Bandgap Anomaly, Atomic Ordering, and Their Applications
- PDF / 1,683,332 Bytes
- 5 Pages / 576 x 777.6 pts Page_size
- 19 Downloads / 251 Views
own on GaAs(llO); the type was CuAu I.3 CuPt-type ordering was observed in SiGe grown on Si(OOl).4 Since the (110) orientation of the substrate for AlGaAs was clumsy for practical purposes and the degree of ordering in SiGe was extremely low, these phenomena seemed to have little relevance in practical applications. Other observations 5 were made such as famatinite-type ordering in GalnAs and chalcopyrite-type ordering in GaAsSb. These observations however were quite rare, with only isolated cases reported. CuPt ordering in III-V alloys was first observed in 1986 for GaAsSb6 during the study of phase separation in the material. We encountered the CuPt ordering from a different direction. In 1982 we first accomplished red-laser operation by optical pumping at 90 K in a metalorganicvapor-phase-epitaxy (MOVPE)-grown MRS BULLETIN/JULY 1997
next-generation high-density opticaldisk systems (digital versatile disk [DVD] and DVD-read-only memory). The ordering and bandgap anomaly have unique and indispensable applications in the design and fabrication technologies of these laser diodes. This article describes how the ordering in GalnP, which is now regarded as the most typical CuPt ordering, was noticed and has been studied, and how this study has been useful in laser-diode technologies. Bandgap Anomaly and CuPt Ordering in Gao.5lno.5P Even a very small effect sometimes requires profound changes in our ideas. —R.P. Feynman
In 111*111? AV or IUVfVl3 v-type terAlInP/GalnP/AlInP double heterostruc- nary semiconductor alloys, it had been ture.7 The wavelength was 647 nm, which firmly believed that if the alloy composiwas emitted from the Gan.5Ino.5P laser ac- tion (x) were determined, the bandgap tive layer. We then noticed that the nor- would be uniquely determined. However mal wavelength for Gao.5Inn.5P should our Gao.5Ino.5P grown on GaAs(OOl) subhave been around 630 nm. While Jo- strates by MOVPE consistently showed hansson and Linde discovered CuPt or- about a 50-meV lower room-temperature determined by pho tolu minesdering in the study on the anomaly in bandgap, 9 resistivity, very strong CuPt-type order- cence, than the expected bandgap of ing in Gao.5Ino.5P was discovered in 1987 1.9 eV. This anomaly was also manifested during research on the origin of this in the lasing wavelength anomaly menbandgap anomaly8 A relation between tioned previously. The difference of the bandgap anomaly and the degree of 50 meV is 2.6% of 1.9 eV. This amount is ordering was established, showing a rele- in a sense "small" and could have been vance to the basic optoelectronic proper- neglected, considering the scattering of ties. Importance of surface structure and the reported bandgap values. surface steps was pointed out at that However if this is truly an anomaly in time—that is, the formation mechanism bandgap, then the concept of the uniquecould not be explained only by homoge- ness of bandgap energy based on componeous bulk thermodynamics but was re- sition needs to be reexa mined. A possible lated to surface effects.8 interpretation
Data Loading...
