History of the creation of an integrated board. Series of microcircuits. Functional control of ICs and test circuits
Introduction
Since the advent of the first computers, software developers have dreamed of hardware designed to solve exactly their problem. Therefore, the idea of creating special integrated circuits that can be tailored to effectively perform a specific task has appeared for quite some time. There are two development paths here:
- The use of so-called specialized custom integrated circuits (ASIC - Application Specific Integrated Circuit). As the name suggests, such chips are made by manufacturers hardware custom-made to effectively perform a specific task or range of tasks. They do not have the versatility of conventional microcircuits, but they solve the tasks assigned to them many times faster, sometimes by orders of magnitude.
- Creation of microcircuits with reconfigurable architecture. The idea is that such chips arrive to the developer or software user in an unprogrammed state, and he can implement on them the architecture that best suits him. Let's take a closer look at their formation process.
Over time, a large number of different chips with reconfigurable architecture appeared (Fig. 1).
Fig. 1 Variety of chips with reconfigurable architecture
For quite a long time, only PLD (Programmable Logic Device) devices existed on the market. This class includes devices that implement the functions necessary to solve the assigned problems in the form of a perfect disjunctive normal shape(perfect DNF). The first to appear in 1970 were EEPROM chips, which belong specifically to the class of PLD devices. Each circuit had a fixed array of AND logic functions connected to a programmable set of OR logic functions. For example, consider a PROM with 3 inputs (a, b and c) and 3 outputs (w, x and y) (Fig. 2).
Rice. 2. PROM chip
Using a predefined AND array, all possible conjunctions over input variables are implemented, which can then be arbitrarily combined using OR elements. Thus, at the output you can implement any function of three variables in the form of a perfect DNF. For example, if you program those OR elements that are circled in red in Figure 2, then the outputs will produce the functions w=a x=(a&b) ; y=(a&b)^c.
Initially, PROM chips were intended to store program instructions and constant values, i.e. to perform computer memory functions. However, developers also use them to implement simple logic functions. In fact, the chip's PROM can be used to implement any logical block, provided that it has a small number of inputs. This condition follows from the fact that in EEPROM microcircuits the matrix of AND elements is strictly defined - all possible conjunctions from the inputs are implemented in it, that is, the number of AND elements is equal to 2 * 2 n, where n is the number of inputs. It is clear that as the number n increases, the size of the array grows very quickly.
Next, in 1975, the so-called programmable logic arrays (PLMs) appeared. They are a continuation of the idea of PROMs of microcircuits - PLMs also consist of AND and OR arrays, however, unlike PROMs, both arrays are programmable. This provides greater flexibility for such chips, but they have never been common because signals take much longer to travel through programmable connections than through their predefined counterparts.
In order to solve the speed problem inherent in PLMs, a further class of devices called programmable array logic (PAL) appeared in the late 1970s. A further development of the idea of PAL chips was the emergence of GAL (Generic Array Logic) devices - more complex varieties of PAL using CMOS transistors. The idea used here is exactly the opposite of the idea of PROM chips - a programmable array of AND elements is connected to a predefined array of OR elements (Fig. 3).
Rice. 3. Unprogrammed PAL device
This imposes a limitation on functionality, however, such devices require arrays of a much smaller size than in EPROM chips.
A logical continuation of simple PLDs was the emergence of so-called complex PLDs, consisting of several blocks of simple PLDs (usually PAL devices are used as simple PLDs), united by a programmable switching matrix. In addition to the PLD blocks themselves, it was also possible to program the connections between them using this switch matrix. The first complex PLDs appeared in the late 70s and early 80s of the 20th century, but the main development of this area occurred in 1984, when Altera introduced a complex PLD based on a combination of CMOS and EPROM technologies.
The advent of FPGA
In the early 1980s, in the digital ASIC environment, a gap opened up between the main types of devices. On the one hand, there were PLDs, which can be programmed for each specific task and are quite easy to manufacture, but they cannot be used to implement complex functions. On the other hand, there are ASICs that can implement extremely complex functions, but have a rigidly fixed architecture and are time-consuming and expensive to manufacture. An intermediate link was needed, and FPGA (Field Programmable Gate Arrays) devices became such a link.
FPGAs, like PLDs, are programmable devices. The main fundamental difference between FPGA and PLD is that functions in FPGA are implemented not using DNF, but using programmable lookup tables (LUTs). In these tables, the function values are specified using a truth table, from which the required result is selected using a multiplexer (Fig. 4):
Rice. 4. Correspondence table
Each FPGA device consists of programmable logic blocks (Configurable Logic Blocks - CLBs), which are interconnected by connections that are also programmable. Each such block is intended for programming a certain function or part of it, but can be used for other purposes, for example, as memory.
In the first FPGA devices, developed in the mid-80s, the logic block was very simple and contained one 3-input LUT, one flip-flop and a small number of auxiliary elements. Modern FPGA devices are much more complex: each CLB block consists of 1-4 “slices”, each of which contains several LUT tables (usually 6-input), several triggers and a large number of service elements. Here is an example of a modern "slice":
Rice. 5. The device of a modern "cut"
Conclusion
Since PLD devices cannot implement complex functions, they continue to be used to implement simple functions in portable devices and communications, while FPGA devices ranging from 1000 gate sizes (the first FPGA developed in 1985) this moment exceeded the 10 million gate mark (Virtex-6 family). They are actively developing and are already replacing ASIC chips, allowing the implementation of a variety of extremely complex functions without losing the ability to reprogram.
Now, even more or less advanced Cell phones cannot do without a microprocessor, let alone tablets, laptops and desktops personal computers. What is a microprocessor and how did the history of its creation develop? To put it in plain language, a microprocessor is a more complex and multifunctional integrated circuit.
The history of the microcircuit (integrated circuit) begins since 1958, when an employee of the American company Texas Instruments, Jack Kilby, invented a certain semiconductor device containing several transistors in one case, connected by conductors. The first microcircuit - the ancestor of the microprocessor - contained only 6 transistors and was a thin plate of germanium with tracks made of gold applied to it. All this was located on a glass substrate. For comparison, today there are units and even tens of millions of semiconductor elements.
By 1970 quite a lot of manufacturers were engaged in the development and creation of integrated circuits of various capacities and different functional areas. But this year can be considered the date of birth of the first microprocessor. It was this year that Intel created a memory chip with a capacity of only 1 Kbit - negligible for modern processors, but incredibly large for that time. At that time, this was a huge achievement - the memory chip was capable of storing up to 128 bytes of information - much higher than similar analogues. In addition, around the same time, the Japanese manufacturer of calculators Busicom ordered the same Intel 12 microcircuits of various functional areas. Intel specialists managed to implement all 12 functional areas in one chip. Moreover, the created microcircuit turned out to be multifunctional, since it made it possible to programmatically change its functions without changing the physical structure. The microcircuit performed certain functions depending on the commands sent to its control pins.
Within a year in 1971 Intel releases the first 4-bit microprocessor, codenamed 4004. Compared to the first microcircuit with 6 transistors, it contained as many as 2.3 thousand semiconductor elements and performed 60 thousand operations per second. At that time, this was a huge breakthrough in the field of microelectronics. 4-bit meant that the 4004 could process 4-bit data at once. In two more years in 1973 The company produces an 8-bit processor 8008, which already worked with 8-bit data. Beginning since 1976, the company begins to develop a 16-bit version of the 8086 microprocessor. It was he who began to be used in the first IBM personal computers and, in fact, laid one of the building blocks in
Analog and digital microcircuits are produced in series. A series is a group of microcircuits that have a single design and technological design and are intended for joint use. Microcircuits of the same series, as a rule, have the same power supply voltages and are matched in terms of input and output resistances and signal levels.
Housings
Microcircuits are available in two design options - packaged and uncased.
The microcircuit housing is a supporting system and part of the structure designed to protect against external influences and for electrical connection with external circuits through pins. The cases are standardized to simplify the manufacturing technology of finished products.
A packaged chip is a semiconductor crystal intended for installation in a hybrid chip or microassembly (direct mounting on a printed circuit board is possible).
Specific names
Intel was the first to produce a chip that performed the functions of a microprocessor (English microproccessor) - Intel 4004. Based on the improved microprocessors 8088 and 8086, IBM released its famous personal computers)
The microprocessor forms the core of the computer; additional functions, such as communication with peripherals, were performed using specially designed chipsets (chipset). For the first computers, the number of microcircuits in sets was in the tens and hundreds, in modern systems This is a set of one, two or three microcircuits. Recently, there has been a tendency to gradually transfer chipset functions (memory controller, PSI Express bus controller) to the processor.
Microprocessors with built-in RAM and ROM, memory and I/O controllers, and other additional functions are called microcontrollers.
Legal protection
Russian legislation provides legal protection to integrated circuit topologies. Topology integrated circuit is the spatial-geometric arrangement of the set of elements of an integrated circuit and the connections between them recorded on a material medium (Article 1448 of the Civil Code of the Russian Federation).
The exclusive right to the topology is valid for ten years. During this period, the copyright holder may, at his own discretion, register the topology with the Federal Service for Intellectual Property, Patents and Trademarks.
History of creation
On May 7, 1952, British radio engineer Geoffrey Dummer first proposed the idea of integrating multiple standard electronic components into a monolithic semiconductor chip, and a year later Harwick Johnson filed the first-ever patent application for a prototype integrated circuit (IC). The implementation of these proposals in those years could not take place due to insufficient development of technology.
At the end of 1958 and in the first half of 1959, a breakthrough took place in the semiconductor industry. Three men, representing three private American corporations, solved three fundamental problems that were preventing the creation of integrated circuits. Jack Kilby of Texas Instruments patented the principle of integration, created the first, imperfect, prototypes of the IC and brought them to mass production. Kurt Legowec of Sprague Electric Company invented a method for electrically insulating components formed on a single semiconductor chip (p-n junction insulation). Robert Noyce of Fairchild Semiconductor invented a method for electrically interconnecting IC components (aluminum metallization) and proposed an improved version of component insulation based on the latest planar technology of Jean Erny. On September 27, 1960, Jay Last's group created the first workable semiconductor IP based on the ideas of Noyce and Ernie. Texas Instruments, which owned the patent for Kilby's invention, launched a patent war against its competitors, which ended in 1966 with a settlement agreement on cross-licensing technologies.
Early logic ICs of the mentioned series were literally built from standard components, the sizes and configurations of which were specified by the technological process. Circuit designers who designed logic ICs of a particular family operated with the same standard diodes and transistors. In 1961-1962, the leading Sylvania developer Tom Longo broke the design paradigm by using it for the first time in a single IC. various configurations of transistors depending on their functions in the circuit. At the end of 1962, Sylvania released the first family of transistor-transistor logic (TTL) developed by Longo - historically the first type of integrated logic that managed to gain a long-term foothold in the market. In analog circuitry, a breakthrough of this level was made in 1964-1965 by Fairchild operational amplifier designer Bob Widlar.
The first semiconductor integrated circuit in the USSR was created on the basis of planar technology developed in early 1960 at NII-35 (then renamed Pulsar Research Institute) by a team that was later transferred to NIIME (Mikron). The creation of the first domestic silicon integrated circuit was concentrated on the development and production with military acceptance of the TC-100 series of integrated silicon circuits (37 elements - the equivalent of the circuit complexity of a flip-flop, an analogue of the American SN-51 series ICs from Texas Instruments). Prototype samples and production samples of silicon integrated circuits for reproduction were obtained from the USA. The work was carried out at NII-35 (director Trutko) and the Fryazino Semiconductor Plant (director Kolmogorov) for a defense order for use in an autonomous altimeter for a ballistic missile guidance system. The development included six standard integrated silicon planar circuits of the TS-100 series and, with the organization of pilot production, took three years at NII-35 (from 1962 to 1965). It took another two years to develop factory production with military acceptance in Fryazino (1967)
First integrated circuits
Dedicated to the 50th anniversary of the official date
B. Malashevich
On September 12, 1958, Texas Instruments (TI) employee Jack Kilby demonstrated to management three strange devices - devices made of two pieces of silicon measuring 11.1 x 1.6 mm glued together with beeswax on a glass substrate (Fig. 1). These were three-dimensional mock-ups - prototypes of an integrated circuit (IC) of the generator, proving the possibility of manufacturing all circuit elements based on one semiconductor material. This date is celebrated in the history of electronics as the birthday of integrated circuits. But is it?
Rice. 1. Layout of the first IP by J. Kilby. Photo from the site http://www.computerhistory.org/semiconductor/timeline/1958-Miniaturized.html
By the end of the 1950s, the technology of assembling electronic equipment (REA) from discrete elements had exhausted its capabilities. The world had come to an acute crisis of REA; radical measures were required. By this time, integrated technologies for the production of both semiconductor devices and thick-film and thin-film ceramic circuit boards had already been industrially mastered in the USA and the USSR, i.e. the prerequisites were ripe for overcoming this crisis by creating multi-element standard products - integrated circuits.
Integrated circuits (chips, ICs) include electronic devices of varying complexity, in which all similar elements are manufactured simultaneously in a single technological cycle, i.e. using integrated technology. Unlike printed circuit boards (in which all connecting conductors are simultaneously manufactured in a single cycle using integrated technology), resistors, capacitors, and (in semiconductor ICs) diodes and transistors are similarly formed in ICs. In addition, many ICs are manufactured simultaneously, from tens to thousands.
ICs are developed and produced by industry in the form of series, combining a number of microcircuits for various functional purposes, intended for joint use in electronic equipment. The series ICs have a standard design and a unified system of electrical and other characteristics. ICs are supplied by the manufacturer to various consumers as independent commercial products that meet a certain system of standardized requirements. ICs are non-repairable products; when repairing electronic equipment, failed ICs are replaced.
There are two main groups of ICs: hybrid and semiconductor.
In hybrid ICs (HICs), all conductors and passive elements are formed on the surface of a microcircuit substrate (usually ceramic) using integrated technology. Active elements in the form of packageless diodes, transistors and semiconductor IC crystals are installed on the substrate individually, manually or automatically.
In semiconductor ICs, connecting, passive and active elements are formed in a single technological cycle on the surface of a semiconductor material (usually silicon) with partial invasion of its volume using diffusion methods. At the same time, on one semiconductor wafer, depending on the complexity of the device and the size of its crystal and wafer, from several tens to several thousand ICs are manufactured. The industry produces semiconductor ICs in standard packages, in the form of individual chips or in the form of undivided wafers.
The introduction of hybrid (GIS) and semiconductor ICs to the world occurred in different ways. GIS is a product of the evolutionary development of micromodules and ceramic board mounting technology. Therefore, they appeared unnoticed; there is no generally accepted date of birth of GIS and no generally recognized author. Semiconductor ICs were a natural and inevitable result of the development of semiconductor technology, but they required the generation of new ideas and the creation of new technology, which have their own dates of birth and their own authors. The first hybrid and semiconductor ICs appeared in the USSR and the USA almost simultaneously and independently of each other.
The first hybrid ICs
Hybrid ICs include ICs, the production of which combines the integral technology of manufacturing passive elements with individual (manual or automated) technology for installing and assembling active elements.
Back in the late 1940s, the Centralab company in the USA developed the basic principles for the manufacture of thick-film ceramic-based printed circuit boards, which were then developed by other companies. The basis was the manufacturing technology of printed circuit boards and ceramic capacitors. From printed circuit boards we took an integrated technology for forming the topology of connecting conductors - silk-screen printing. From capacitors - the substrate material (ceramics, often sital), as well as the materials of the pastes and the thermal technology of their fixation on the substrate.
And in the early 1950s, the RCA company invented thin-film technology: by spraying various materials in a vacuum and depositing them through a mask onto special substrates, they learned how to simultaneously produce many miniature film connecting conductors, resistors and capacitors on a single ceramic substrate.
Compared to thick-film technology, thin-film technology provided the possibility of more precise manufacturing of smaller-sized topology elements, but required more complex and expensive equipment. Devices manufactured on ceramic circuit boards using thick-film or thin-film technology are called “hybrid circuits.” Hybrid circuits were produced as components of products of their own production; each manufacturer had their own design, dimensions, and functional purposes; they did not enter the free market, and therefore are little known.
Hybrid circuits have also invaded micromodules. At first, they used discrete passive and active miniature elements, united by traditional printed wiring. The assembly technology was complex, with a huge share of manual labor. Therefore, micromodules were very expensive, and their use was limited to on-board equipment. Then thick film miniature ceramic scarves were used. Next, resistors began to be manufactured using thick-film technology. But the diodes and transistors used were still discrete, individually packaged.
The micromodule became a hybrid integrated circuit at the moment when unpackaged transistors and diodes were used in it and the structure was sealed in a common housing. This made it possible to significantly automate the process of their assembly, sharply reduce prices and expand the scope of application. Based on the method of forming passive elements, thick-film and thin-film GIS are distinguished.
The first GIS in the USSR
The first GIS (modules of the “Kvant” type, later designated IS series 116) in the USSR were developed in 1963 at NIIRE (later NPO Leninets, Leningrad) and in the same year its pilot plant began their serial production. In these GIS, semiconductor ICs “R12-2”, developed in 1962 by the Riga Semiconductor Devices Plant, were used as active elements. Due to the inextricability of the histories of the creation of these ICs and their characteristics, we will consider them together in the section devoted to P12-2.
Undoubtedly, the Kvant modules were the first in the world of GIS with two-level integration - they used semiconductor ICs rather than discrete packaged transistors as active elements. It is likely that they were also the first in the world of GIS - structurally and functionally complete multi-element products, supplied to the consumer as an independent commercial product. The earliest foreign similar products identified by the author are the IBM Corporation SLT modules described below, but they were announced the following year, 1964.
The first GIS in the USA
The appearance of thick-film GIS as the main element base of the new IBM System /360 computer was first announced by IBM in 1964. It seems that this was the first use of GIS outside the USSR; the author was unable to find earlier examples.
Already known at that time in specialist circles, the semiconductor IC series “Micrologic” from Fairchild and “SN-51” from TI (we will talk about them below) were still inaccessibly rare and prohibitively expensive for commercial applications, such as the construction of a large computer. Therefore, the IBM corporation, taking the design of a flat micromodule as a basis, developed its series of thick-film GIS, announced under the general name (as opposed to “micromodules”) - “SLT modules” (Solid Logic Technology - solid logic technology. Usually the word “solid” translated into Russian as “solid", which is absolutely illogical. Indeed, the term “SLT modules” was introduced by IBM as opposed to the term “micromodule" and should reflect their difference. But both modules are “solid”, i.e. this translation is not The word “solid” has other meanings – “solid”, “whole”, which successfully emphasize the difference between “SLT-modules” and “micromodules” - SLT-modules are indivisible, non-repairable, i.e. “whole.” We did not use the generally accepted translation into Russian: Solid Logic Technology - technology of solid logic).
The SLT module was a half-inch square ceramic thick-film microplate with pressed-in vertical pins. Connecting conductors and resistors were applied to its surface using silk-screen printing (according to the diagram of the device being implemented), and unpackaged transistors were installed. Capacitors, if necessary, were installed next to the SLT module on the device board. While externally almost identical (micromodules are slightly taller, Fig. 2.), SLT modules differed from flat micromodules in their higher density of elements, low power consumption, high performance and high reliability. In addition, SLT technology was quite easy to automate, therefore they could be produced in huge quantities at a cost low enough for use in commercial equipment. This is exactly what IBM needed. The company built an automated plant in East Fishkill near New York for the production of SLT modules, which produced them in millions of copies.
Rice. 2. USSR micromodule and SLT module f. IBM. Photo STL from the site http://infolab.stanford.edu/pub/voy/museum/pictures/display/3-1.htm
Following IBM, other companies began to produce GIS, for which GIS became a commercial product. The standard design of flat micromodules and SLT modules from IBM has become one of the standards for hybrid ICs.
The first semiconductor ICs
By the end of the 1950s, the industry had every opportunity to produce cheap elements of electronic equipment. But if transistors or diodes were made of germanium and silicon, then resistors and capacitors were made of other materials. Many then believed that when creating hybrid circuits there would be no problems in assembling these elements, manufactured separately. And if it is possible to produce all the elements of a standard size and shape and thereby automate the assembly process, then the cost of the equipment will be significantly reduced. Based on such reasoning, supporters of hybrid technology considered it as the general direction for the development of microelectronics.
But not everyone shared this opinion. The fact is that mesa transistors, and especially planar transistors, already created by that period, were adapted for group processing, in which a number of operations for the manufacture of many transistors on one substrate plate were carried out simultaneously. That is, many transistors were manufactured on one semiconductor wafer at once. Then the plate was cut into individual transistors, which were placed in individual cases. And then the equipment manufacturer combined the transistors on one printed circuit board. There were people who thought this approach was ridiculous - why separate the transistors and then connect them again. Is it possible to combine them immediately on a semiconductor wafer? At the same time, get rid of several complex and expensive operations! These people came up with semiconductor ICs.
The idea is extremely simple and completely obvious. But, as often happens, only after someone first announced it and proved it. He proved that simply announcing it is often, as in this case, not enough. The idea of an IC was announced back in 1952, before the advent of group methods for manufacturing semiconductor devices. On annual conference on electronic components, held in Washington, an employee of the British Royal Radar Office in Malvern, Jeffrey Dummer, presented a report on the reliability of radar equipment components. In the report he made a prophetic statement: “ With the advent of the transistor and work in the field of semiconductor technology, it is generally possible to imagine electronic equipment in the form of a solid block containing no connecting wires. The block may consist of layers of insulating, conducting, rectifying and reinforcing materials in which certain areas are cut out so that they can directly perform electrical functions.”. But this forecast went unnoticed by experts. They remembered it only after the appearance of the first semiconductor ICs, that is, after the practical proof of a long-publicized idea. Someone had to be the first to reinvent and implement the semiconductor IC idea.
As in the case of the transistor, the generally recognized creators of semiconductor ICs had more or less successful predecessors. Dammer himself made an attempt to realize his idea in 1956, but failed. In 1953, Harvick Johnson of RCA received a patent for a single-chip oscillator, and in 1958, together with Torkel Wallmark, announced the concept of a “semiconductor integrated device.” In 1956, Bell Labs employee Ross produced a binary counter circuit based on n-p-n-p basis structures in a single single crystal. In 1957, Yasuro Taru from the Japanese company MITI received a patent for combining various transistors in one crystal. But all these and other similar developments were of a private nature, were not brought to production and did not become the basis for the development of integrated electronics. Only three projects contributed to the development of IP in industrial production.
The lucky ones were the already mentioned Jack Kilby from Texas Instruments (TI), Robert Noyce from Fairchild (both from the USA) and Yuri Valentinovich Osokin from the design bureau of the Riga Semiconductor Device Plant (USSR). The Americans created experimental samples of integrated circuits: J. Kilby - a prototype of an IC generator (1958), and then a trigger on mesa transistors (1961), R. Noyce - a trigger using planar technology (1961), and Yu. Osokin – the logical IC “2NOT-OR” immediately went into mass production in Germany (1962). These companies began serial production of IP almost simultaneously, in 1962.
First semiconductor ICs in the USA
IP by Jack Kilby. IS series SN - 51”
In 1958, J. Kilby (a pioneer in the use of transistors in hearing aids) moved to Texas Instruments. The newcomer Kilby, as a circuit designer, was “thrown” into improving the micromodular filling of rockets by creating an alternative to micromodules. The option of assembling blocks from parts was considered standard form, similar to assembling toy models from LEGO figures. However, Kilby was fascinated by something else. The decisive role was played by the effect of a “fresh look”: firstly, he immediately stated that micromodules are a dead end, and secondly, having admired the mesa structures, he came to the idea that the circuit should (and can) be implemented from one material - a semiconductor. Kilby knew about Dummer's idea and his unsuccessful attempt to implement it in 1956. After analyzing, he understood the reason for the failure and found a way to overcome it. “ My credit is that I took this idea and turned it into reality.”, J. Kilby said later in his Nobel speech.
Having not yet earned the right to leave, he worked in the laboratory without interference while everyone was resting. On July 24, 1958, Kilby formulated a concept in a laboratory journal called the Monolithic Idea. Its essence was that “. ..circuit elements such as resistors, capacitors, distributed capacitors and transistors can be integrated into a single chip - provided that they are made of the same material... In a flip-flop circuit design, all elements must be made of silicon, with resistors being use the volume resistance of silicon, and capacitors - the capacitance of p-n junctions". The “monolith idea” met with a condescending and ironic attitude from the management of Texas Instruments, which demanded proof of the possibility of manufacturing transistors, resistors and capacitors from a semiconductor and the operability of a circuit assembled from such elements.
In September 1958, Kilby realized his idea - he made a generator from two pieces of germanium measuring 11.1 x 1.6 mm, glued together with beeswax on a glass substrate, containing two types of diffusion regions (Fig. 1). He used these areas and the existing contacts to create a generator circuit, connecting the elements with thin gold wires with a diameter of 100 microns using thermocompression welding. A mesatransistor was created from one area, and an RC circuit was created from the other. The assembled three generators were demonstrated to the company management. When the power was connected, they started working at a frequency of 1.3 MHz. This happened on September 12, 1958. A week later, Kilby made an amplifier in a similar manner. But these were not yet integrated structures, these were three-dimensional mock-ups of semiconductor ICs, proving the idea of manufacturing all circuit elements from one material - a semiconductor.
Rice. 3. Trigger Type 502 J. Kilby. Photo from the site http://www.computerhistory.org/semiconductor/timeline/1958-Miniaturized.html
Kilby's first truly integrated circuit, made in a single piece of monolithic germanium, was the experimental Type 502 trigger IC (Fig. 3). It used both the volume resistance of germanium and the capacitance of the p-n junction. Its presentation took place in March 1959. A small number of such ICs were manufactured in laboratory conditions and sold to a small circle for $450. The IC contained six elements: four mesa transistors and two resistors, placed on a silicon wafer with a diameter of 1 cm. But the Kilby IC had a serious drawback - mesa transistors, which in the form of microscopic “active” columns towered above the rest, “passive” part of the crystal. The connection of mesa columns to each other in the Kilby IS was carried out by boiling thin gold wires - the “hairy technology” hated by everyone. It became clear that with such interconnections a microcircuit with a large number of elements cannot be made - the wire web will break or reconnect. And germanium at that time was already considered as a non-promising material. There was no breakthrough.
By this time, Fairchild had developed planar silicon technology. Given all this, Texas Instruments had to put everything Kilby had done aside and begin, without Kilby, to develop a series of ICs based on planar silicon technology. In October 1961, the company announced the creation of a series of ICs of the SN-51 type, and in 1962 it began their mass production and deliveries in the interests of the US Department of Defense and NASA.
IP by Robert Noyce. IS seriesMicrologic”
In 1957, for a number of reasons, W. Shockley, the inventor of the planar transistor, left a group of eight young engineers who wanted to try to implement their own ideas. “The Eight Traitors,” as Shockley called them, whose leaders were R. Noyce and G. Moore, founded the company Fairchild Semiconductor (“beautiful child”). The company was headed by Robert Noyce, he was then 23 years old.
At the end of 1958, physicist D. Horney, who worked at Fairchild Semiconductor, developed planar technology for manufacturing transistors. And Czech-born physicist Kurt Lehovec, who worked at Sprague Electric, developed a technique for using a reverse-connected n-p junction to electrically isolate components. In 1959, Robert Noyce, having heard about Kilby's IC design, decided to try to create an integrated circuit by combining the processes proposed by Horney and Lehovec. And instead of “hairy technology” of interconnects, Noyce proposed selective deposition of a thin layer of metal on top of silicon dioxide-insulated semiconductor structures with connection to the contacts of the elements through holes left in the insulating layer. This made it possible to “immerse” the active elements in the body of the semiconductor, insulating them with silicon oxide, and then connect these elements with sputtered tracks of aluminum or gold, which are created using the processes of photolithography, metallization and etching at the last stage of product manufacturing. Thus, a truly “monolithic” version of combining components into a single circuit was obtained, and the new technology was called “planar”. But first the idea had to be tested.
Rice. 4. Experimental trigger by R. Noyce. Photo from the site http://www.computerhistory.org/semiconductor/timeline/1960-FirstIC.html
Rice. 5. Photo of Micrologic IC in Life magazine. Photo from the site http://www.computerhistory.org/semiconductor/timeline/1960-FirstIC.html
In August 1959, R. Noyce commissioned Joy Last to develop a version of the IC based on planar technology. First, like Kilby, they made a prototype of a trigger on several silicon crystals, on which 4 transistors and 5 resistors were made. Then, on May 26, 1960, the first single-chip trigger was manufactured. To isolate the elements in it with reverse side The silicon wafer was etched with deep grooves filled with epoxy resin. On September 27, 1960, a third version of the trigger was manufactured (Fig. 4), in which the elements were isolated by a reverse-connected p-n junction.
Until that time, Fairchild Semiconductor was only involved in transistors; it did not have circuit designers to create semiconductor ICs. Therefore, Robert Norman from Sperry Gyroscope was invited as a circuit designer. Norman was familiar with resistor-transistor logic, which the company, at his suggestion, chose as the basis for its future “Micrologic” series of ICs, which found its first application in the equipment of the Minuteman rocket. In March 1961, Fairchild announced the first experimental IC of this series (F-flip-flop containing six elements: four bipolar transistors and two resistors placed on a plate with a diameter of 1 cm) with the publication of its photograph (Fig. 5) in the magazine Life(dated March 10, 1961). Another 5 IPs were announced in October. And from the beginning of 1962, Fairchild launched mass production of ICs and their supply also in the interests of the US Department of Defense and NASA.
Kilby and Noyce had to listen to a lot of criticism about their innovations. It was believed that the practical yield of suitable integrated circuits would be very low. It is clear that it should be lower than that of transistors (since it contains several transistors), for which it was then no higher than 15%. Secondly, many believed that inappropriate materials were used in integrated circuits, since resistors and capacitors were not made from semiconductors at that time. Thirdly, many could not accept the idea of non-repairability of the IP. It seemed blasphemous to them to throw away a product in which only one of many elements had failed. All doubts were gradually cast aside when integrated circuits were successfully used in the US military and space programs.
One of the founders of Fairchild Semiconductor, G. Moore, formulated the basic law of the development of silicon microelectronics, according to which the number of transistors in an integrated circuit crystal doubled every year. This law, called “Moore's Law,” operated quite clearly for the first 15 years (starting in 1959), and then this doubling occurred in about a year and a half.
Further, the IP industry in the United States began to develop at a rapid pace. In the United States, an avalanche-like process of the emergence of enterprises oriented exclusively “for planar” began, sometimes reaching the point that a dozen companies were registered a week. Striving for veterans (the firms of W. Shockley and R. Noyce), as well as thanks to tax incentives and service provided by Stanford University, the “newcomers” clustered mainly in the Santa Clara Valley (California). Therefore, it is not surprising that in 1971, with the light hand of journalist and popularizer of technical innovations Don Hofler, the romantic-technological image of “Silicon Valley” came into circulation, forever becoming synonymous with the Mecca of the semiconductor technological revolution. By the way, in that area there really is a valley that was previously famous for its numerous apricot, cherry and plum orchards, which before the appearance of the Shockley company had another, more pleasant name - the Valley of Heart's Delight, now, unfortunately, almost forgotten.
In 1962, mass production of integrated circuits began in the United States, although their volume of deliveries to customers amounted to only a few thousand. The strongest incentive for the development of the instrument-making and electronics industry on a new basis was rocket and space technology. The United States did not then have the same powerful intercontinental ballistic missiles as the Soviet ones, and in order to increase the charge they were forced to minimize the mass of the carrier, including control systems, through the introduction of the latest advances in electronic technology. Texas Instrument and Fairchild Semiconductor have entered into large contracts for the design and manufacture of integrated circuits with the US Department of Defense and NASA.
The first semiconductor ICs in the USSR
By the late 1950s, Soviet industry was so desperate for semiconductor diodes and transistors that radical measures were required. In 1959, semiconductor device factories were founded in Aleksandrov, Bryansk, Voronezh, Riga, etc. In January 1961, the CPSU Central Committee and the USSR Council of Ministers adopted another Resolution “On the development of the semiconductor industry,” which provided for the construction of factories and research institutes in Kyiv, Minsk, Yerevan, Nalchik and other cities.
We will be interested in one of the new factories - the above-mentioned Riga Semiconductor Devices Plant (RZPP, it changed its names several times, for simplicity we use the most famous one, which is still in operation today). The building of the cooperative technical school under construction with an area of 5300 m2 was allocated as a launching pad for the new plant, and at the same time construction of a special building began. By February 1960, the plant had already created 32 services, 11 laboratories and pilot production, which began in April to prepare for the production of the first devices. The plant already employed 350 people, 260 of whom were sent to study at the Moscow Research Institute-35 (later the Pulsar Research Institute) and the Leningrad Svetlana plant during the year. And by the end of 1960, the number of employees reached 1,900 people. Initially, the technological lines were located in the rebuilt sports hall of the cooperative technical school building, and the OKB laboratories were located in the former classrooms. The plant produced the first devices (alloy-diffusion and conversion germanium transistors P-401, P-403, P-601 and P-602 developed by NII-35) 9 months after the order for its creation was signed, in March 1960. And by the end of July, he manufactured the first thousand P-401 transistors. Then he mastered the production of many other transistors and diodes. In June 1961, construction of a special building was completed, in which mass production of semiconductor devices began.
Since 1961, the plant began independent technological and development work, including mechanization and automation of the production of transistors based on photolithography. For this purpose, the first domestic photo repeater (photo stamp) was developed - an installation for combining and contact photo printing (developed by A.S. Gotman). Great assistance in financing and manufacturing unique equipment was provided by enterprises of the Ministry of Radio Industry, including KB-1 (later NPO Almaz, Moscow) and NIIRE. At that time, the most active developers of small-sized radio equipment, not having their own technological semiconductor base, were looking for ways to creatively interact with newly created semiconductor factories.
At RZPP, active work was carried out to automate the production of germanium transistors of the P401 and P403 types based on the Ausma production line created by the plant. Its chief designer (GC) A.S. Gottman proposed making current-carrying paths on the surface of germanium from the electrodes of the transistor to the periphery of the crystal to make it easier to weld the transistor leads in the housing. But most importantly, these tracks could be used as external terminals of the transistor when they were assembled into boards (containing connecting and passive elements) without packaging, soldering them directly to the corresponding contact pads (in fact, the technology for creating hybrid ICs was proposed). The proposed method, in which the current-carrying paths of the crystal seem to kiss the contact pads of the board, received the original name - “kissing technology”. But due to a number of technological problems that turned out to be insoluble at that time, mainly related to problems with the accuracy of obtaining contacts on a printed circuit board, it was not possible to practically implement the “kiss technology”. A few years later, a similar idea was implemented in the USA and the USSR and found wide application in the so-called “ball leads” and in “chip-to-board” technology.
However, hardware companies cooperating with RZPP, including NIIRE, hoped for “kiss technology” and planned its use. In the spring of 1962, when it became clear that its implementation was postponed indefinitely, chief engineer of NIIRE V.I. Smirnov asked the director of the RZPP S.A. Bergman to find another way to implement a multi-element 2NOR circuit, universal for building digital devices.
Rice. 7. Equivalent circuit of IC R12-2 (1LB021). Drawing from the 1965 IP prospectus.
The first IS and GIS by Yuri Osokin. Solid scheme R12-2(IS series 102 And 116 )
The director of the RZPP entrusted this task to the young engineer Yuri Valentinovich Osokin. We organized a department consisting of a technology laboratory, a laboratory for the development and production of photo masks, a measuring laboratory and a pilot production line. At that time, the technology for manufacturing germanium diodes and transistors was supplied to RZPP, and it was taken as the basis for the new development. And already in the fall of 1962, the first prototypes of the germanium solid circuit 2NOT-OR were obtained (since the term IS did not exist then, out of respect for the affairs of those days, we will retain the name “hard circuit” - TS), which received the factory designation “P12-2”. An advertising booklet from 1965 on P12-2 has survived (Fig. 6), information and illustrations from which we will use. TS R12-2 contained two germanium p - n - p -transistors (modified transistors of type P401 and P403) with a common load in the form of a distributed germanium p-type resistor (Fig. 7).
Rice. 8. Structure of IC R12-2. Drawing from the 1965 IP prospectus.
Rice. 9. Dimensional drawing of vehicle R12-2. Drawing from the 1965 IP prospectus.
External leads are formed by thermocompression welding between the germanium regions of the TC structure and the gold of the lead conductors. This ensures stable operation of the circuits under external influences in tropical and sea fog conditions, which is especially important for operation in naval quasi-electronic automatic telephone exchanges produced by the Riga VEF plant, which was also interested in this development.
Structurally, the R12-2 TS (and the subsequent R12-5) were made in the form of a “tablet” (Fig. 9) from a round metal cup with a diameter of 3 mm and a height of 0.8 mm. The TC crystal was placed in it and filled with a polymer compound, from which came the short outer ends of the leads made of soft gold wire with a diameter of 50 microns, welded to the crystal. The mass of P12-2 did not exceed 25 mg. In this design, the vehicles were resistant to relative humidity of 80% at an ambient temperature of 40 ° C and to cyclic temperature changes from -60 ° to 60 ° C.
By the end of 1962, the pilot production of RZPP produced about 5 thousand R12-2 vehicles, and in 1963 several tens of thousands of them were made. Thus, 1962 became the year of birth of the microelectronic industry in the USA and the USSR.
Rice. 10. Groups TS R12-2
Rice. 11. Basic electrical characteristics of R12-2
Semiconductor technology was then in its infancy and did not yet guarantee strict repeatability of parameters. Therefore, operable devices were sorted into groups of parameters (this is often done in our time). The residents of Riga did the same, installing 8 standard ratings of the R12-2 vehicle (Fig. 10). All other electrical and other characteristics are the same for all standard ratings (Fig. 11).
The production of TS R12-2 began simultaneously with the R&D “Hardness”, which ended in 1964 (GK Yu.V. Osokin). As part of this work, an improved group technology for the serial production of germanium vehicles was developed based on photolithography and galvanic deposition of alloys through a photomask. Its main technical solutions are registered as an invention by Yu.V. Osokin. and Mikhalovich D.L. (A.S. No. 36845). Several articles by Yu.V. were published in the classified journal Spetsradioelectronics. Osokina in collaboration with KB-1 specialists I.V. Nothing, G.G. Smolko and Yu.E. Naumov with a description of the design and characteristics of the R12-2 vehicle (and the subsequent R12-5 vehicle).
The design of the P12-2 was good in everything, except for one thing - consumers did not know how to use such small products with the thinnest leads. As a rule, hardware companies had neither the technology nor the equipment for this. Over the entire period of production of R12-2 and R12-5, their use was mastered by NIIRE, the Zhigulevsky Radio Plant of the Ministry of Radio Industry, VEF, NIIP (since 1978 NPO Radiopribor) and a few other enterprises. Understanding the problem, the TS developers, together with NIIRE, immediately thought of a second level of design, which at the same time increased the density of the equipment layout.
Rice. 12. Module of 4 vehicles R12-2
In 1963, at NIIRE, within the framework of the Kvant design and development work (GK A.N. Pelipenko, with the participation of E.M. Lyakhovich), a module design was developed that combined four R12-2 vehicles (Fig. 12). From two to four R12-2 TCs (in a housing) were placed on a microboard made of thin fiberglass, which collectively implemented a certain functional unit. Up to 17 pins (the number varied for a specific module) with a length of 4 mm were pressed onto the board. The microboard was placed in a stamped metal cup measuring 21.6 ? 6.6 mm and 3.1 mm deep and filled with a polymer compound. The result is a hybrid integrated circuit (HIC) with double sealing of elements. And, as we already said, it was the world's first GIS with two-level integration, and, perhaps, the first GIS in general. Eight types of modules were developed with the general name “Quantum”, which performed various logical functions. As part of such modules, the R12-2 vehicles remained operational when exposed to constant accelerations of up to 150 g and vibration loads in the frequency range of 5–2000 Hz with acceleration up to 15 g.
The Kvant modules were first produced by the pilot production of NIIRE, and then they were transferred to the Zhigulevsky Radio Plant of the USSR Ministry of Radio Industry, which supplied them to various consumers, including the VEF plant.
TS R12-2 and “Kvant” modules based on them have proven themselves well and are widely used. In 1968, a standard was issued establishing a unified designation system for integrated circuits in the country, and in 1969 - General technical specifications for semiconductor (NP0.073.004TU) and hybrid (NP0.073.003TU) ICs with unified system requirements. In accordance with these requirements, the Central Bureau for the Application of Integrated Circuits (TsBPIMS, later CDB Dayton, Zelenograd) on February 6, 1969 approved new technical specifications ShT3.369.001-1TU for the vehicle. At the same time, the term “integrated circuit” of the 102 series appeared for the first time in the designation of the product. TS R12-2 began to be called IS: 1LB021V, 1LB021G, 1LB021Zh, 1LB021I. In fact, it was one IC, sorted into four groups according to output voltage and load capacity.
Rice. 13. 116 and 117 series ICs
And on September 19, 1970, TsBPIMS approved the technical specifications AB0.308.014TU for the Kvant modules, designated IS series 116 (Fig. 13). The series included nine ICs: 1ХЛ161, 1ХЛ162 and 1ХЛ163 – multifunctional digital circuits; 1LE161 and 1LE162 – two and four logical elements 2NOR; 1TP161 and 1TP1162 – one and two triggers; 1UP161 – power amplifier, as well as 1LP161 – logic element"ban" on 4 inputs and 4 outputs. Each of these ICs had from four to seven design options, differing in output signal voltage and load capacity, for a total of 58 IC types. The designs were marked with a letter after the digital part of the IS designation, for example, 1ХЛ161ж. Subsequently, the range of modules expanded. The ICs of the 116 series were actually hybrid, but at the request of RZPP they were labeled as semiconductor (the first digit in the designation is “1”, hybrid ones should have “2”).
In 1972, by a joint decision of the Ministry of Electronics Industry and the Ministry of Radio Industry, the production of modules was transferred from the Zhigulevsky Radio Plant to RZPP. This eliminated the possibility of transporting the 102 series ICs over long distances, so they abandoned the need to seal the die of each IC. As a result, the design of both the 102 and 116 series ICs was simplified: there was no need to package the 102 series ICs in a metal cup filled with compound. Unpackaged ICs of the 102 series in technological containers were delivered to a neighboring workshop for the assembly of ICs of the 116 series, mounted directly on their microboard and sealed in the module housing.
In the mid-1970s, a new standard for the IP designation system was released. After this, for example, IS 1LB021V received the designation 102LB1V.
Second IS and GIS by Yuri Osokin. Solid scheme R12-5(IS series 103 And 117 )
By the beginning of 1963, as a result of serious work on the development of high-frequency n - p - n transistors, the team of Yu.V. Osokina has accumulated extensive experience working with p-layers on the original n-germanium wafer. This and the presence of all the necessary technological components allowed Osokin in 1963 to begin developing new technology and the design of a faster version of the vehicle. In 1964, by order of NIIRE, the development of the R12-5 vehicle and modules based on it was completed. Based on its results, the Palanga R&D was opened in 1965 (GK Yu.V. Osokin, his deputy - D.L. Mikhalovich, completed in 1966). Modules based on the R12-5 were developed within the framework of the same R&D project “Kvant” as the modules based on the R12-2. Simultaneously with the technical specifications for the 102 and 116 series, the technical specifications ShT3.369.002-2TU for the 103 series IC (R12-5) and AV0.308.016TU for the 117 series IC (modules based on the 103 series IC) were approved. The nomenclature of types and standard ratings of TS R12-2, modules on them and IS series 102 and 116 was identical to the nomenclature of TS R12-5 and IS series 103 and 117, respectively. They differed only in speed and manufacturing technology of the IC crystal. The typical propagation delay time of the 117 series was 55 ns versus 200 ns for the 116 series.
Structurally, the R12-5 TS was a four-layer semiconductor structure (Fig. 14), where the n-type substrate and p + -type emitters were connected to a common ground bus. The main technical solutions for constructing the R12-5 vehicle are registered as the invention of Yu.V. Osokin, D.L. Mikhalovich. Kaydalova Zh.A and Akmensa Ya.P. (A.S. No. 248847). When manufacturing the four-layer structure of the TC R12-5, an important know-how was the formation of an n-type p-layer in the original germanium plate. This was achieved by diffusion of zinc in a sealed quartz ampoule, where the plates are located at a temperature of about 900 ° C, and zinc is located at the other end of the ampoule at a temperature of about 500 ° C. The further formation of the TS structure in the created p-layer is similar to the P12-2 TS. New technology has made it possible to avoid the complex shape of the TS crystal. Wafers with P12-5 were also ground from the back to a thickness of about 150 microns, preserving part of the original wafer, and then they were scribed into individual rectangular IC chips.
Rice. 14. Structure of the TS R12-5 crystal from AS No. 248847. 1 and 2 – ground, 3 and 4 – inputs, 5 – output, 6 – power
After the first positive results production of experimental R12-5 vehicles, by order of KB-1, the Mezon-2 research project was opened, aimed at creating a vehicle with four R12-5s. In 1965, working samples in a flat metal-ceramic case were obtained. But P12-5 turned out to be difficult to manufacture, mainly due to the difficulty of forming a zinc-doped p-layer on the original n-Ge wafer. The crystal turned out to be labor-intensive to produce, the yield percentage is low, and the cost of the vehicle is high. For the same reasons, the R12-5 TC was produced in small volumes and could not displace the slower, but more technologically advanced R12-2. And the Mezon-2 research project was not continued at all, including due to interconnection problems.
By this time, the Pulsar Research Institute and the NIIME were already carrying out extensive work on the development of planar silicon technology, which has a number of advantages over germanium technology, the main of which is a higher operating temperature range (+150°C for silicon and +70°C for germanium) and the presence of natural silicon protective film SiO2. And the specialization of RZPP was reoriented to the creation of analog ICs. Therefore, RZPP specialists considered the development of germanium technology for the production of ICs inappropriate. However, in the production of transistors and diodes, germanium did not lose its position for some time. In the department of Yu.V. Osokin, after 1966, RZPP germanium planar low-noise microwave transistors GT329, GT341, GT 383, etc. were developed and produced. Their creation was awarded the State Prize of the Latvian USSR.
Application
Rice. 15. Arithmetic device on solid-circuit modules. Photo from the TS booklet dated 1965.
Rice. 16. Comparative dimensions of the automatic telephone exchange control device, made on a relay and a vehicle. Photo from the TS booklet dated 1965.
The customers and first consumers of the R12-2 TS and modules were the creators of specific systems: the Gnome computer (Fig. 15) for the Kupol on-board aircraft system (NIIRE, GK Lyakhovich E.M.) and naval and civil automatic telephone exchanges (plant VEF, GK Misulovin L.Ya.). Actively participated in all stages of the creation of the R12-2, R12-5 vehicles and modules on them and KB-1, the main curator of this cooperation from KB-1 was N.A. Barkanov. They helped with financing, equipment manufacturing, and research of vehicles and modules in various modes and operating conditions.
TS R12-2 and “Kvant” modules based on it were the first microcircuits in the country. And in the world they were among the first - only in the USA did Texas Instruments and Fairchild Semiconductor begin to produce their first semiconductor ICs, and in 1964 the IBM Corporation began producing thick-film hybrid ICs for its computers. In other countries, IP has not yet been thought about. Therefore, integrated circuits were a curiosity to the public; the effectiveness of their use made a striking impression and was played up in advertising. The surviving booklet on the R12-2 vehicle from 1965 (based on actual applications) says: “ The use of solid-state P12-2 circuits in on-board computing devices makes it possible to reduce the weight and dimensions of these devices by 10–20 times, reduce power consumption and increase operational reliability. ... The use of solid P12-2 circuits in control systems and switching of information transmission paths of automatic telephone exchanges makes it possible to reduce the volume of control devices by approximately 300 times, as well as significantly reduce electricity consumption (30-50 times)" . These statements were illustrated by photographs of the arithmetic device of the Gnome computer (Fig. 15) and a comparison of the relay-based ATS rack produced by the VEF plant at that time with a small block on the girl’s palm (Fig. 16). There were other numerous applications of the first Riga ICs.
Production
Now it is difficult to restore a complete picture of the production volumes of IC series 102 and 103 by year (today RZPP has turned from a large plant into a small production and many archives have been lost). But according to the memoirs of Yu.V. Osokin, in the second half of the 1960s, production amounted to many hundreds of thousands per year, in the 1970s - millions. According to his surviving personal notes, in 1985, ICs of the 102 series were produced - 4,100,000 pcs., modules of the 116 series - 1,025,000 pcs., ICs of the 103 series - 700,000 pcs., modules of the 117 series - 175,000 pcs.
At the end of 1989, Yu.V. Osokin, then the general director of Alpha Production Association, turned to the leadership of the Military-Industrial Commission under the USSR Council of Ministers (MIC) with a request to remove series 102, 103, 116 and 117 from production due to their obsolescence and high labor intensity (in 25 years, microelectronics is far from went ahead), but received a categorical refusal. Deputy Chairman of the Military-Industrial Complex V.L. Koblov told him that the planes fly reliably, replacement is excluded. After the collapse of the USSR, IC series 102, 103, 116 and 117 were produced until the mid-1990s, i.e. for more than 30 years. The Gnome computers are still installed in the navigation cabin of the Il-76 and some other aircraft. “This is a supercomputer,” our pilots are not at a loss when their foreign colleagues are surprised by their interest in this unprecedented device.
About priorities
Despite the fact that J. Kilby and R. Noyce had predecessors, they are recognized by the world community as the inventors of the integrated circuit.
R. Kilby and J. Noyce, through their firms, filed applications for a patent for the invention of an integrated circuit. Texas Instruments applied for a patent earlier, in February 1959, and Fairchild did not do so until July of that year. But patent number 2981877 was issued in April 1961 to R. Noyce. J. Kilby sued and only in June 1964 received his patent number 3138743. Then there was a ten-year war about priorities, as a result of which (in a rare case) “friendship won.” Ultimately, the Court of Appeal upheld Noyce's claim to technological primacy, but ruled that J. Kilby should be credited with creating the first working microcircuit. And Texas Instruments and Fairchild Semiconductor signed an agreement on cross-licensing technologies.
In the USSR, patenting inventions did not give authors anything other than hassle, an insignificant one-time payment and moral satisfaction, so many inventions were not registered at all. And Osokin was in no hurry either. But for enterprises, the number of inventions was one of the indicators, so they still had to be registered. Therefore, Yu. Osokina and D. Mikhalovich received the USSR Author's Certificate No. 36845 for the invention of the R12-2 vehicle only on June 28, 1966.
And J. Kilby in 2000 became one of the Nobel Prize laureates for the invention of IP. R. Noyce did not receive world recognition; he died in 1990, and according to the regulations, the Nobel Prize is not awarded posthumously. Which, in this case, is not entirely fair, since all microelectronics followed the path begun by R. Noyce. Noyce’s authority among specialists was so high that he even received the nickname “mayor of Silicon Valley,” since he was then the most popular of the scientists working in that part of California, which received the unofficial name Silicon Valley (V. Shockley was called “Moses of Silicon Valley”) . But the path of J. Kilby (“hairy” germanium) turned out to be a dead end, and was not implemented even in his company. But life is not always fair.
The Nobel Prize was awarded to three scientists. Half of it was received by 77-year-old Jack Kilby, and the other half was divided between academician of the Russian Academy of Sciences Zhores Alferov and professor at the University of California at Santa Barbara, German-American Herbert Kremer, for “the development of semiconductor heterostructures used in high-speed optoelectronics.”
Evaluating these works, experts noted that “integrated circuits are, of course, the discovery of the century, which has had a profound impact on society and the world economy.” For the forgotten J. Kilby, the Nobel Prize was a surprise. In an interview with the magazine Europhysics News He admitted: " At that time, I was only thinking about what would be important for the development of electronics from an economic point of view. But I didn’t understand then that the reduction in the cost of electronic products would cause an avalanche of growth in electronic technologies.”.
And the works of Yu. Osokin are not appreciated not only by the Nobel Committee. They are also forgotten in our country; the country’s priority in the creation of microelectronics is not protected. And he undoubtedly was.
In the 1950s, the material basis was created for the formation of multi-element products - integrated circuits - in one monolithic crystal or on one ceramic substrate. Therefore, it is not surprising that almost simultaneously the idea of IP independently arose in the minds of many specialists. And the speed of implementation of a new idea depended on the technological capabilities of the author and the interest of the manufacturer, i.e., on the presence of the first consumer. In this regard, Yu. Osokin found himself in a better position than his American colleagues. Kilby was new to TI, he even had to prove to the company's management the fundamental possibility of implementing a monolithic circuit by making its prototype. Actually, the role of J. Kilby in the creation of the IP comes down to re-educating the management of TI and provoking R. Noyce to take active action with his layout. Kilby's invention did not go into mass production. R. Noyce, in his young and not yet strong company, went to create a new planar technology, which indeed became the basis for subsequent microelectronics, but did not immediately yield to the author. In connection with the above, both of them and their companies had to spend a lot of effort and time to practically implement their ideas for building mass-produced ICs. Their first samples remained experimental, but other microcircuits, not even developed by them, went into mass production. Unlike Kilby and Noyce, who were far from production, factory owner Yu. Osokin relied on industrially developed semiconductor RZPP technologies, and he had guaranteed consumers of the first vehicles in the form of the initiator of the development of NIIRE and the nearby VEF plant, which helped in this work. For these reasons, the first version of his vehicle immediately went into experimental production, which smoothly transitioned into mass production, which continued continuously for more than 30 years. Thus, having started developing the TS later than Kilby and Noyce, Yu. Osokin (not knowing about this competition) quickly caught up with them. Moreover, the works of Yu. Osokin are in no way connected with the works of the Americans, evidence of this is the absolute dissimilarity of his vehicle and the solutions implemented in it from the Kilby and Noyce microcircuits. Texas Instruments (not Kilby's invention), Fairchild and RZPP began production of their ICs almost simultaneously, in 1962. This gives every right to consider Yu. Osokin one of the inventors of the integrated circuit on a par with R. Noyce and more than J. Kilby, and it would be fair to share part of the Nobel Prize for J. Kilby with Yu. Osokin. As for the invention of the first GIS with two-level integration (and possibly GIS in general), here priority A. Pelipenko from NIIRE is absolutely indisputable.
Unfortunately, it was not possible to find samples of vehicles and devices based on them, necessary for museums. The author would be very grateful for such samples or photographs of them.
Integrated circuit (IC, microcircuit), chip, microchip (English microchip, silicon chip, chip - thin plate - originally the term referred to a plate of a microcircuit crystal) - microelectronic device - electronic circuit of arbitrary complexity (crystal), manufactured on a semiconductor substrate (wafer or film) and placed in a non-separable housing, or without one, if included in a microassembly.
Microelectronics is the most significant and, as many believe, the most important scientific and technical achievement of our time. It can be compared with such turning points in the history of technology as the invention of printing in the 16th century, the creation of the steam engine in the 18th century, and the development of electrical engineering in the 19th century. And when today we talk about the scientific and technological revolution, we primarily mean microelectronics. Like no other technical achievement of our days, it permeates all spheres of life and makes reality what was simply unimaginable just yesterday. To be convinced of this, it is enough to remember pocket calculators, miniature radios, electronic control devices in household appliances, watches, computers and programmable computers. And this is only a small part of its application area!
Microelectronics owes its emergence and very existence to the creation of a new subminiature electronic element - an integrated circuit. The appearance of these circuits, in fact, was not some kind of fundamentally new invention - it directly followed from the logic of the development of semiconductor devices. At first, when semiconductor elements were just coming into use, each transistor, resistor or diode was used separately, that is, it was enclosed in its own individual case and included in the circuit using its individual contacts. This was done even in cases where it was necessary to assemble many similar circuits from the same elements.
Gradually, the understanding came that it was more rational not to assemble such devices from individual elements, but to immediately manufacture them on one common crystal, especially since semiconductor electronics created all the prerequisites for this. In fact, all semiconductor elements are very similar to each other in their structure, have the same principle of operation and differ only in the relative position of the p-n regions.
These p-n regions, as we remember, are created by introducing impurities of the same type into the surface layer of a semiconductor crystal. Moreover, reliable and from all points of view satisfactory operation of the vast majority of semiconductor elements is ensured with a thickness of the surface working layer of thousandths of a millimeter. The smallest transistors typically use only the top layer of the semiconductor chip, which is only 1% of its thickness. The remaining 99% acts as a carrier or substrate, since without a substrate the transistor could simply collapse at the slightest touch. Consequently, using the technology used for the manufacture of individual electronic components, it is possible to immediately create a complete circuit of several tens, hundreds, or even thousands of such components on a single chip.
The benefits from this will be huge. Firstly, costs will immediately decrease (the cost of a microcircuit is usually hundreds of times less than the total cost of all the electronic elements of its components). Secondly, such a device will be much more reliable (as experience shows, thousands and tens of thousands of times), and this is of enormous importance, since finding a fault in a circuit consisting of tens or hundreds of thousands of electronic components turns into an extremely complex problem. Thirdly, due to the fact that all the electronic elements of an integrated circuit are hundreds and thousands of times smaller than their counterparts in a conventional circuit, their energy consumption is much lower and their performance is much higher.
The key event that heralded the arrival of integration in electronics was the proposal of the American engineer J. Kilby from Texas Instruments to obtain equivalent elements for the entire circuit, such as registers, capacitors, transistors and diodes, in a monolithic piece of pure silicon. Kilby created the first integrated semiconductor circuit in the summer of 1958. And already in 1961, Fairchild Semiconductor Corporation released the first serial chips for computers: a coincidence circuit, a half-shift register and a trigger. In the same year, the production of semiconductor integrated logic circuits mastered by the Texas company.
The following year, integrated circuits from other companies appeared. IN a short time in integral design were created Various types amplifiers. In 1962, RCA developed integrated memory matrix chips for computer storage devices. Gradually, the production of microcircuits was established in all countries - the era of microelectronics began.
The starting material for an integrated circuit is usually a raw wafer of pure silicon. It has a relatively large size, since several hundred of the same type of microcircuits are simultaneously manufactured on it. The first operation is that under the influence of oxygen at a temperature of 1000 degrees, a layer of silicon dioxide is formed on the surface of this plate. Silicon oxide is characterized by great chemical and mechanical resistance and has the properties of an excellent dielectric, providing reliable insulation to the silicon located underneath.
The next step is the introduction of impurities to create p or n conduction bands. To do this, the oxide film is removed from those places on the plate that correspond to individual electronic components. The selection of the desired areas occurs using a process called photolithography. First, the entire oxide layer is coated with a photosensitive compound (photoresist), which plays the role of photographic film - it can be exposed and developed. After this, through a special photomask containing a pattern of the surface of the semiconductor crystal, the plate is illuminated with ultraviolet rays.
Under the influence of light, a flat pattern is formed on the oxide layer, with unexposed areas remaining light and all others darkened. In the place where the photoresistor is exposed to light, insoluble areas of the film are formed that are resistant to acid. The wafer is then treated with a solvent, which removes the photoresist from the exposed areas. From the exposed areas (and only from them), the silicon oxide layer is etched away using acid.
As a result, silicon oxide dissolves in the right places and “windows” of pure silicon open, ready for the introduction of impurities (ligation). To do this, the surface of the substrate at a temperature of 900-1200 degrees is exposed to the desired impurity, for example, phosphorus or arsenic, to obtain n-type conductivity. Impurity atoms penetrate deep into pure silicon, but are repelled by its oxide. Having treated the wafer with one type of impurity, it is prepared for ligation with another type - the surface of the wafer is again covered with a layer of oxide, new photolithography and etching are carried out, as a result of which new “windows” of silicon are opened.
This is followed by a new ligation, for example with boron, to obtain p-type conductivity. So, p and n regions are formed on the entire surface of the crystal in the right places. Insulation between individual elements can be created in several ways: a layer of silicon oxide can serve as such insulation, or blocking p-n junctions can also be created in the right places.
The next stage of processing is associated with the application of conductive connections (conducting lines) between the elements of the integrated circuit, as well as between these elements and contacts for connecting external circuits. To do this, a thin layer of aluminum is sprayed onto the substrate, which settles in the form of a thin film. It is subjected to photolithographic processing and etching similar to those described above. As a result, only thin conductive lines and contact pads remain from the entire metal layer.
Finally, the entire surface of the semiconductor chip is covered with a protective layer (most often silicate glass), which is then removed from the contact pads. All manufactured microcircuits are subjected to the strictest testing at a control and test bench. Defective circuits are marked with a red dot. Finally, the crystal is cut into individual chip plates, each of which is enclosed in a durable housing with leads for connection to external circuits.
The complexity of an integrated circuit is characterized by an indicator called the degree of integration. Integrated circuits with more than 100 elements are called low-integration circuits; circuits containing up to 1000 elements - integrated circuits with a medium degree of integration; circuits containing up to tens of thousands of elements are called large integrated circuits. Circuits containing up to a million elements are already being manufactured (they are called ultra-large). The gradual increase in integration has led to the fact that every year the schemes become more and more miniature and, accordingly, more and more complex.
Great amount electronic devices, which previously had large dimensions, now fit on a tiny silicon wafer. An extremely important event on this path was the creation in 1971 by the American company Intel of a single integrated circuit for performing arithmetic and logical operations - a microprocessor. This entailed a grandiose breakthrough of microelectronics into the field of computer technology.
Read and write useful
