GaN LED N-Layer Conductive Intervening Sublayers

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The challenge of growing high-quality GaN epitaxial layers on silicon substrates is hindered by lattice constant mismatch, thermal expansion differences, and defects, making it difficult to fabricate cost-effective blue LEDs, as GaN-based LEDs are typically grown on expensive substrates like sapphire.

Innovation Solution

The approach involves epitaxially growing a Low Resistance Layer (LRL) with a superlattice structure on a silicon substrate, using a ZnS buffer layer, and incorporating conductive AlGaN:Si intervening layers to reduce lattice defects and enhance current spreading, along with a novel wafer bonding process using a eutectic metal layer and reduced platinum usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If GaN is epitaxially grown on a silicon substrate, then manufacturing cost is reduced, but lattice defects increase due to lattice constant mismatch

Engineering Contradiction:
Improvemanufacturing costVSAvoidlattice defect density
Core Design Contradiction:
Ease of manufactureVSManufacturing precision

Solution Approach 1:

A multi-layer buffer structure comprising AlN, GaN, and AlGaN layers is introduced between the silicon substrate and the active GaN layer. This intermediary buffer structure gradually transitions the lattice constant from silicon to GaN, reducing dislocation density and improving crystal quality while enabling cost-effective silicon substrate usage.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The buffer structure uses a composite of different nitride materials (AlN, GaN, AlGaN) with varying lattice constants and aluminum compositions. This composite approach allows optimization of both mechanical stress management and lattice matching, achieving low defect density in the final GaN layer.

Inventive Principle:
Principle #40Composite materials

2Reliability

If the GaN layer is grown thick enough, then device performance improves, but stress-induced cracking occurs

Engineering Contradiction:
Improvedevice performanceVSAvoidstress resistance
Core Design Contradiction:
ReliabilityVSStrength

Solution Approach 1:

The aluminum composition in the AlGaN buffer layers is varied (e.g., 15%, 30%, 45% Al) to create a composition gradient. This parameter change allows progressive stress management through the buffer structure, enabling the growth of thick, high-performance GaN layers without stress-induced cracking.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The buffer structure is divided into multiple discrete layers with different thicknesses and compositions rather than a single continuous layer. This segmentation allows each layer to independently manage stress, preventing crack propagation through the entire structure while supporting thick active layers.

Inventive Principle:
Principle #1Segmentation

3Reliability

If multiple periods of platinum and titanium-tungsten are used as barrier metal layers, then tin penetration is prevented, but manufacturing cost increases

Engineering Contradiction:
Improvetin barrier effectivenessVSAvoidmanufacturing cost
Core Design Contradiction:
ReliabilityVSEase of manufacture

Solution Approach 1:

The complex multi-period platinum and titanium-tungsten barrier structure is replaced with a simplified single-layer titanium barrier. This extraction of unnecessary layers maintains the essential function of preventing tin penetration while dramatically reducing manufacturing complexity and cost.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The barrier metal design transitions from a multi-layer structure with varying thicknesses and compositions to a single titanium layer with optimized thickness (50-200 nm). This parameter change simplifies the manufacturing process while maintaining effective tin barrier performance through proper thickness control.

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 method enables the production of high-quality blue LEDs on silicon substrates with reduced lattice defects and improved current spreading, while lowering manufacturing costs by utilizing inexpensive silicon substrates and minimizing platinum usage.

Implementation Method 1

growing a first gallium-nitride layer over the zinc-sulfide buffer layer reduces a density of lattice defects in the first gallium-nitride layer

Methodology Applied
Scientific EffectLattice mismatch reduction:

Implementation Method 2

wafer bonding a carrier wafer structure to a device wafer structure by melting a layer of eutectic metal

Methodology Applied
Scientific EffectEutectic melting: Melting

Implementation Method 3

the aluminum-gallium-nitride sublayers provide a compressive strain to the gallium-nitride sublayers and prevent cracking

Methodology Applied
Scientific EffectCompressive strain:

Data Source

PatentUS9012939B2N-type gallium-nitride layer having multiple conductive intervening layers
Publication Date: 2015.04.21 SEOUL SEMICONDUCTOR
  • US9012939B2 patent drawing
  • US9012939B2 patent drawing
  • US9012939B2 patent drawing

AI summary

A vertical GaN-based blue LED has an n-type layer comprising multiple conductive intervening layers. The n-type layer contains a plurality of periods. Each period of the n-type layer includes a gallium-nitride (GaN) sublayer and a thin conductive aluminum-gallium-nitride (AlGaN:Si) intervening sublayer. In one example, each GaN sublayer has a thickness substantially more than 100 nm and less than 1000 nm, and each AlGaN:Si intervening sublayer has a thickness less than 25 nm. The entire n-type layer is at least 2000 nm thick. The AlGaN:Si intervening layer provides compressive strain to the GaN sublayer thereby preventing cracking. After the epitaxial layers of the LED are formed, a conductive carrier is wafer bonded to the structure. The silicon substrate is then removed. Electrodes are added and the structure is singulated to form a finished LED device. Because the AlGaN:Si sublayers are conductive, the entire n-type layer can remain as part of the finished LED device.