Semi-Insulating SiC Substrate Transfer for Low RF Loss III-N Epitaxy

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The existing methods for fabricating substrates for epitaxial growth of III-N alloys, such as gallium nitride, face challenges with high RF losses and poor heat dissipation, particularly due to the limitations of semi-insulating SiC substrates, which are expensive and available only in limited sizes, and composite structures that introduce thermal barriers.

Innovation Solution

A process involving the transfer of a single-crystal semi-insulating SiC layer from a donor substrate to a receiver substrate, followed by the formation of an additional semi-insulating SiC layer, allowing direct contact and minimizing thermal barriers, while using a temporary carrier to optimize the orientation and reduce dislocations, thereby creating a low-cost, high-electrical resistivity substrate with enhanced heat dissipation and reduced RF losses.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Temperature

If semi-insulating SiC substrates are used for heteroepitaxy of III-N alloys, then thermal conductivity is improved and heat dissipation is enhanced, but substrate cost increases and substrate size is limited

Engineering Contradiction:
Improveheat dissipationVSAvoidsubstrate cost and size
Core Design Contradiction:
TemperatureVSEase of manufacture

Solution Approach 1:

The substrate structure is segmented into multiple functional layers: a large-area low-cost polycrystalline SiC substrate provides mechanical support and heat dissipation, while a transferred thin layer of single-crystal semi-insulating SiC provides the epitaxial growth interface. This segmentation allows each layer to optimize its function without requiring the entire substrate to have expensive single-crystal properties.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

A bonding layer is introduced as an intermediary between the polycrystalline SiC substrate and the transferred single-crystal SiC layer. This bonding layer facilitates the transfer process and creates a stable interface, enabling the combination of low-cost polycrystalline material with high-performance single-crystal material.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of manufacture

If composite structures with single-crystal silicon or SiC layer on polycrystalline SiC substrate are used, then substrate cost is reduced and substrate size is increased, but thermal barrier is formed and heat dissipation is hindered

Engineering Contradiction:
Improvesubstrate cost and sizeVSAvoidheat dissipation
Core Design Contradiction:
Ease of manufactureVSTemperature

Solution Approach 1:

The invention changes the material parameters at the interface by using a bonding layer that allows direct thermal contact between the single-crystal SiC layer and polycrystalline SiC substrate. This eliminates the thermal barrier effect that would occur with traditional oxide interfaces, maintaining high thermal conductivity while enabling the use of low-cost polycrystalline substrates.

Inventive Principle:
Principle #35Parameter changes

3Adaptability or versatility

If heteroepitaxy is performed on substrates with large lattice parameter difference or thermal expansion difference, then material availability is improved, but crystal defects increase and mechanical stresses are generated

Engineering Contradiction:
Improvematerial availabilityVSAvoidcrystal quality
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

Instead of directly growing III-N alloys on substrates with large lattice mismatches (which would create defects), the invention creates a copied interface by transferring a thin layer of single-crystal semi-insulating SiC onto a polycrystalline SiC substrate. This copied single-crystal interface provides a high-quality epitaxial surface while the bulk substrate can be made from abundant, low-cost polycrystalline material.

Inventive Principle:
Principle #26Copying

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 process enables the creation of a substrate with minimized RF losses and maximized heat dissipation, suitable for high-frequency, high-power electronic devices, using a low-cost, large-sized substrate with optimal semi-insulating SiC quality, avoiding thermal barriers and dislocation issues.

Implementation Method 1

implanting ionic species into the donor substrate so as to form a weakened region that defines a thin layer of single-crystal semi-insulating SiC to be transferred

Methodology Applied
Scientific EffectIonic species implantation: Ion Implantation

Implementation Method 2

bonding the donor substrate to a first receiver substrate via a bonding layer

Methodology Applied
Scientific EffectBonding: Welding

Data Source

PatentUS20230411140A1Method for producing a substrate for epitaxial growth of a gallium-based iii-n alloy layer
Publication Date: 2023.12.21 SOITEC SA
  • US20230411140A1 patent drawing
  • US20230411140A1 patent drawing
  • US20230411140A1 patent drawing

AI summary

A method of producing a substrate for epitaxial growth of a gallium-based III-N alloy layer comprises the following consecutive steps: —providing a donor substrate of semi-insulating monocrystalline silicon carbide, —implanting ionic species in the donor substrate so as to form a zone of weakness defining a thin layer of semi-insulating monocrystalline SiC to be transferred, —bonding the donor substrate to a first receiving substrate by means of a bonding layer, —detaching the donor substrate along the zone of weakness so as to transfer the thin layer of semi-insulating monocrystalline SiC on to the first receiving substrate, —forming an additional layer of semi-insulating SiC on the transferred thin layer, —bonding the additional layer to a second receiving substrate having a high electrical resistivity, —removing at least a portion of the bonding layer so as to detach the first receiving substrate and expose the layer of transferred semi-insulating monocrystalline SiC.