Fabrication of InGaP light emitting diodes on a graded buffer substrate

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Materials Science and Engineering, 2007.Includes bibliographical references (leaf 33).Introduction: Computer display panels create a vast color palette by combining color from three light emitting diodes (LEDs), each producing red, green, or blue light. The light from these three LEDs is chosen so that the display can generate the largest color combination possible. Tri-color LED displays typically use one of two combinations of LEDs. In the first, InGaN blue (480 nm), AlGaAs red (637 nm), and GaP green (572 nm). This combination provides a full color spectrum, but is limited in its display of deep green colors1. The second option uses substitutes the GaP green for InGaN green (515 nm). This shift to 515 nm, a more yellowish green, increases the depth of greens and reds in the display. However, the incorporation of a green LED with a wavelength of 530 nm in current LED displays would result in a color range and color depth that would significantly exceed anything currently available. Such a technology would increase the richness of displays and would also make the mixing of green colors less power intensive given the availability of 'true green' in the display. Another technologically relevant application for such a "true green" technology is in lasers. Current green LED lasers are a combination of an infrared laser diode and a frequency multiplier. Light emitted by the laser diode is passed through a frequency doubler, which is a non-linear device that combines photons to create photons of higher energy within the material. For commercially available green lasers, an infrared laser diode emitting 1064 nm light is converted into 532 nm green light. While the fabrication of such a device has become inexpensive, it is still a multi-component system. The development of a green laser diode would reduce the costs associated with assembly of such a system. In addition, the reduction in size of the entire laser diode would enable the manufacturing of very small green laser-based devices. The development of green emitting materials has been limited by the ability to grow lattice-matched semiconductor material of the appropriate bandgap on commercially available substrates. A III-V semiconductor material such as InGaP can be grown with a bandgap of -2.3 eV to emit 532 nm light. However, the lattice constant of InGaP with this bandgap is in the 5.45 to 5.5 A range. A suitable substrate might be GaAs with a bandgap of roughly 5.65 A, but the lattice mismatch between these two materials is roughly 4%. The deposited InGaP layer is populated by dislocations that propagate through the material to the substrate, resulting in poor electrical conductivity and minimal light emission if any. The development of graded buffer substrates has enabled the lattice matching of semiconductor materials whose lattice constants are significantly different 2. Layers of incrementally changing percent composition are deposited on a substrate such as GaAs, enabling a gradual change in lattice constant to one that more closely matches that of the active material being deposited. This technology holds great promise for the fabrication of devices with optical emission spectra never before available with traditional substrates. Scope of Thesis: The development of "true green" LEDs and lasers using a graded buffer substrate is explored in this thesis. The aim of this project was to fabricate a green InGaP LED in order to determine the feasibility of making such a device on a graded buffer substrate. The results of such an endeavor would outline the potential to develop a green InGaP laser diode using a similar heterostructure and a ridge waveguide geometry. The first step in this process was the fabrication of an All InGaP red LED using the waveguide geometry to test the operation of a known LED heterostructure in this geometry. The second step was to fabricate an InGaP LED using the waveguide structure to characterize the optical and electrical properties of the emitting material. This project was successful in the completion of the first step, but was only partially successful in the execution of the second step. The inability to successfully complete a working InGaP LED was the result of a chemistry problem-wet etching. The results of the working All InGaP LED are presented, in addition to the problems encountered during fabrication of InGaP devices.by Josué F. Martínez.S.B