In their continuing pursuit of a replacement for the incandescent light bulb, researchers at Princeton University in New Jersey and at the University of Southern California in Los Angeles, with the support of Universal Display Corp. of Ewing, N.J., and the US Department of Energy (DoE), have reported the development of white organic LEDs (OLEDs) with a total external quantum efficiency of 18.4 percent and a total power efficiency of 23.8 lm/W at a luminance of 500 cd/m2. To maximize performance, the devices incorporate a fluorescent dopant to generate blue light from singlet excitons and phosphorescent dopants to generate green and red light from triplet excitons.The white organic LED employs phosphors to generate green and red light from triplet excitons and a fluorophore to generate blue light from singlet excitons. Courtesy of Stephen R. Forrest. Stephen R. Forrest, who is relocating his team from Princeton to the University of Michigan in Ann Arbor, and who has assumed the positions of vice president for research and professor of electrical engineering and computer science and of physics at the institution, explained that the work is driven by two simple facts. “The incandescent bulb is 135 years old, and it has seen few improvements in efficiency in the last 120 years,” he said. That lack of improvement has significant practical consequences. According to a 2002 report produced for the DoE, 63 percent of lamps in the US are incandescent sources, which have a power efficiency of approximately 15 lm/W. The cumulative effect of this inefficiency becomes stark when it is recognized that they consume 42 percent of the electricity produced for lighting in the country, yet return only 12 percent of the generated light. As a result, significant benefits in terms of energy savings are to be had by the replacement of the incandescent bulb, which would translate to reduced carbon emissions and to significant financial rewards. Because they promise high efficiencies and manufacturing by spin-casting or ink-jet printing from low-cost organics, OLEDs are considered particularly attractive alternatives, and a 2005 report for the DoE calculated an annual energy savings of 215 GW with the replacement of incandescent and fluorescent lamps with OLEDs. When a voltage is applied to an OLED, the injected electrons and holes combine to form higher-energy excitons with a spin of zero, called singlets, and lower-energy excitons with a spin of one, called triplets. Although the exact ratio is a matter of debate, spin statistics state that 75 percent of the excitons will be triplets and that 25 percent will be singlets.Because they can harvest both triplets and singlets, phosphorescent dopants promise to enable OLEDs with an internal quantum efficiency of 100 percent and, thus, the greatest savings. Fluorescent materials, in contrast, can use only singlets. A problem limiting the development of electrophosphorescent OLEDs that generate white light by mixing red, green and blue, however, has been the lack of a good blue phosphor. The researchers have circumvented that issue by embracing the correlation between exciton spin statistics and the contribution to a white-light spectrum from blue light. Forrest recounted that he and collaborator Mark E. Thompson of the University of Southern California were struck with the idea at a conference early last year. “Good white light is about 25 percent blue,” he explained, “so we thought, ‘Why not do it with fluorescence and make the green and red with phosphors?’ ” In the new OLEDs, they accomplish this using a blue fluorophore in a layer spatially separated by undoped spacers from the layers with the red and green phosphors. The singlet excitons thus are transferred only to the fluorophore, and the triplets are transferred to the phosphors. The approach provides a power efficiency 20 percent higher than all-phosphorescent OLEDs because it avoids the exchange energy loss in converting a singlet in the conductive organic host to a triplet in a phosphorescent dopant.Forrest said that employing multiple emissive layers and spacers offers an additional benefit in that it is possible to attain a higher brightness at a high efficiency simply by increasing the voltage because the density of excitons across the emission zone is reduced. Separating the transfer channels further results in a stable output color. In tests, the devices yielded a color-rendering index of 85 (a value of 100 corresponds to an ideal white-light source) at applied current densities of 1, 10 and 100 mA/cm2, and they shifted CIE coordinates only from (0.40, 0.41) to (0.38, 0.40) with an increase from 1 to 100 mA/cm2. Although speculating that white OLEDs such as these could find a niche in specialty architectural and consumer applications in two years, he noted that substrate and electrode issues currently drive costs beyond what the market will bear for general illumination. The ITO-coated glass used in the demonstration devices is too brittle, heavy and expensive, and a move to plastics would require the development of costly moisture barriers to prevent the degradation of the OLED materials. Metal foils might offer a low-cost solution, he said, but the issues associated with ITO would remain. The scientists calculate that a proper choice of dopants would yield white OLEDs with a total external quantum efficiency and a total power efficiency of 34 percent and 60 lm/W, respectively. Forrest said that the team plans to work on improving the efficiency of the device and to explore issues involving exciton transport and light extraction.Nature, April 13, 2006, pp. 908-912.