Access Official Technical PDF

Share Summary:LinkedInEmail

White LEDs are commonly made by combining a blue-emitting LED with a phosphor that converts some of its output into broader yellowish light. A different approach is to generate multiple colors within one semiconductor structure. A Crosslight TCAD study examined a monolithic white LED based on InGaN/GaN/InGaN multiple quantum wells (MQWs), with separate wells intended to produce blue and yellow emission. The simulations suggest that barrier composition can influence the balance between those emissions—and therefore the resulting color—while also exposing design trade-offs that matter for further development.

Why generate white light within one LED?

An LED produces light through electroluminescence: electrons and holes recombine, and radiative recombination releases photons. The photon energy is related to the semiconductor band gap and the allowed energy states of the carriers. As a result, an LED can be engineered to emit at a selected wavelength, but producing white light requires combining emissions across different parts of the visible spectrum.

Two established approaches are additive mixing from multiple colored LEDs and blue LED emission combined with a phosphor. The first requires multiple emitters and is constrained by the difficulty of producing efficient green emission, often called the “green gap.” The phosphor approach avoids that particular limitation but adds a conversion layer and can have color-rendering limitations. A monolithic design aims to create more than one emission band within a single device, without relying on a phosphor coating or separate LEDs.

How the InGaN/GaN quantum-well design works

A quantum well is a thin semiconductor layer placed between materials with wider band gaps. This arrangement confines electrons and holes, creating allowed energy states that affect the energy of emitted photons. In an InGaN well, changing the indium composition changes the band gap and provides a way to target different emission wavelengths.

The proposed structure assigns one well near the p-side to blue emission and two other wells to yellow emission. The study selected a three-well design partly to limit simulation time and avoid adding too many wells. More wells may increase opportunities for radiative recombination, but they also add material through which current must pass, increasing internal resistance.

The modeled wells were set to 10 nm wide, with 5 nm barriers between them. The blue and yellow well compositions were adjusted to target approximate emission wavelengths of 470 nm and 580 nm, respectively. The final compositions used in the simulations differed from initial design estimates, reflecting the effects of quantum confinement and differences between material parameters used in calculations and those in the simulation file.

Barrier composition provides a color-tuning mechanism

In the simulated structure, the barrier between the blue well and the first yellow well influences how readily carriers pass between them. A higher barrier can restrict carrier transfer; changing the indium content changes the barrier height and can alter the relative contribution of blue and yellow emission. The two yellow wells were separated by a GaN barrier, while the blue-to-yellow barrier was varied.

The study compared three barrier compositions: pure GaN, In0.01Ga0.99N, and In0.02Ga0.98N. These were used to produce simulated cool, neutral, and warm white outputs, respectively. In the model, allowing more carriers to reach the yellow-emitting wells shifts the mixture toward warmer light. This is a design-level indication of tunability, not evidence that a finished device will achieve the same color response without further optimization and validation.

The neutral-emission simulation produced a color temperature of approximately 5914 K, near the study’s 6000 K target. The result was not an exact match: the output remained blue-shifted, and the study noted that stronger cyan or greenish emission could move the color closer to the target. Adding a green well was not pursued because of the green-gap challenge; adding a gradient of wells was also avoided because it would increase resistivity.

What the simulation does—and does not—establish

The work used Crosslight TCAD to investigate the device structure and emission behavior. Its results support the possibility of producing white light from a single multi-quantum-well structure, and illustrate how band engineering and carrier distribution can be used to influence the spectrum. The CIE 1931 color space provides a way to relate an emission spectrum to perceived color: tristimulus values are calculated from the spectrum and represented as chromaticity coordinates.

However, the simulation was shaped by simplifying design decisions. The well width was fixed, the bulk thickness was selected arbitrarily, and the model did not account for band shrinkage from doping. The study also identifies further optimization of well and barrier dimensions, doping, and bulk material as necessary. Its results should therefore be treated as an initial design exploration rather than a demonstration of production performance.

Practical implications for LED development

For semiconductor device teams, the proposed architecture highlights a useful set of interacting variables:

  • Well composition sets the target band gap and contributes to the emission wavelength.
  • Barrier composition and thickness affect carrier confinement and transfer between wells.
  • Well count and geometry influence recombination opportunities as well as resistance and design complexity.
  • Doping and material properties affect electrical behavior and may also influence the band structure.

These variables cannot be optimized independently. A design that improves the balance of color components may create other challenges, such as weaker confinement or higher resistance. In addition, the study identifies the cost of InGaN and GaN material growth as an important consideration: even a technically feasible phosphor-free structure must be assessed against the economics of manufacturing.

Conclusion

InGaN/GaN/InGaN quantum wells offer a simulated route to generating blue and yellow light within one monolithic LED. Adjusting the barrier between those wells changed the modeled color balance, bringing the neutral output close to the intended 6000 K target without reaching an exact match. The practical takeaway is that spectral tuning depends on coordinated control of composition, confinement, carrier flow, and electrical resistance. Further device-physics work, broader simulation studies, and manufacturing-cost analysis are needed before the approach can be judged as a practical alternative to phosphor-assisted white LEDs.

Access Official Technical PDF

About The Author