Susan Gormley | Kubos Semiconductor: Why is it so hard to make efficient green LEDs? See the data that shows how a different crystal structure requires almost 40% less of the key ingredient to hit green wavelengths.
07:43 - 09:12
Other snippets from this talk
Summary of the clip:
Why is it so hard to make efficient green LEDs? See the data that shows how a different crystal structure requires almost 40% less of the key ingredient to hit green wavelengths.
The "green gap" is a long-standing problem in the LED industry where the efficiency of conventional hexagonal Indium Gallium Nitride (InGaN) LEDs drops dramatically when trying to produce green light. Reaching these longer wavelengths requires incorporating a high percentage of indium into the quantum wells. This high indium content not only introduces material defects but also worsens the Quantum Confined Stark Effect (QCSE), crippling the device's light-generating efficiency.
Data from photoluminescence (PL) studies on Kubos's cubic GaN quantum wells demonstrates a clear solution. A plot of PL intensity versus quantum well thickness shows a very weak dependency. This is a critical result, as it implies the internal electric fields that cause QCSE in hexagonal GaN have been successfully eliminated in the cubic structure, allowing for greater design freedom and maintaining high efficiency even in thicker wells designed for longer wavelengths.
The most compelling evidence is a direct quantitative comparison. The data shows a cubic GaN sample emitting green light at 540 nm with only approximately 10% indium content in its quantum well. In stark contrast, a standard hexagonal GaN LED requires around 17% indium just to achieve blue emission at 450 nm. This proves that the cubic GaN platform provides a much more efficient and stable pathway to reaching green and red wavelengths, directly addressing the core material science challenge of the green gap.
In this short video, you can learn:
* Why high indium content kills efficiency in traditional hexagonal GaN green LEDs.
* How stable photoluminescence intensity across different quantum well thicknesses proves the absence of QCSE.
* The quantitative advantage: achieving green emission with significantly less indium compared to what's needed for blue in hexagonal GaN.
๐ **Clip Abstract** This clip presents compelling data on how cubic GaN quantum wells can solve the infamous "green gap" in LED technology. By requiring significantly less indium to achieve green emission, this approach avoids the efficiency-killing effects that plague conventional hexagonal GaN.
๐ Link in comments ๐
#CubicGaN, #GreenGap, #QuantumConfinedStarkEffect, #InGaNLEDs, #MicroLEDDisplays, #ARdisplays
This is a highlight of the presentation:
Next generation material for microLEDs; the relevance of cubic-GaN
More Highlights from the same talk.
03:55 - 06:26
How can changing a crystal's symmetry from hexagonal to cubic eliminate the biggest roadblock for efficient green and red microLEDs?
How can changing a crystal's symmetry from hexagonal to cubic eliminate the biggest roadblock for efficient green and red microLEDs?
The fundamental problem with conventional Gallium Nitride (GaN) used in blue LEDs is its hexagonal crystal structure. This asymmetry creates strong internal electric fields due to piezoelectric and spontaneous polarization. These fields cause a phenomenon known as the Quantum Confined Stark Effect (QCSE), which physically pulls the electron and hole wavefunctions apart within the quantum well, severely reducing the probability of efficient light-generating recombination. This issue becomes progressively worse as more indium is added to the alloy to push the emission to green and red wavelengths.
Kubos Semiconductor's solution is to grow GaN in its metastable cubic phase. The high symmetry of the cubic crystal structure inherently eliminates these internal electric fields. As shown in the band gap diagrams, this results in flat energy bands within the quantum well, maximizing the spatial overlap between electrons and holes. This directly translates to a much higher probability of radiative recombination and, therefore, higher device efficiency, especially at the longer wavelengths needed to close the "green gap."
This fundamental change unlocks several other critical performance benefits. The absence of QCSE means thicker quantum wells can be grown without losing efficiency, providing a larger active volume. Furthermore, cubic GaN exhibits reduced efficiency droop at high drive currents and significantly less spectral drift (wavelength shift) with changing current. These factors are crucial for creating stable, high-brightness microLED displays for applications like AR/VR that demand high performance.
In this short video, you can learn:
* The fundamental physics behind the Quantum Confined Stark Effect (QCSE) in hexagonal GaN.
* How cubic GaN's crystal symmetry eliminates internal electric fields to enhance electron-hole overlap.
* The multiple performance benefits of cubic GaN, including higher efficiency, reduced droop, and better wavelength stability.
๐ **Clip Abstract** This clip explains the core physics advantage of cubic Gallium Nitride (GaN) over the conventional hexagonal form for LED applications. By eliminating the Quantum Confined Stark Effect, cubic GaN dramatically improves radiative efficiency, paving the way for high-performance green and red microLEDs.
๐ Link in comments ๐
#CubicGaN, #QuantumConfinedStarkEffect, #MicroLEDs, #GreenGap, #ARVRDisplays, #DisplayTechnology
09:12 - 11:10
Growing a perfect cubic crystal on a standard silicon wafer is a massive strain engineering challenge. How is it done without the crystal cracking or relaxing into the wrong phase?
Growing a perfect cubic crystal on a standard silicon wafer is a massive strain engineering challenge. How is it done without the crystal cracking or relaxing into the wrong phase?
The primary challenge in growing high-quality cubic GaN is heteroepitaxyโdepositing crystalline layers on a substrate with a different lattice structure and thermal properties. Growing directly on a large, cost-effective silicon wafer introduces significant stress from both lattice mismatch and thermal expansion coefficient mismatch. If not managed, this strain leads to a high density of performance-killing defects and can even cause the wafer to crack or bow, making it unusable for manufacturing.
Kubos's proprietary approach involves a multi-layer buffer stack engineered to manage this strain. The process starts with a standard (001) silicon wafer, which is compatible with CMOS foundries. A key enabling layer of 3C-Silicon Carbide (3C-SiC) is grown first, for which Kubos holds an exclusive license. This is followed by an Aluminum Gallium Nitride (AlGaN) buffer layer, which provides the final template for growing the cubic GaN device structure on top.
Evidence of successful strain management is provided by X-Ray Reciprocal Space Mapping (RSM), a powerful characterization technique. The RSM plot shows the diffraction peaks for the cubic GaN and the AlGaN buffer layers are perfectly aligned on a vertical line. This alignment confirms that the layers have grown coherently and remain under compressive strain, exactly as designed, without relaxing into a lower-quality state. This precise control over the crystal growth is essential for maintaining the pure cubic phase and achieving high material quality and device yield.
In this short video, you can learn:
* The core challenges of lattice and thermal mismatch in growing GaN on silicon.
* The specific multi-layer stack (Si -> 3C-SiC -> AlGaN -> GaN) used to control strain.
* How to interpret a Reciprocal Space Map (RSM) to verify that the crystal layers are strained and not relaxed.
๐ **Clip Abstract** This clip delves into the critical materials engineering required to grow high-quality cubic GaN on silicon wafers. It details the proprietary buffer layer stack used to manage crystal strain and presents X-ray diffraction data as proof of successful, unrelaxed epitaxial growth.
๐ Link in comments ๐
#CubicGaN, #StrainEngineering, #Heteroepitaxy, #ReciprocalSpaceMapping, #MicroLEDs, #ARdisplays