all of microelectronics

(I) The intensive effort  of electronics to increase the reliability and performance  of its products while reducing their size and cost has led to the results that hardly anyone would have dared  to predict.

The evolution of electronic technology is sometimes called a revolution. True, there has been a real revolution: a quantitative change in technology has given rise to qualitative change in human capabilities . There appeared smaller and smaller electronic components performing increasingly complex electronic functions at ever higher speeds.

It all began with the development of the transistor.

Prior to  the invention of the transistor in 1947, its function in an electronic circuit could be performed only by a vacuum tube.

The first transistors had no striking advantage in size over the smallest tubes and they were more costly. The one great advantage the transistor had over the best vacuum tubes was exceedingly  low power consumption. Besides they promised greater reliability and longer life. However, it took years to demonstrate other transistor advantages.

With the invention of the transistor all essential circuit functions could be carried out inside solid  bodies. The goal  of creating electronic circuits with entirely solid-state components had finally been realized .

However, early transistors were actually enormous on the scale  at which electronic events  take place. They could respond  at a rate  of a few million times a second; this was fast enough to serve in radio and hearing-aid circuits but far below  the speed needed for high-speed computers or for microwave communication systems. Besides, the early transistors were slow.

The effort was to reduce the size of transistors so that they could operate at higher speeds. That gave rise to the whole technology of microelectronics.

A microelectronic technology has shrunk  transistors and other circuit elements to dimensions  almost invisible to unaided eye .

The point of this extraordinary miniaturization is not so much to make circuits small perse  as to make circuits capable of performing electronic functions at extremely high speeds.
 
It is known that the speed of response depends primarily on the size of transistor: the smaller the transistor, the faster it is.

The performance benefit resulting from microelectronics stems  directly from the reduction of distances between circuit components. If a circuit is to operate a few billion times a second the conductors that tie the circuit together must be measured in fractions of an inch. The microelectronics technology makes close coupling attainable.

During the past decade the performance of electronic systems increased manifold by the use of ever larger numbers of components and they continue to evolve. Modern scientific and business computers, electronic switching systems contain more than a million components.

The problem of handling many discrete electronic devices began to concern the scientists as early as 1950. The overall reliability of the electronic system is related to the number of individual components.

A more serious shortcoming was that it was once the universal practice to manufacture each of the components separately and then assemble the complete device by wiring the components together with metallic conductors. It was no good : the more components and interactions, the less reliable the system.

What ultimately provided the solution was the semiconductor integrated circuit, the concept of which had begun to take shape a few years after the invention of the transistor. Roughly  between 1960 and 1963, a new circuit technology became a reality. It was microelectronics development that solved the problem.

The advent of microelectronic circuits has not, for the most part, changed the nature of the basic functional units: microelectronic devices were still made up of transistors, resistors, capacitors, and similar components. The major difference is that all these elements and their interconnections are now fabricated on a single substrate in a single series of operations.
(II) Several key developments were required before the exciting potential of integrated circuits could be realized.

The development of microelectronics depended on the invention of techniques for making the various functional units or on a crystal of semiconductor materials. In particular, a growing number of functions have been given over to circuit elements that perform best: transistors. Several kinds of microelectronic transistors have been developed, and for each of them families of associated circuit elements and circuit patterns have evolved.

The bipolar transistor was invented in 1948 by John Bardeen, Walter H. Brattain and William Shockley of the Bell Telephone Laboratories. In bipolar transistors charge carriers of both polarities are involved in their operation. They are also known as junction transistors. The npn and pnp transistors make up the class of devices called junction transistors.

A second kind of transistor was actually conceived almost 25 years before the bipolar devices, but its fabrication in quantity did not become practical until the early 1960’s. This is field-effect transistor. The one that is common in microelectronics is the metal-oxide-semiconductor field-effect transistor. The term refers to the three materials employed in its construction and is abbreviated MOSFET (metal-oxide-semiconductor field-effect transistor).

The two basic types of transistor, bipolar and MOSFET, divide microelectronic circuits into two large families. Today the greatest density of circuit elements per- chip can be achieved with the newer MOSFET technology.

An individual integrated circuit (1C) on a chip now can embrace (ad. BKjnoiarb) more electronic elements than most complex pieces of electronic equipment that could be built in 1950.

In the first 15 years since the inception (sd. noHBJieinie) of integrated circuits, the number of transistors that could be placed on a single chip has doubled every year. The 1980 state of art is about 70K density per chip. Nowadays we can put millions of transistors on a single chip.

The first generations of the commercially produced microelectronic devices are now referred to as small-scale integrated circuits (SSI). They included a few gates. The circuitry defining a logic array had to be provided by external conductors.

Devices with more than about 10 gates on a chip but fewer than about 200 are medium-scale integrated circuits (MSI). The upper boundary of medium-scale integrated circuits technology is marked  by chips that contain a complete arithmetic and logic unit (ALU). This unit accepts as inputs two operands and can perform any one of a dozen or so operations on them. The operations include addition, sub-straction, comparison, logical “and” and “or” and shifting one bit to the left or right.

A large-scale integrated circuit (LSI) contains tens of thousands of elements, yet each element is so small that the complete circuit is typically less than a quarter of an inch on a side.

Integrated circuits are evolving from large-scale to very-large-scale (VLSI) and wafer-scale integration (WSI).

The change in scale can be measured by counting the number of transistors that can be fitted onto a chip.

Continued evolution of the microcomputer will demand further increases in packing density.

There appeared a new mode of integrated circuits, microwave integrated circuits. In broadest sense, a microwave integrated circuit is any combination of circuit functions which are packed together without a user accessible  interface.

The evolution of microwave integrated circuits must begin with the development of planar transmission lines.

As we moved into the 1970’s, stripline and microstrip assemblies became commonplace and accepted as the everyday method of building microwave integrated circuits. New forms of transmission lines were on the horizon, however. In 1974 new integrated-circuit components in a transmission line called fineline appeared. Other more exotic techniques, such as dielectric waveguide integrated circuits emerge . Major efforts currently are directed at such areas as image guide, co-planar waveguide, fineline and dielectric waveguide, all with emphasis on techniques which can be applied to monolithic integrated circuits. These monolithic circuits encompass all of the traditional microwave functions of analog circuits as well as new digital applications.

Microelectronic technique will continue to displace other modes. As the limit of optical resolution is now being reached, new lithographic and fabrication techniques will be required. Circuit patterns will have to be formed with radiation having wavelength shorter than those of light, and fabrication techniques capable of greater definition will be needed.

Electronics has extended man’s intellectual power. Microelectronics extends that power still further.

LSI technology has been the cutting edge of the innovate semiconductor industry. In the field of the process technology, much effort has been made to improve microfabrication and thinner-film formation technology. In particular, improvements in photolithographic and etching techniques are the keys to the integration of more devices on smaller chips, increases in circuit performance, and improvement in wafer process yield. Continue reading…

RTF is one of the exciting new wafer fabrication technologies. Its origin can be traced to the laser annealing  research of the early 1980’s, but it is only with the very recent appearance of techniques and equipment suitable for use in production that the technique has begun to attract serious attention of process engineers.
Current applications for RTF include ion implant annealing, glass reflow, silicide formation, and deposition of thin gate oxides. The RTF equipment market is expected to have one of the highest growth factors in the equipment industry.

In the case of evaporation, the material to be deposited is heated by a resistive heating unit until the molecules acquire the thermal energy necessary to leave the surface at a suitable speed to ensure deposition.
Sputtering differs from evaporation in that an electrical field accelerates the positive gas ions toward a cathode that is covered with a material to be deposited. An ion striking the cathode causes a molecule to be ejected and deposited on the substrate.

Most often, thin-film deposition on a ceramic substrate is done in a vacuum chamber by evaporating or sputtering conductive, resistive, or dielectric material on a carefully cleaned substrate.
The vacuum prevents oxidation and allows the molecules of material being deposited to travel to the target with minimum collisions with gas molecules.

(I) The manufacture of silicon microcircuits consists of a number of carefully controlled processes, all of which have to be performed to well-defined specifications.
Processing a “wafer” of silicon, a substrate on which the microelectronic circuits are made, is not a simple technological process.
In order to understand how transistors and other circuit elements can be made from silicon, it is necessary to consider the physical nature of semiconductor materials.
In a conductor current is known to be carried by electrons that are free to flow through the lattice of the substance.
In an insulator all the electrons are tightly bound to atoms or molecules and hence none are available to serve as a carrier of electric charge.
The situation in a semiconductor is intermediate between the two: free charge carriers are not ordinarily present, but they can be generated with a modest expenditure of energy.

Semiconductors are similar to insulators in that they have their lower bands completely filled . The semiconductor will conduct if more than a certain voltage is applied. At voltages in excess of this critical voltage, the electrons are raised from the top on the band 1 (the valence band) to the bottom of band 2 (the conducting band). Below this critical voltage, the semiconductor material acts as an insulator. Semiconductors such as that described above are called intrinsic  semiconductors — they are pure materials (for example, silicon or germanium). It should be noted that a crystal of pure silicon is a poor conductor of electricity. Thus , conductivity poses  a problem.
Several other requirements are imposed on materials. The basic demand appears to be conductivity because it can substantially improve  the resistance and delay times for VLSI. The improvement of conductivity has been made in several ways. Most semiconductor devices are known to be made by introducing controlled numbers of impurity atoms into a crystal, the process called doping.
Two independent lines of development are considered to lead to microscopic technique that produced the present integrated circuits. One involves the semiconductor technology; the other is a film technology.
Let us consider the former one first. To improve the semiconductor crystal the impurities known as dopants are added to the silicon to produce a special type of conductivity characterized by either positive (p-type) charge carriers or negative (n-type) ones. The dopants are diffused into semiconductor crystals at high temperatures. In the furnace  the crystals are surrounded by vapour containing atoms of the desired dopant. These atoms enter the crystal by substituting  for the semiconductor atoms at regular sites in the crystal lattice and move into the interior of the crystal by jumping from one site to an adjacent vacancy .
Silicon crystals may be doped with different elements. Suppose silicon is doped with boron. Each atom inserted in the silicon lattice creates a deficiency of the electron, a state that is called a hole. A hole also remains associated with an impurity atom under ordinary circumstances24 but can become mobile in response to an applied voltage. The hole is not a real particle, of course, but merely  the absence of an electron at a position where one would be found in a pure lattice of silicon atoms. Nevertheless , the hole has a positive electric charge and can carry electric current. The hole moves through the lattice in much the same way that the bubble  moves through a liquid medium. An adjacent atom transfers  an electron to the impurity atom, “filling” the hole there but creating a new one in its own cloud of electrons; the process is then repeated, so that the hole is passed  along from atom to atom.
Silicon doped with phosphorus or another pentavalent element is called an «-type semiconductor. Doping with boron or another triva-lent element gives rise to ap-type semiconductor.
Impurities may be introduced by the diffusion process. At each diffusion step  in which n-type or p-type regions are to be created in certain areas, the adjacent areas are protected by surface layer of silicon dioxide, which effectively blocks the passage of impurity atoms. This protective layer is created very simply by exposing  the silicon wafer at high temperatures to an oxidizing atmosphere. The silicon dioxide is then etched away in conformity  with a sequence of masks that accurately delineates  multiplicity  of n-type and p-type regions.
To define the microscopic regions that are exposed to diffusion in various stages  of the process, extremely precise  photographic procedures  have been developed. The surface of the silicon dioxide is
 
Coated with a photosensitive organic compound that polymerizes wherever it is struck by ultraviolet radiation and that can be dissolved and washed away everywhere else. By the use of a high-resolution photographic mask the desired configurations can thus be transferred to the coated wafer. In areas where the mask prevents the ultraviolet radiation from reaching the organic coating the coating is removed. An etching acid can then attack the silicon dioxide layer and leave the underlying silicon exposed to diffusion.
A transistor can be made by adding a third doped region to a diode so that, for example, a p-type region is said to be sandwiched between two n-type regions. One of the n-doped areas is called the emitter and the other, the collector; thep-region between them is the base.
The transistor described is called an npn transistor. There may be pnp transistors. The terms are likely to denote  the sequence of doped regions in the silicon.
The first transistor structures were formed by alloying or diffusion in bulk  single-crystal Ge or Si, but with the development of “planar technology” in the early 1960s the possibility of forming high frequency transistors and integrated circuits using epitaxial semiconductor films was realized.
The success of silicon in microelectronics is believed to be largely attributed to excellent properties of SiO interface and ease of thermal oxidation of silicon.
The recent years have seen considerable interest in the subject of oxygen and its precipitates in silicon. It has now been established  that their presence can have a variety of effects, harmful  as well as beneficial. Oxygen concentration is known to influence many silicon wafer properties, such as wafer strength, resistance to thermal warping, minority carrier lifetime, and instability in resistivity. Oxidation is widely used to create insulating areas. However, many phenomena happen not to be understood at present.
An important aspect of the oxidation process is its low cost. Several hundred wafers can be oxidized simultaneously in a single operation.
Reactive gas plasma technology is reported to be presently in widespread use in the semiconductor industry. This technology is being applied to the deposition and removal of selected materials during manufacture of semiconductor devices.
 
Contributing greatly to the manufacturing technique is a unique crystal forming method known as epitaxial growth.
Epitaxial growth in combination with oxide masking and diffusion has given the device designer extremely flexible tools  for making an almost limitless variety of structures.
After 1964, epitaxial growth remains an important technique in semiconductor device fabrication, and the demand for improved device yield per slice , still higher device operating frequencies and more sophisticated device structures, has needed continuing innovation  and development.
Advances  in silicon crystal growth technology have encouraged advances in the automation of crystal growing equipment. Crystal pulling  equipment now available uses computer software to control all the growing parameters.

(II) Let us see what a film technique is like.
Even before the invention of the transistor the electronic industry had studied the properties of thin film of metallic and insulating materials. Such films range in thickness from a fraction of a micron, or less than a wavelength of light, to several microns.
The techniques for the deposition  of thin films are numerous and include the following methods: evaporation, sputtering , anod-ization, radiation, included “cracking” or polymerization, chemical reduction, thermal reduction of oxidation and electrophoresis. The first three are the major techniques used in integrated thin film circuit construction and are also applicable to silicon integrated circuitry and device work. These methods singly or in combination enable a variety of resistive, insulating and constructive materials to be laid down onto a suitable substrate.
The two most important processes for the deposition of thin films are chemical-vapour deposition and evaporation. The film technology has proved to provide precise dimensions.
In the fabrication of a typical large-scale integrated circuit there are more thin-film steps than diffusion steps. Therefore thin-film technology is probably more critical to the overall yield  and performance of the circuit than the diffusion and oxidation steps are. A thin film happens even to be employed to select the areas on a wafer that are to be oxidized.

The deposition of layers is followed by shaping operations, such as etching, to form the required outlines . Alternatively, the film can be deposited through a mask onto the substrate to define the outlines directly.
In this way many identical thin-film devices can be made on a single sheet of material, which then are cut apart to yield individual devices.
Plasma etching, which is expected to play an important role in manufacture of semiconductor and other devices requiring fine-line lithography, involves the use of a glow discharge to generate reactive species63 from relatively inert molecular gases. These reactive species combine chemically with certain solid materials to form volatile  compounds which are then removed by vacuum pumping system.
This plasma-etching process has been shown to have important advantages in terms of cost, cleanliness, fine-line resolution, and potential for production line automation.
Additionally, the inside of a water-fabrication must be extremely clean and orderly: a single particle happens to cause a defect that will result in the malfunction of a circuit. The larger the die , the greater the chance for a defect.
The structure of an integrated circuit is sure to be complex both in the topology of its surface and in its internal composition. Each element of such a device has intricate  three-dimensional architecture that must be reproduced exactly in every circuit. The structure is made up of many layers, each of which is a detailed pattern. Some of the layers lie within the silicon wafer and others are stacked on the top. The manufacturing process consists in forming the sequence of layers precisely in accordance with the plan of the circuit designer.
Nowadays much of the procedure by which ICs are transformed from the conception of the circuit designer to a physical reality is done with the aid  of computers. In the first stage of the development of new microelectronic circuits the designers themselves used to work at specifying the functional characteristics of the device. They also selected the processing steps that will be required to manufacture it. The process was difficult and not always exact. A computer can simulate  the operations of the circuit. Besides, computer simulation is less expensive than assembling a “bread-board”  circuit made up of discrete circuit elements; it is also more accurate.
 
The layout is known to specify the pattern of each layer of the 1C. The goal of the layout is to achieve the desired function of each circuit in the smallest possible space. At present much of the preliminary work is done with the aid of computers. The final layout is also made with that of a computer.
Increasing interest in submicron layer now poses new problems. New developments in materials are believed to be due to new manufacturing forms and vice versa.
Integrated circuit technology is evolving so rapidly that even a period as short as six months can produce a significant change.

Ceramic-to-metal seals are a natural extension of the state-of-the-art where adverse temperature, shock and vibration conditions prevail. Alumina ceramics are widely used for high-performance electronic applications because of their excellent properties and moderate costs. Beryllia ceramic-to-metal seals are available but generally limited to where high heat transfer is needed. Continue reading…

Photoresists are high-sensitive materials used to generate etched patterns in substrates. The quality of the etched images depends upon the success of every step in the process, and the image flaws may be due to resist or nonresist imperfections, or to conditions which underline resist performance. Some fundamental factors influencing resist performance include adherence coating thickness, heat treatment, and resist response to various energy sources. Let us start with adherence. Continue reading…

Numerous experiments carried out at the Soviet orbital stations have paved the way to the development of methods and means of industrial production hi space.

In recent years active research has been going on in one of the fields of space industrialization — space material study and production of new materials of better quality on board the spacecraft, ranging from semiconductors for microelectronics to unique and more efficient medicines for the treatment of quite a number of diseases. Continue reading…