III-Nitride Lateral PN Junctions via Co-Doping and Selective Activation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Selective area doping of gallium nitride (GaN) and related semiconductors is challenging, particularly in creating patterned p- and n-type regions for vertical transistors, due to issues like implantation damage, high thermal budgets, and interfacial impurities from regrowth processes.
Innovation Solution
The method involves co-doping III-Nitride layers with n-type and p-type dopants, where the p-type dopant concentration is higher, and using passivating species like hydrogen to selectively activate or deactivate p-type dopants, creating p-n junctions without diffusion or ion-implantation techniques, allowing for the generation of p-type and n-type regions in the same layer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If implantation is used to create patterned p- and n-type regions, then selective area doping is achieved, but implantation damage and high thermal budget for activation anneal occur
Solution Approach 1:
The patent extracts the harmful implantation process entirely and replaces it with in-situ co-doping during MOCVD growth. The p-type dopant (Mg) and n-type dopant (Si) are introduced simultaneously during the epitaxial growth of the GaN layer, eliminating implantation damage and the need for high-temperature activation annealing.
Solution Approach 2:
The patent applies preliminary action by incorporating the p-type dopant (Mg) at high concentration during the initial growth stage, then using selective hydrogen removal (via plasma or thermal treatment) in specific regions to activate the p-type conductivity. This preliminary doping followed by selective activation avoids the need for implantation and subsequent high-temperature annealing.
2Manufacturing precision
If regrowth is used to create patterned p- and n-type regions, then selective area doping is achieved, but high interfacial impurities and additional process complexity occur
Solution Approach 1:
The patent merges the doping process with the epitaxial growth process itself. By using MOCVD to co-introduce p-type (Mg) and n-type (Si) dopants during the same growth run, the method eliminates the need for separate regrowth steps, reducing both process complexity and the risk of interfacial impurities that would arise from multiple growth cycles.
Solution Approach 2:
The MOCVD process serves multiple functions simultaneously: it grows the GaN layer, introduces p-type dopants, introduces n-type dopants, and enables selective activation through hydrogen removal. This multi-functionality replaces what would otherwise require separate implantation, regrowth, and annealing steps, significantly reducing process complexity.
3Adaptability or versatility
If co-doping with higher p-type dopant concentration is used, then p-n junctions can be formed, but selective activation of p-type dopants is required
Solution Approach 1:
The patent applies local quality by creating regions with different hydrogen concentrations within the same co-doped layer. Through selective plasma treatment or thermal processing of specific areas, hydrogen is removed only in regions where p-type activation is desired, while other regions retain hydrogen to keep the p-type dopant deactivated. This creates locally different electrical properties from a uniformly co-doped structure.
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 enables the creation of selectively patterned p-type and n-type regions within the same III-Nitride semiconductor layer, reducing process complexity and improving the performance of devices like vertical transistors by achieving activated p-type regions in specific areas while maintaining n-type behavior elsewhere.
Implementation Method 1
a passivating species, such that the passivating species deactivates an acceptor
Data Source
AI summary
Methods are provided of selectively obtaining n-type and p-type regions from the same III-Nitride layer deposited on a substrate without using diffusion or ion-implantation techniques. The III-Nitride layer is co-doped simultaneously with n-type and p-type dopants, with p-type dopant concentration higher than n-type dopant to generate p-n junctions. The methods rely on obtaining activated p-type dopants only in selected regions to generate p-type layers, whereas the rest of the regions effectively behave as an n-type layer by having deactivated p-type dopant atoms.