Seed Substrate Layer Stack for Low-Defect III-Nitride Epitaxy

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing group III nitride substrates such as AlN and GaN have high crystal defects, low quality, and high costs, limiting their widespread use in devices like LEDs, lasers, and high-frequency devices due to thermal stress, contamination, and lattice constant differences.

Innovation Solution

Optimize the composition and thickness of encapsulating, planarizing, and seed crystal layers to minimize thermal expansion coefficient differences, use a stress-adjusting layer, and employ thin-film transfer of Si single crystals with low oxidation-induced stacking faults to reduce defects and stress.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If sublimation method is used to manufacture AlN single-crystal substrates, then dislocation density is reduced to 5 cm−2, but substrates are colored by carbon and metallic impurities, have low resistivity and low UV transmission, and are extremely expensive

Engineering Contradiction:
Improvedislocation densityVSAvoidcarbon and metallic impurity contamination
Core Design Contradiction:
Manufacturing precisionVSObject-generated harmful factors

Solution Approach 1:

The invention divides the substrate system into multiple functional layers: a sacrificial AlN layer that provides initial growth support, an intermediate layer that prevents impurity diffusion, and a final AlN layer that serves as the seed crystal. This segmentation isolates the seed crystal from contamination sources while maintaining low dislocation density through controlled layer-by-layer growth.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The invention uses a sacrificial AlN layer that is intentionally designed to be consumed or removed during the growth process. This disposable layer provides necessary support during manufacturing but is discarded afterward, eliminating the need for expensive, contamination-prone substrates while maintaining high crystal quality.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

2Area of stationary object

If heteroepitaxial growth is performed on sapphire substrates, then larger size and higher crystal quality can be achieved, but lattice constant differences result in numerous crystal defects and cracks

Engineering Contradiction:
Improvesubstrate sizeVSAvoidcrystal defect density
Core Design Contradiction:
Area of stationary objectVSManufacturing precision

Solution Approach 1:

The invention changes the material parameter by using AlN as the sacrificial layer instead of sapphire, matching the lattice constant of the target AlN crystal. This parameter matching eliminates lattice mismatch defects while maintaining the ability to grow large-area high-quality crystals through the AlN-based layer structure.

Inventive Principle:
Principle #35Parameter changes

3Productivity

If conventional seed substrates are used for epitaxial growth, then growth can proceed, but thermal expansion coefficient differences cause thermal stress and contamination during deposition

Engineering Contradiction:
Improveepitaxial growth capabilityVSAvoidthermal stress and contamination
Core Design Contradiction:
ProductivityVSObject-generated harmful factors

Solution Approach 1:

The invention addresses thermal expansion issues by using AlN layers throughout the structure, ensuring matching thermal expansion coefficients between all layers. This eliminates thermal stress during temperature cycling in the deposition process while maintaining productivity through compatible material properties.

Inventive Principle:
Principle #37Thermal expansion

Solution Approach 2:

The intermediate layer acts as a mediator between the sacrificial AlN layer and the final seed crystal layer. It prevents contamination diffusion from lower layers while accommodating thermal stress, enabling productive epitaxial growth without the harmful effects of direct contact between incompatible materials.

Inventive Principle:
Principle #24Intermediary (Mediator)

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

Produces high-quality, low-cost seed substrates with few defects for epitaxial growth, suitable for deep ultraviolet LEDs and high-frequency 5G applications with reduced warpage and voids, enhancing device characteristics and yield.

Implementation Method 1

A seed substrate with a minimized thermal expansion coefficient difference between layers, optimized composition and thickness of encapsulating, planarizing, and seed crystal layers, and the use of Si single crystals with low oxidation-induced stacking faults (OSF) to reduce thermal stress and contamination

Methodology Applied
Scientific EffectThermal expansion coefficient matching: Thermal Expansion

Implementation Method 2

A 0.1 to 1.5 μm thin film of Si single crystal with oxidation-induced stacking faults (OSF) of 10 defects/cm2 or less was transferred to the top surface of the planarizing layer and used as the seed crystal layer

Methodology Applied
Scientific EffectThin-film transfer:

Implementation Method 3

Seed substrate for epitaxial growth use and method for manufacturing same, and semiconductor substrate and method for manufacturing same

Methodology Applied
Scientific EffectEpitaxial growth: Epitaxy

Data Source

PatentUS12618172B2Seed substrate for epitaxial growth use and method for manufacturing same, and semiconductor substrate and method for manufacturing same
Publication Date: 2026.05.05 SHIN ETSU CHEMICAL CO LTD
  • US12618172B2 patent drawing
  • US12618172B2 patent drawing

AI summary

A seed substrate for epitaxial growth has a support substrate, a planarizing layer of 0.5 to 3 μm provided on the top surface of the support substrate, and a seed crystal layer provided on the top surface of the planarizing layer. The support substrate includes a core of group III nitride polycrystalline ceramics and a 0.05 to 1.5 μm encapsulating layer that encapsulates the core. The seed crystal layer is provided by thin-film transfer of 0.1 to 1.5 μm of the surface layer of Si<111> single crystal with oxidation-induced stacking faults (OSF) of 10 defects/cm2 or less. High-quality, inexpensive seed substrates with few crystal defects for epitaxial growth of epitaxial substrates and solid substrates of group III nitrides such as AlN, AlxGa1-xN (0<X<1) and GaN are obtained.