Free-Standing III-N Substrate via Two-Stage Epitaxy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for producing free-standing III-N layers face challenges such as high defect densities, impurity diffusion, and limited availability due to complex and time-consuming processes, which hinder the production of high-quality substrates suitable for advanced semiconductor devices.
Innovation Solution
A process involving the deposition of a first III-N layer at a low temperature on an Li(Al,Ga)Ox substrate using Molecular Beam Epitaxy (MBE), followed by a second layer at a higher temperature using Hydride Vapor Phase Epitaxy (HVPE), significantly reduces impurity diffusion and defect density, enabling the production of high-quality, large-area free-standing III-N substrates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a single high-temperature HVPE process is used to grow thick GaN layers, then productivity is improved, but impurity diffusion and defect density increase
Solution Approach 1:
The growth process is divided into two distinct stages: first a low-temperature MBE stage to form a high-quality template layer, then a high-temperature HVPE stage to rapidly grow the thick GaN layer. This segmentation allows each stage to optimize for its specific function, resolving the contradiction between growth rate and quality.
Solution Approach 2:
A thin high-quality GaN template layer is grown by MBE before the main HVPE growth. This preliminary layer serves as a defect-filtering interface that prevents impurity diffusion into the subsequent thick layer grown by HVPE, enabling high productivity without sacrificing quality.
2Device complexity
If conventional single-step HVPE is used, then the process is simple, but the defect density is high
Solution Approach 1:
The conventional single-step HVPE process is segmented into two separate processes: MBE for template formation and HVPE for bulk growth. This increases process complexity slightly but dramatically improves crystal quality by preventing impurity diffusion.
Solution Approach 2:
The process uses different temperature parameters for different stages: low temperature (≤700°C) for MBE template growth to ensure high quality, then high temperature for HVPE to achieve rapid growth. This parameter optimization resolves the quality-complexity contradiction.
3Ease of manufacture
If LiAlO2 substrate is used for GaN growth, then substrate availability is improved, but lithium diffusion into the GaN layer occurs
Solution Approach 1:
A thin GaN template layer is grown by MBE on the LiAlO2 substrate before the main HVPE growth. This preliminary layer acts as a diffusion barrier that prevents lithium from the substrate from contaminating the bulk GaN layer, while still allowing the process to use readily available LiAlO2 substrates.
Solution Approach 2:
The thin MBE-grown GaN layer serves as an intermediary barrier between the LiAlO2 substrate and the bulk GaN layer. It mediates the interface interaction, blocking lithium diffusion while maintaining the benefits of using LiAlO2 substrates.
4Ease of manufacture
If mechanical polishing is used to remove substrate, then free-standing layers are obtained, but the process is time-consuming and reduces crystal quality
Solution Approach 1:
The process exploits the natural thermal expansion mismatch between GaN and LiAlO2 substrates. During cooling after HVPE growth, the GaN layer self-detaches from the substrate without requiring mechanical polishing or chemical etching, saving time and preserving crystal quality.
Solution Approach 2:
The substrate removal utilizes differential thermal contraction: after high-temperature HVPE growth, cooling causes the GaN layer to contract more than the LiAlO2 substrate, creating stress that causes the GaN layer to delaminate and become free-standing, eliminating the need for time-consuming mechanical removal processes.
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
This approach results in III-N substrates with low impurity concentrations, reduced defect densities, and increased crystal quality, allowing for the production of substrates that are both efficient and scalable, suitable for advanced semiconductor applications.
Implementation Method 1
depositing on an Li(Al,Ga)Ox substrate, where 1 ≤ x ≤ 3, a first III-N layer at a first temperature
Implementation Method 2
depositing on the first III-N layer a second III-N layer at a second temperature
Implementation Method 3
such that during deposition at the first temperature, contaminants in the substrate, such as Li and O, diffuse to a lesser extent into the first III-N layer
Data Source
Figure 1a~1e
Figure 2
Figure 3
AI summary
A process for producing a free-standing III-N layer, where III denotes at least one element from group III of the periodic system, selected from Al, Ga and In, comprises depositing on a Li(Al,Ga)Ox substrate, where x is in a range between 1 and 3 inclusive, at least one first III-N layer by means of molecular beam epitaxy. A thick second III-N layer is deposited on the first III-N layer by means of a hydride vapor phase epitaxy. During cooling of the structure produced in this way, the Li(Al,Ga)Ox substrate completely or largely flakes off the III-N layers, or residues can be removed if necessary, by using etching liquid, such as aqua regia. A free-standing III-N substrate being substantially free of uncontrolled impurities and having advantageous properties is provided.