Light Constructions - Stacked LED structure combines red, green and blue emitters

From OE Reports Number 166 - October 1997
01 October 1997
By Sunny Bains

Princeton researchers have taken a major step towards an integrated full-color LED array. They have demonstrated a device that can produce light from three different materials at once, each independently tunable. Though there are serious problems with their current demonstrators, the new LED structure has fabrication, resolution, and functional advantages that could make it an important stepping stone to a commercial product.

A long time coming

New display technologies seem to pop up and shout for attention every day. Though many are hugely exciting, few have the potential to completely change the face of the industry. One that undoubtedly does is the light-emitting diode (or device). There has been a lot of hype on this subject, but for once it may well be deserved. If the technology reaches its full potential, it may succeed the cathode ray tube as the most common type of computer monitor (an event that would be long overdue). And this replacement of huge heavy boxes with thin, lightweight, low-power screens could be just the beginning. The LED array is not just a light modulator, it is an efficient light source, and therefore has applications in all sorts of areas of optoelectronics.

In the past 10 years or so, three major advances have taken LEDs out of the digital watch to the threshold of the high-quality display market. Two of these advances were materials related. The first was the demonstration, at Eastman Kodak Co. about ten years ago, of electroluminescent polymers: organic materials that emit light when a voltage is applied across them. Conventional crystalline light-emitting diode materials could only be manufactured on silicon wafers, thus limiting their size and increasing their weight. Organic materials, on the other hand, can be coated over large areas and, because the active layers are flexible, the substrates can be made of plastic as well as glass.

The first blue organic LEDs, which started to appear in the early nineties, were also crucial. First, they allowed LEDs to make up color "triads": sets of red, green, and blue pixels that can together produce white and, potentially, millions of other colors. More importantly, these triads could be extremely efficient.

Conventional laptop computer screens, based on liquid crystal technology, are subtractive before being additive. They are backlit with white light and the red, green, and blue pixels are created by using arrays of color filters. The RGB light is then recombined (as triad) to produce white light again. Even pixels that are "off" are backlit and the unwanted light is filtered out. As a result, such displays are extremely inefficient. When all the pixels are bright, at least a third of the light produced (and the energy used) is just thrown away. And that's the best case. If all the pixels are off, the display consumes exactly the same amount of energy and 100 percent of the light is discarded.

Blue LEDs are important because, when integrated with red and green devices, they can be combined to produce a display where each color can be produced as needed. Where a red, green, or blue photon is not needed, it need not be produced. This is efficiency.

Putting it all together

The third step, of course, is to actually build a three-color array that works. This has been tricky. In order to work independently, each LED has to have three things: a front electrode (usually in common with its neighbors), a layer of electroluminescent material, and then a second electrode that is unique to it. The fact that the red, green, and blue devices had to be deposited sequentially caused serious problems. The first set fabricated would be eaten away by the chemicals used to process the second set of LEDs, and both would be affected by the fabrication of a third set. Basically, such displays just didn't work well.

One group that has been trying to tackle this, as well as other problems related to organic LEDs, is at Princeton University's Advanced Technology Center for Photonics and Optoelectronic Materials (known as POEM). There, researchers have tried to tackle this problem by designing an LED structure that would protect the vulnerable emissive layer: entirely encasing it in a tougher material. Though they produced a design that worked, (see figure 1)1 they soon came up with an even better idea. Rather than produce three sets of single color devices, they could produce one set of three-color devices: each device made of a stack of three light-emitting layers and four electrodes.2-3

The stacked, organic light emitting device (SOLED) that the Princeton group came up with is shown in figure 2. The bottom electrode, next to the glass or plastic substrate that each device is built on, is made of indium tin oxide (ITO). This material is commonly used in displays because it is transparent, but can only be processed using chemicals that tend to harm organics. It could be used as a first electrode, therefore, but not a second, third, or fourth without protective layers in between. Instead, the Princeton team chose to use metal electrodes, grown in such thin layers that they were semi-transparent. The color of the SOLED emissions could be controlled by varying the voltages across each of the layers. Because the green and red voltages are interdependent, these colors have to be tuned together. They need not be the same, but they have to be adjusted with respect to one another in order to get the right color combination out.

In terms of resolution, this structure has a huge advantage. Whereas before each white pixel had to be made up of a triad of colored devices, now a pixel could consist of just one LED. This means triple the number of pixels for the same number of devices. Fabrication is also much more straightforward with a homogeneous array of emitters.

Initial experiments have been partially successful. As the relevant voltages were set, light was emitted by each organic layer, passed through the metal layers, and eventually emerged through the glass front of the display. However, the color of the light had been distorted by the device structure. In one experiment, light from the blue and green electroluminescents became almost indistinguishable. The red was shifted too.

The problem appears to arise from the fact that each stack is, itself, much more complicated than a conventional LED device. When they examined their results, Princeton researchers found that this color distortion seemed to be caused by microcavities formed in between the many layers of the SOLED structure. Figure 3 shows the kind of effects that these can cause: three green layers can produce completely different spectra depending on where they are in the stack. Researchers are now in the process of modeling these effects. With luck, this optimization process, combined with some new materials, should produce a true RGB LED.

If they succeed, Princeton researchers may significantly increase the size of the high-quality display market, as well as taking it by storm. Because their devices are transparent, the RGB LEDs can be used as head-up displays; their industrial collaborators, Universal Display Corporation in Bala Cynwyd, Pennsylvania, are already aiming sheets of single-color LEDs at this application.


1. C. C. Wu, J. C. Sturm, R. A. Register and M. E. Thompson, Integrated three-color organic light-emitting devices, Appl. Phys. Lett. 69 (21), p 3117, 18 November 1996.

2. Zilan Shen, Paul E. Burrows, Vladimir Bulovi, Stephen R Forrest, and Mark E Thompson, Three-color, tunable, organic light-emitting devices, Science 276, p. 2009, 27 June 1997.

3. P. E. Burrows, G. Gu, V. Bulovi, Z Shen, S R Forrest, and M E Thompson, Achieving full-color organic light emitting devices for lightweight, flat-panel displays, IEEE Transactions on Electron Devices 44 (8), p. 1188, August 1997.

Sunny Bains
Sunny Bains is a technical journalist based in Edinburgh, UK.

Recent News
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?