SiC Composite Substrate Mismatch Interface Adhesion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for bonding monocrystalline SiC and polycrystalline SiC substrates face challenges such as high cost, defects, and stress-induced bowing due to thermal expansion differences and high interfacial energy, leading to suboptimal quality and increased costs in manufacturing large-diameter SiC composite substrates with low defect density.

Innovation Solution

A method involving a mismatch interface between the monocrystalline SiC and polycrystalline SiC layers, where the polycrystalline SiC substrate has randomly oriented close-packed planes and a rugged surface, reducing internal stresses and preventing dislocation propagation into the monocrystalline SiC layer, achieved through mechanical working and chemical vapor deposition.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If monocrystalline SiC substrates are prepared by the improved Rayleigh method, then high purity and monocrystalline quality are achieved, but the preparation process becomes complex and costly

Engineering Contradiction:
Improvesubstrate qualityVSAvoidpreparation process complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The substrate is divided into two functional parts: a thin monocrystalline SiC layer (5-50 μm) for device fabrication and a thick polycrystalline SiC substrate for mechanical support. This segmentation allows each part to be optimized independently - the monocrystalline layer provides high quality for devices while the polycrystalline substrate provides mechanical strength, avoiding the need to manufacture entire large-diameter monocrystalline substrates which is complex and costly.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent combines monocrystalline SiC layer and polycrystalline SiC substrate into a composite structure. The monocrystalline layer is grown on the polycrystalline substrate using CVD or epitaxial methods, creating a bonded composite that leverages the advantages of both materials - the crystal quality of monocrystalline SiC and the mechanical properties of polycrystalline SiC.

Inventive Principle:
Principle #5Merging (Combining)

2Strength

If a thin monocrystalline SiC layer is bonded to a polycrystalline SiC substrate via ceramic or metal interposing layers, then bonding is achieved, but electrical conduction is blocked or metal impurities are introduced

Engineering Contradiction:
Improvebonding strengthVSAvoiddevice performance
Core Design Contradiction:
StrengthVSReliability

Solution Approach 1:

The patent removes the problematic intermediate layers (ceramic or metal bonding layers) entirely. Instead, it uses a direct bonding approach where the monocrystalline SiC layer is grown epitaxially on the polycrystalline SiC substrate, eliminating the need for separate bonding layers that would block electrical conduction or introduce impurities.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent uses a buffer layer composed of SiC with intermediate crystal structure (such as 3C-SiC between 4H-SiC and polycrystalline SiC) as a mediator. This buffer layer facilitates the transition between different crystal structures, reducing misfit dislocations and enabling direct bonding without ceramic or metal intermediaries, thus maintaining electrical conduction while preventing impurity introduction.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Strength

If polycrystalline SiC is deposited on an intermediate support with different thermal expansion coefficient, then substrate support is achieved, but thermal stress causes bowing

Engineering Contradiction:
Improvesubstrate supportVSAvoidsubstrate flatness
Core Design Contradiction:
StrengthVSShape

Solution Approach 1:

The patent changes the material parameter of the substrate from materials with different thermal expansion coefficients (such as sapphire or silicon) to polycrystalline SiC, which has a thermal expansion coefficient matching that of SiC. This parameter change eliminates thermal stress during heating and cooling cycles, preventing bowing while maintaining substrate support functionality.

Inventive Principle:
Principle #35Parameter changes

4Area of stationary object

If large diameter substrates are manufactured, then device application area is increased, but defect density increases due to stress and bowing

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

Solution Approach 1:

The patent segments the substrate into a thin monocrystalline layer (5-50 μm) and a thick polycrystalline substrate. The thin monocrystalline layer can be grown with high precision and low defect density using CVD or epitaxial methods, while the thick polycrystalline substrate provides mechanical support. This segmentation allows large diameter substrates to be manufactured with low defect density in the active monocrystalline region.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the thickness parameter of the monocrystalline layer to be thin (5-50 μm), which reduces the impact of stress and defects on device performance. Even if some defects occur in the bulk polycrystalline substrate, they do not propagate into the thin monocrystalline layer where devices are fabricated, maintaining low defect density in the active region.

Inventive Principle:
Principle #35Parameter changes

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 improved adhesion and reduced defect density, eliminating bowing and metal contamination, while allowing for cost-effective production of high-quality SiC composite substrates suitable for power semiconductor devices.

Implementation Method 1

depositing polycrystalline SiC on the rugged surface of the monocrystalline SiC layer by chemical vapor deposition

Methodology Applied
Scientific EffectChemical vapor deposition: Chemical Vapour Deposition

Data Source

PatentUS11208719B2SiC composite substrate and method for manufacturing same
Publication Date: 2021.12.28 SHIN ETSU CHEMICAL CO LTD
  • US11208719B2 patent drawing
  • US11208719B2 patent drawing
  • US11208719B2 patent drawing

AI summary

Provided is an SiC composite substrate 10 having a monocrystalline SiC layer 12 on a polycrystalline SiC substrate 11, wherein: some or all of the interface at which the polycrystalline SiC substrate 11 and the monocrystalline SiC layer 12 are in contact is an unmatched interface I12/11 that is not lattice-matched; the monocrystalline SiC layer 12 has a smooth obverse surface and has, on the side of the interface with the polycrystalline SiC substrate 11, a surface that has more pronounced depressions and projections than the obverse surface; and the close-packed plane (lattice plane 11p) of the crystals of the polycrystalline SiC in the polycrystalline SiC substrate 11 is randomly oriented with reference to the direction of a normal to the obverse surface of the monocrystalline SiC layer 12. The present invention improves the adhesion between the polycrystalline SiC substrate and the monocrystalline SiC layer.