and the quality of beta-Ga2O3 epitaxial layer grown by MOCVD is ensured.