Biaxially Oriented SiC Substrate for Crack-Resistant Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
SiC substrates are difficult to process due to high hardness, leading to yield loss from breakings and cracks during grinding, polishing, and cutting, primarily caused by uneven distribution of basal plane dislocations.
Innovation Solution
A SiC substrate with a biaxially oriented SiC layer where basal plane dislocations (BPDs) are evenly distributed, reducing strains and minimizing continuous regions of high BPD density to prevent cracks during processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If SiC substrate is processed through grinding, polishing, and cutting, then the substrate can be manufactured into final products, but breakings and cracks occur due to high hardness and uneven dislocation distribution
Solution Approach 1:
The patent applies local quality by creating a specific layered structure where a first SiC layer with low dislocation density is positioned at the surface region (within 30μm from surface) while a second SiC layer with higher dislocation density is positioned at the deep region. This non-uniform distribution of dislocation density across different depths and regions optimizes both surface integrity during processing and overall substrate performance, directly resolving the contradiction between manufacturability and substrate integrity.
2Ease of manufacture
If dislocations are present in SiC single crystal, then epitaxial growth can proceed, but basal plane dislocations expand to stacking defects under energization, adversely affecting device reliability
Solution Approach 1:
The patent extracts and removes basal plane dislocations from the surface region by positioning a low dislocation density first SiC layer at the surface (within 30μm depth) where epitaxial growth occurs. This separation allows epitaxial growth to proceed while preventing BPDs from reaching the device formation region, thereby maintaining both manufacturability and device reliability.
Solution Approach 2:
The patent solves the dislocation problem by transitioning from a uniform 2D distribution to a 3D stratified distribution of dislocations. By controlling dislocation density as a function of depth (z-direction), BPDs are confined to deep regions away from the surface, allowing epitaxial growth while preventing stacking fault formation in the device region through vertical spatial separation.
3Device complexity
If BPDs are unevenly distributed in SiC substrate, then dislocation management is simplified, but strains in the crystal increase leading to breakings and cracks during processing
Solution Approach 1:
The patent applies local quality by creating a specific layered structure where a first SiC layer with low dislocation density is positioned at the surface region (within 30μm from surface) while a second SiC layer with higher dislocation density is positioned at the deep region. This non-uniform distribution of dislocation density across different depths and regions optimizes both surface integrity during processing and overall substrate performance, directly resolving the contradiction between manufacturability and substrate integrity.
Data Source
Figure 1~2(c)
Figure 3~4
AI summary
There is provided a SiC substrate that can reduce breakings and cracks that occur during substrate processing. This SiC substrate includes a biaxially oriented SiC layer. In an XRT image obtained by subjecting the biaxially oriented SiC layer to X-ray topography (XRT) measurement, when the entire XRT image is divided into a lattice pattern giving a region of 4 mm longitudinal length × 4 mm lateral length × 28 µm depth per square and an average value of a volume density of basal plane dislocations (BPDs) per square is defined as X (cm/cm3), regions having 5X (cm/cm3) or more per square do not extend for 10 or more continuous squares in a straight line in either a longitudinal or lateral direction.