Prepared by Arkady Maltsev, Nauka v Sibiri, (Science in Siberia), 2000

One of this country's leading experts on micro- and macroelectronics - Academician Kamil Valiev, was once asked by correspondents what he thinks about the future of this branch of modem technology. "What, in your view"-he was asked-"will succeed silicon? Will it be gallium arsenide, fullerenes, or diamonds?" "I think,"-replied the scientist-"that silicon will always be with us, at least till the time when the advent of some completely new materials turns super-large silicon integrated circuits into 'stone age' products." This was the academician's way of saying that in areas like micro and power electronics, solar energy, micromechanics and other related branches silicon has been and remains the basic material of choice. With reference to this country in particular, it is Siberia that possesses some truly unique potential for a rapid expansion of the production of semiconductor silicon.

The region boasts a rich raw material base-deposits of high-purity quartzites, and large-scale production of what we call technical (metallurgical) silicon has been established at the Irkutsk and Bratsk aluminum smelters. Geared to the project are several other big plants, scheduled for conversion, including one of the main participants in the national nuclear power program - the Mining and Chemical Works in Zheleznogorsk (Krasnoyarsk Territory). In line with this general trend, the Siberian Branch of the Russian Academy of Sciences has been focusing its efforts on what is known as an integrated project entitled "Fundamental problems of studies of semiconductor silicon". Focusing on this project from 1997 to 1999 were the Novosibirsk Institutes of Physics of Semiconductors, Inorganic Chemistry, the Kutateladze

page 24

Institute of Heat Physics, the Institute of Theoretical and Applied Mechanics, the Institute of Hydrodynamics named after M. Lavrentyev and the Irkutsk Institutes of Geochemistry and Chemistry.

The project was divided into four main sections, with the first of them called "Fundamental problems of growth of high-quality monocrystals of silicon of large diameter by the Chokhralsky and crucible-free zonal smelting at extreme parameters of purity, homogeneity and structural perfection". Within this framework experts studied and developed mathematical models of the connection between thermal, hydrodynamic and structural- physical characteristics of melt with the electronic and structural properties of silicon monocrystals of large diameter. The latter were obtained by means of crucible-free zonal smelting conducted at the Institute of Semiconductor Physics (in conjunction with the Danish firm Haldor Topse) and with the financial backing of Russia's Ministry of Science and Technologies. And the Chokhralsky method (of silicon monocrystal growth by means of extrusion from melt in a quartz crucible) was implemented at the Zhe-leznogorsk Mining and Chemical Plants within the framework of a specially established Kremniy (silicon) R&D Complex. Considerable support for this work was provided by the Ministry of Economics and Atomic Power Engineering of the Russian Federation.

The second part of the project was called "Development of physical principles of new types of multilayer structures of silicon-on-insulator, heteroepitaxial and nanodis-persed structures". These studies were conducted by the institutes of semiconductors and heat physics which possess the necessary equipment and use some unique vacuum fine-film technologies.

The third part of the project was entitled "Investigation of scientific problems involved in the production of high-quality semiconductor silicon and initial products (chlorsilanes, poly- and granulated silicon, 'quartz crucibles')." The studies were conducted by the Institutes of inorganic chemistry, theoretical and applied mechanics and the Irkutsk Mining Institute. On the basis of these studies production of polysilicon was launched at the Zheleznogorsk Mining and Chemical Plant.

And the last section of the project, entitled "Development of silicon production technologies for solar power generation" was the responsibility of the Institute of Geochemistry with the participation of ZAO Kremniy (Irkutsk Aluminum Smelter in the town of Shelekhov).

page 25

Discussions on the main results of the work on the Project were on the agenda of the Second Conference on Material Studies and Physico-chemical Basics of Technologies of Production of Alloyed Silicon Crystals. The conference, which met in Moscow early this year, was attended by scientists from Moscow and the Moscow Region, St. Petersburg, Nizhni Novgorod, Novosibirsk, Krasnoyarsk, Irkutsk, Kiev and Minsk-a total of more than 120 participants. Speakers at the conference stressed that so far only four countries-the United States, Japan, Germany and Russia-have mastered the technologies for polysilicon production although the demand for it is steadily growing. This stems from the needs of such branches as microelectronics and solar power generation which will provide a tangible contribution to electricity generation in the 21st century.

The development of technologies of manufacture of silicon monocrystals is oriented at producing ingots of larger size with mounting demands for getting more perfect crystalline structures and greater uniformity in the distribution of electrophysical properties throughout the volume of the material. The most serious problem encountered in this connection is the need to reduce the size of microflaws which have the strongest impact on the performance characteristics of integrated circuits. Speakers at the conference discussed the results of studies (Institute of Semiconductor Physics) of new types of what are called extension defects in silicon crystals and also the method of modelling of processes of heat and mass transfer, crystallization and defect formation (Institute of Heat Physics) which help reduce the numbers of surface submicronic defects.

Polished silicon plates have been mainly used so far for the production of integrated circuits. But with the current transition to submicronic and nanometer levels preference is being given to what are called epitaxial(*) structures especially in view of the prospects of using them for super-fast circuits of the future. Today epitaxial processes (chiefly molecular-ray ones), combined with ionic implantation(*) and pulsed radiation processing of materials are becoming increasingly important for the formation of silicon structures.

There has been growing interest among experts in recent time towards what are called microcrystalline and amorphous silicon films upon glass and metal base. These can be used as solar panels, fine-film transistors for liquid-crystal displays, light emitters and photocells. What is more, methods have been developed of producing such films with preset characteristics. One of the most effective of these consists in using a supersonic gas jet with gas activation by an electron beam. And the rate of precipitation of silicon layers by this technique has turned out to be the greatest.

The attention of experts has also been attracted by yet one more modification of this remarkable material-porous silicon. When some of the associated problems (ensuring stability and reproducibility) are resolved the new material will have a future as light emitter in the visible band. So far, however, there have been even more successful studies into what experts call controlled formation of pores massif of preset configuration in the process of deep photoanodic etching, or scouring, of silicon. This kind of structures can be used for the making of matrixes of parabolic short-focus X-ray lenses and of components of three- dimensional photon crystals.

Another direction of studies of this porous material is linked with the production of unique bases for homo- and heteroepitaxy with the subsequent development on this basis of semiconductors on dielectrics with the help of which it should be possible to reduce appreciably the parasitic electric effects, ensure reliable insulation of the base layer and achieve a reduction of working voltage and power levels.

The smaller are the topological dimensions of the elements in electronic circuits, the greater is their density and the more complex is the architecture of traditional wiring circuits. The latter fact is a tangible obstacle to increasing the rates of response of various instruments. An attractive alternative, therefore, is offered by fiber optics systems which can take care in principle of generation, modulation, amplification, transmission and detection of light signals. But the problem of fiber optics runs into a "snag" of developing an effective radiation source, because pure silicon cannot be used for that purpose due to a number of reasons. As proved by experience-the problem can be resolved by the introduction of erbium which forms effective centers of emission recombination. In that case the generated emission of 1.54 Mkm is practically not absorbed by silicon and matches the maximum transparency "window" of optical waveguides of quartz glass. The main snag, however, is the low solubility of erbium in silicon. As proved by Siberian researchers, however, this obstacle can be overcome by using what are called non-equilibrium methods of generation of strongly alloyed layers resorting to ionic implantation, molecular-ray epitaxy and ionic-ray spray- coating.

Summing it up, in the process of implementation of the above integration project it has been possible to obtain within relatively short time high-quality silicon monocrystals and a range of products on this basis. This opens up new prospects for applied research in the centers of the Siberian Branch of the Russian Academy of Sciences and at the local industrial plants. The aim of it all is obtaining high-tech equipment on the basis of semiconductor silicon.

* Epitaxy-oriented growth of a monocrystal on the surface of another.- Ed.

* Ionic implantation-introduction of foreign atoms into a solid body by means of ionic bombardment. -Ed.

Permanent link to this publication: