Abstract

Recording experiments on thin metallic films of lead and tantalum using a HeNe laser were first performed by Carlson, et al.¹ in 1966. The laser beam focused to a microscopic spot onto the film raised its temperature and produced a permanent change which was detectable by optical means. This forms the basis of thermal recording. Later experiments by Harnisch² used Bi monolayers and also Ge-Bi and Se-Bi bilayers for improved optical efficiency. In 1971 Maydan ^3,4 did a comprehensive study of ablative recording in Bi films. He computed the temperature rise in the film due to absorption of the laser beam and demonstrated the effect of the thermal conductivity of the film substrate (glass, mylar) on the thermal efficiency of the process. Maydan speculated that hole formation was initiated by vaporization at the center of a molten mass of Bi, after which the molten material retracted under the influence of surface tension forces. Scanning electron microscope pictures revealed the formation of a rim at the periphery of the pit containing a large percentage of the displaced material. The following 7-8 years of research concentrated on finding materials with higher writing sensitivity in which clean pits with regularly shaped rims could be formed. The research group at Hitachi⁵ experimented with thermal recording on chalcogenide thin films. Extremely clean holes were produced in As-Te-Se amorphous films. It is believed that the high viscosity of the films in the liquid state is responsible for the regular shape and cleanliness of the holes. The research group at Bell Telephone Laboratories⁶ concentrated on the Bi-Se system for a high resolution facsimile printer using a GaAs semiconductor laser. By optimizing the thicknesses of Se and Bi one increases the film absorptance but the main factor responsible for the greatly increased sensitivity (x5) over bare Bi is the large exothermic reaction which occurs when molten Se and Bi mix to form the Bi₂Se₃ composition. In addition a number of subbing layers between the mylar substrate and the Bi layer were investigated. Substrate overcoating layers not only modify the thermal impedance between the metal film and the substrate, but also the bonding energy at the film-substrate interface. An overcoat layer of poly-isobutyl methacrylate on top of a mylar substrate decreases the required threshold energy for machining a Se-Bi layer by a factor of 2 over an uncoated mylar substrate. Experimental results on laser machining also indicate that a definite temperature profile must be achieved in the unruptured film at the end of the pulse independent of its duration, before hole initiation is possible. Hole nucleation is believed to originate in thin films of bismuth at pinholes and requires only local balling up of the molten metal to reduce the surface energy. In thicker films which are pinhole free one hypothesis is that the rupture of the molten film is due to surface tension gradients. Another possibility is the formation of a blister due to evolution of gas from the heated substrate. The nucleation then proceeds by rupture of the blister. An interesting paper on the effect of overcoat layers on the ablative writing sensitivity of Te films was published by the group at IBM⁷. The hole formation was monitored directly by observing the drop in reflected signal at hole initiation. It was found that a thick overcoat of SiO₂ on a Te layer greatly increased the amount of power necessary for hole formation. This effect cannot be attributed to thermal losses alone, it must include the additional energy required to deform the rigid overcoat in order to produce a raised rim. Furthermore, the quality of the read-during-write signal is degraded when a rigid overcoat and substrate (glass) are combined. By using a thick capping layer of SiO₂ it is possible to inhibit hole formation in Te over a certain power range and produce a reversible amorphous-crystalline transition. A reversible trilayer structure using this effect was described by Bell and Spong⁸.

© 1983 Optical Society of America