Search

We humans have employed and improved materials for millennia, but it required the Industrial Revolution of the last century to birth the systematic, science-based development of materials. During this time, effort expended in understanding the process-microstructure-properties relationships of materials conferred great economic and military advantage upon the successful. The introduction of machine power in this era created great leverage for improvements in the strength, ductility, corrosion resistance, formability, and similar properties of materials. Response to this opportunity led to the emergence of the materials profession. Stimulated by opportunity, materials scientists and engineers of the day met many of the challenges by first understanding and then controlling the composition and microstructure of materials. In the process, they defined the materials-engineering profession and left their names as a part of its vocabulary: Martens(ite), Bain(ite), Austen(ite), Schmid, Bessemer, Charpy, and Jomminy, to name a few. In fact the understanding and control of microstructure is the hallmark of materials science and engineering. Of course the ancient art of finding, mining, concentrating, and refining materials from the earth's crust does not apply to this definition since we wish to focus on the engineeringofmaterials.Five decades ago, a new chapter in the evolution of this profession began by the invention of the transistor. This invention and the development of integrated circuitry that followed from it spawned a new era of materials achievement, again stimulated by the enormous economic and performance gains available. In this arena however, the object of the game was to completely eliminate microstructure while doing away with impurities, save for a desired few, to levels previously unimagined. Today a material thus prepared is a blank slate upon which we can write the microstructure of an integrated circuit.

135 nsec pulses ofλ = 1.06μm light from a Nd: YAG laser have been used to form Schottky barriers by irradiation of a 500A thick metal film on n-type silicon. Large area barriers were fabricated by over-lapping individual 30μ diameter laser pulses of from 4 to 12 J/cm2. The barrier height was 0.73 ± 0.03 V, independent of the laser power. The barrier quality, as assessed by measurement of the forward current characteristic, decreased with laser power to a value of n = 1.5 at 12 J/cm2.

Profound acknowledgment of the importance of materials research to the well-being of society has recently been provided by the government of the United States of America in the form of Public Law 98-373: National Critical Materials Act of 1984. This law provides for the establishment of a National Critical Materials Council within the Executive Office of the President that is responsible for coordination of the government's materials-related policies, programs, and research and technology activities. This new legislation, the major features of which are reproduced on the next page in this issue of the BULLETIN, promises to touch all of us in some way during the coming years.The law establishes a National Critical Materials Council consisting of three presidential appointees who are qualified in materials policy or materials science and engineering. The Council is primarily responsible for the formulation of national materials policies consistent with other Federal policies. This activity includes establishing responsibilities for programs and priorities for materials activities in each Federal department or agency. The Council will also review materials programs and activities of the government for consistency with the National Materials and Minerals Policy Research and Development Act of 1980. It will monitor the critical needs of both industry and government, and it will advise the President of world trends and their implications for national and world economy and national security.

The redistribution of antimony impurity atoms implanted in (111) and (100) substrates was studied as a function of the power density of Q-switched YAG laser radiation (λ = 0.53μm) used to recrystallize 4700Å films of amorphous silicon vapor deposited over blanket and patterned Sb buried layer structures. Results for fully crystallized layers show impurity profiles ranging from a sharp interface displaced outward slightly relative to the as-grown interface, to complete redistribution of antimony into the deposited layer. Studies of patterned buried layer structures indicate that epitaxial crystallization can be achieved on (111) and (100) substrates without serious damage to the surface topography necessary for subsequent mask alignment.

The fabrication of silicon-on-insulator structures was examined utilizing epitaxial growth of Si films on porous silicon and subsequent enhanced oxidation of the porous substrate. Porous silicon films have been formed by local anodic dissolution of silicon wafers in HF, while retaining the single crystalline structure. Low temperature liquid phase epitaxy (LPE) from saturated Ga solutions as well as molecular beam epitaxy (MBE) and laser induced epitaxy have been studied. Transmission electron microscopy (TEM) has demonstrated that all three techniques can achieve crystalline Si film growth on porous silicon. Surface smoothness and crystal perfection of the resulting epitaxial layers varies considerably with each technique.

Beam processing can be applied to thin, polycrystalline silicon films on amorphous substrates to: a( increase grain size, b( produce single, isolated crystals, and c( to produce oriented single crystals. Specific methods for these tasks are outlined. Crystal growth and heat flow consideration appropriate to each are presented.

Laser processing of amorphous thin films of amorphous Ge often results in an explosive or self-sustaining crystallization reaction. The reaction is sustained by the heat liberated during crystallization. In a theoretical analysis of the process that was presented at this symposium last year, Gilmer and Leamy postulated the existence of a thin layer of liquid at the propagating interface. The liquid layer forms at temperatures above Ta, the melting point of amorphous Ge, and is predicted to be ~ 0.02 – 0.1 of the film thickness in width. We have obtained experimental confirmation of the presence of this liquid layer.

Silicon layers evaporated on crystalline Si have been crystallized by Q-switched Nd:YAG laser irradiation. A strong correlation was observed between the density of a-Si films and the quality of the epitaxial regrowth from the liquid phase. Dense films crystallized epitaxially in a wide range of laser energy densities. Layers with 20% lower density, as determined by spectroscopic ellipsometry, had higher crystallization thresholds and suffered from severe pitting of the surface. Coalescence of the excess void volume into microbubbles, stabilized by gaseous contaminants, is responsible for the surface degradation.In polycrystalline films on amorphous insulating substrates laser melting changes the grain distribution. Rapid melting and solidification of small diameter spots creates concentric rings of large crystallites. This characteristic pattern is explained by a simple model, based on the kinetics of crystallization.

The growth of device-worthy silicon crystals on insulating substrates would have a revolutionary impact upon microelectronics, and is consequently the object of intense effort. This article reviews recent progress in the application of laser and electron beams and incoherent light to this end. In every case, radiant energy is employed to melt a precursor material, from which device-worthy layers are produced during resolidification. Techniques include seeded and unseeded growth, partial and complete encapsulation of the liquid layer, and oscillation of the solid-liquid interface. Considerations of liquid zone stability, temperature gradient control, and interface morphological stability are central to each technique. Progress in this field has been rapid and several techniques have been successfully demonstrated.

Seeded epitaxial growth of Si over SiO2 is demonstrated in two types of structures. In the first case, rectangular pads of deposited Si were recessed into a thick SiO2 film. Narrow (≃ 5μm) via holes in SiO2 linked the pads with the bulk Si substrates. In the second embodiment, SiO2 patterns were recessed into the Si wafers which were then covered with a continuous 0.5μm poly-Si layer. In both cases, nearly planar geometries were maintained by use of local oxidation and etching techniques. Silicon was recrystallized with a focused Ar+ laser beam. Depending on the scanning conditions and the Si pattern geometry, single crystal growth propagated between 30 and >500μm. The factors influencing the extent and quality of the crystallized regions are discussed.

Grain boundary diffusion of arsenic and phosphorus in laser processed polycrystalline Si films has been investigated. Mesa diodes were fabricated in LPCVD Si thin films on SiO2 and subsequently recrystallized by cw Ar ion laser processing to form large grain material. The diffusion length of enhanced As and P diffusion along grain boundaries intersecting the p-n junction has been measured by the EBIC technique. Quantitative experimentation allowed determination of the grain boundary diffusion coefficients of As and P in the temperature range from 900°C to 1250°C.