Erasable Phase-Change Optical Materials
- PDF / 756,867 Bytes
- 3 Pages / 576 x 777.6 pts Page_size
- 89 Downloads / 264 Views
conditions. On the other hand, when the information is rewritten, the material must swiftly crystallize—for example within 100 ns (a ratio of 1015) when heated. Does such a material exist? In this article, first I will review the history of phase-change materials research. Next the principles that have led us to the present material composition will be discussed. Last the Ge-Sb-Te system will be introduced as an example of a material solving the very difficult previously mentioned problem.
History of Phase-Change Materials Research What is the best material for an erasable phase-change optical memory? The answer to this question has greatly changed over the past 10 years. Table I shows the history of phasechange materials research for optical memory. Research on laser-induced reversible phase transitions started in the early 1970s.1'2 From then to the first half of the 1980s, eutectic compositions of Te-
based alloys were mainly studied. This is because (1) Te-based alloys near their eutectic point had already been known to be easily vitrified through a melt-quench process and (2) a near-infrared laser diode was available at that time and Tebased alloys were known to show appropriate optical changes in that wavelength region. Since a liquidus line drops near a eutectic point in binary or ternary alloy systems, the viscosity of the liquid becomes very large at the eutectic composition. In the case of Te-based alloys, the atomic mobility in the liquid state tends to be very small because a long-chain structure -Te-Te-Te-, with small mobility, remains even in the liquid state. Therefore a Te-based alloy having a eutectic composition has a tendency to be solidified while keeping the atomic distribution of the liquid state, in addition to being easily vitrified. However this advantage of "easiness for vitrification" unfortunately leads to the disadvantage of "difficulty for crystallization." The words "difficulty for crystallization" in this case do not mean that the vitrified state is thermally stable but that the crystallization speed is very low. Thus many studies during this time period were done to increase the crystallization speed and thereby to shorten the data-rewriting time. The crystallization time was as high as several microseconds for the early compositions. In spite of many efforts, it remains at several hundred nanoseconds even for the most optimized compositions such as the TeGe-Sn-Au system.3 To further shorten the crystallization time, studies of stoichiometric compound materials have been proceeding since the second half of the 1980s.45 There are many stoichiometric compositions even if the material candidates are limited to chalcogenide alloys. How does one find the most suitable compositions among them for an erasable phasechange optical memory?
Compound Phase-Change Materials
Figure 1. Transmission-electron microscope image of a Ge-Sb-Te alloy film. Amorphous recording marks are formed in crystallized area along the tracks. The mark size is about 0.5 fim.
Several important conditions that pha
Data Loading...
