Prospects for LED lighting
Prospects for LED lighting
复制标题
DOI:
10.1038/nphoton.2009.32
复制
发表时间:
2009-04-01
期刊:
影响因子:
35
通讯作者:
Nakamura, Shuji
中科院分区:
文献类型:
--
作者:
Pimputkar, Siddha;Speck, James S.;Nakamura, Shuji
Since the development of incandescent light bulbs in the late 1800s, various methods of producing white light more efficiently have been investigated. Of these, white-light sources based on light-emitting diodes (LEDs) look set to have a considerable impact on issues such as energy consumption, environment and even the health of individuals. Roughly 22% of the electricity generated in the United States is dedicated to lighting applications1. If all conventional whitelight sources in the world were converted to the energy-efficient LED light sources, energy consumption could be reduced by around 1,000 TW h yr–1, the equivalent of about 230 typical 500-MW coal plants, reducing greenhouse gas emission by about 200 million tonnes (ref. 2). White-light sources based on reliable and energy-efficient LEDs have only recently been made possible through developments in semiconductors (Fig. 1 shows a whitelight LED in action). It is possible to generate light in semiconductor materials (such as GaN or AlInGaP) by injecting electrons into the conduction band of the material and providing lower-energy sites (‘holes’) in the valence band into which they can fall, thereby creating light of a colour corresponding to the energy gap between the conduction band and valence band, also called the bandgap. A light-emitting diode is an electronic device integrating electrical access to the bandgap structure and allowing for efficient light generation. LEDs essentially consist of three different types of materials layered on top of each other. The bottom layer has a high concentration of free electrons (for example n-type GaN doped with Si) followed by multiple alternating thin layers (1–30 nm) of material with a smaller bandgap (InGaN/GaN), also called quantum wells. The sandwiching of a smaller-bandgap material (InGaN) between layers of larger-bandgap material (GaN) creates a well that spatially traps electrons and holes, allowing them to recombine efficiently, generating light with the wavelength of the smaller-bandgap material. Above this ‘active layer’, there is a layer of material with a high concentration of holes (p-type GaN doped with Mg). Until recently, the only high-luminosity LEDs available emitted red light. For white light, however, two or more wavelengths are required to generate a broad spectrum of light that is a better approximation of a blackbody radiation curve, such as that of the Sun. One way to produce additional colours is to use a material that absorbs light of one wavelength and emits at longer wavelength.Phosphors are commonly used for this task and a select few have received considerable attention, such as rare-earth-doped yttrium aluminium garnets (YAG: RE). For example, cerium-doped YAG can absorb blue and ultraviolet light and emit yellow light relatively efficiently3. Crucial to this process is the fact that higher-energy light (for example ultraviolet or blue) is converted to lower energy (for example yellow or red). Therefore, LEDs emitting red light cannot be used for white-light generation using phosphors; instead a short-wavelength ultraviolet, violet or blue LED is required. Early attempts to produce blue-emitting semiconductors focused on SiC, but these devices proved inefficient (0.03% efficiency4) owing to the material’s indirect bandgap. The GaN revolution has since provided efficient ultraviolet, violet and blue light emitters. GaN is a direct-bandgap semiconductor material with a 3.45-eV bandgap, which corresponds to near-ultraviolet light (364 nm). GaN was first investigated as a potential material for LEDs in the late 1960s by Paul Maruska and Jacques Pankove at the Radio Corporation of America (RCA) and in later years additionally by …