Circuits in stone: A video history of Japan's electronic industry (Part 2)

the total annual amount of electronics products that Japan currently exports is approximately 10 trillion 50 billion yen vast amounts of semiconductors are used in all of them Japan began to pursue the lead of the United States in this field from a MIT's post-war hunger and became the world's number one producer of germanium transistors in 1959 however the US had advanced to where it was replacing germanium with silicon into which even electronic circuits were being fabricated in 1962 Japanese Prime Minister hi Otto Ikeda who was visiting Europe was criticized as being a transistor salesman by French president Charles de Gaulle this scene shows the recruitment of female factory workers who were called transistor girls the three requisites for passing the aptitude test were good eyes dexterous hands and perseverance patient girls from farming villages were hired in large numbers after intense education and training the girls worked on the assembly lines each company tried to secure and educate outstanding transistor girls at the time this was the shortcut to winning the war of yields despite the quality of the fingertips and the perseverance of their employees manufacturing yields weren't satisfactory hopefully Giulia hoping we need my still mommy I went to report to the department chief that yesterday the yield was 0.3% simple singular some family in other words just three out of 1,000 were good the remainder were all inferior products in fact in my heart I was hoping that I wouldn't have to report 0% tomorrow there couldn't be a worse yield than 0% a nightmare for a manager well so there are even more L&I huh that the semiconductor industry was equated to phishing is because you don't know how many fish there are unless you cast your net not it was also equated to farming is that the crop is bad if the weather is bad the crop yield is poor especially when the temperature and humidity are high in summer the drawbacks with germanium transistors were that they were easily damaged by moisture humidity and heat and they degraded and stopped working thus creating frequent problems one of the most dangerous threats to our nation's security is the possibility of attack by high-speed enemy bombers armed with nuclear weapon these bombers can strike at supersonic speeds from many directions and altitudes to confuse our defense and delay the dispatch of interceptor weapons in a mass raid high-speed bombers could be in on us before we could determine their tracks and then it would be too late to act the poor characteristics of the transistors used in military computers were a big problem amid the Cold War standoff between the United States and the USSR the search for a stable material to replace germanium was an urgent task for the United States silicon is not highly sensitive to heat and unlike germanium which is a scarce resource it is found everywhere Bell Laboratories where scientists were struggling to develop a silicon transistor was the center of this research in order to compensate for the shortcomings of silicon it being highly active various advances such as the method of covering it with an oxide film were made here Carl frosh and Lincoln Derek thought the only way to handle silicon was to cover the surface with a protective film and so looked for a material to make a strong film to bond with a silicon initially they excluded only oxygen they thought if they were to use oxygen which easily synthesizes with other materials then oxidation would not stop at the surface but would completely permeate the silicon but the bonding of oxygen to silicon became the key to using it this chance discovery happened when a valve broke accidentally I don't think we knew at that time I don't think we really know yes he knew it was big but I don't think anybody at that particular same month or to realize how big it really could I would be about bin 2 because later on of course Bell Labs did write up that that was one of the shusei cornerstones and all integrated circuitry without that you couldn't make it he couldn't make a circuit AT&T which foresaw the coming age of communication satellites vigorously pursued research on solar batteries the work of Calvin fuller and his associates was among the most significant at that time fuller discovered the gas phase diffusion method this would eventually lead to a turning point in semiconductor technology and brought about today's semiconductor industry in the meantime a firm in Japan anticipated the switchover from germanium to silicon and aimed to manufacture highly pure silicon located at yoga Aichi city harbor and me a Prefecture is high silicon company a refiner of silicon for semiconductor use with a monthly production of 90 tons it is the world's number-one producer this is metallic silicon in powdered form shipped from Norway it's 98% pure it is placed inside a tank melted with chloride and turned into silicon chloride gas which is then sent to a distillation tower while the gas is continuously circulated the impurities are gradually removed until the gas becomes highly purified with a negligible impurity content of one part per 100 billion this gas is then reverted into a solid state this is a giant electric cauldron for making solid silicon out of the silicon gas the technicians set approximately 100 silicon seed rods which are cut out from manufactured products electricity is passed through the seed rods while the silicon gas is continuously circulated for a full five days during this time the seed rods grow to a thickness of about 10 centimeters this work has never been shown before even to people involved with a semiconductor industry this process is a Japanese technology that was independently developed about 30 years ago hi silicon is a subsidiary that branched off from the chi sao corporation of Kumamoto Prefecture when it developed this technology shiso was called New Japan nitrogen fertilizer company before the war the company had a factory overseas after it hired a large number of repatriates and was struggling to find new work company executives therefore decided to get involved with silicon which they thought would probably replace germanium the purity of silicon has to be nine to the tenth or eleventh power it's useless if the purity figure isn't at these high levels however at this time we didn't know how to achieve a purity of this strict standard their only clue was a rumor that DuPont in the United States had used silicon tetrachloride in 1956 and experimental furnace was built at the Minamata plant this experimental furnace is a recreation of the original model low-grade silicon is melted with hydrochloric acid turning it into a gas and sent to the furnace into which hydrogen is circulated the silicon separates and sticks to the inside we listed it after that we injected an inert gas that wouldn't explode and gradually replace the hydrogen with it but when we open the furnace lid it exploded BAM there was an error in the design and the hydrogen had accumulated though these explosions repeated they gradually raised the purity of the silicon in this furnace in 1958 they began to build a factory in Chiba Prefecture on land that had been part of a pine forest the land price was cheap and there was not much dust or dirt in the air they built a quartz furnace and began their initial production at the end of 1959 everyone watched with bated breath to see if it would work it did inside the quartz tube ultra pure poly crystal silicon began to densely grow the poly crystal silicon that they manufactured to a purity of ten to the ninth power was remade into single crystal silicon using independently developed machinery they cut the silicon and measured its resistance to judge its quality when the technician announced that the resistance was 200 ohms everyone emphatically question is that correct to explain the rate of resistance changes completely with the way it is measured and depends on various conditions but then the technician declared I can absolutely guarantee that it's 100 ohms or above that we had achieved 100 ohms above was fantastic nothing has made such a deep impression on me as that day everyone was cheering the crystals improved each day and finally they completed a silicon with a resistance of 1000 ohms that greatly exceeded the standard value set in the United States we thought this was sufficient so we went to the Ministry of International Trade and Industry with confidence and in high spirits we've done it we told them but mitties reaction was a rebuff they asked is your method of measuring reliable they wouldn't believe us at all without checking they said that it was useless to think about using this type of domestic product we became angry what a disgrace going south on the freeway from San Francisco vast plains gradually open on both sides of the road this area is popularly known as Silicon Valley there are roughly 4,800 firms here everything from semiconductor makers to electrical equipment manufacturers Silicon Valley which has become the number one electronics industrial area in the world was at the end of the 1950's still filled with orchards here in 1956 dr. William Shockley who discovered the junction transistor recruited outstanding young people from throughout the United States and founded Shockley transistor laboratories this was the first sprout that gave birth to today's Silicon Valley these are the people the best and brightest who were selected by dr. Shockley they were the driving force that created the semiconductor industry among them was the late Robert Noyce who is considered the father of the American semiconductor industry this building is where Shockley had his lab today it is an audio good shop the young people inspired by dr. Shockley soon realized that he was not a good business man and they started looking for opportunities elsewhere one group that left established Fairchild Semiconductor corporation in 1957 an aerial camera equipment company of the same name put up the money led by noise this group boldly experimented with new technologies the first product they produced was the Mesa transistor at the time this was the leading edge of technology IBM which was deeply involved in military contracts took notice of this new Fairchild technology initially Fairchild's orders didn't exceed 100 pieces but they gradually increased and Fairchild began setting up a string of factories here's the process and structure of the fairchild mesa transistor first n-type silicon is prepared and placed inside a diffusion furnace and sealed this liquid is a boron compound used as a p-type impurity it is mixed with an inert gas and injected into the furnace the ratio of boron gas forced in is carefully adjusted this is because the thickness of the p-type layer that is made on top of the crystal surface varies according to the density of the gas and the heating time when n-type silicon is heated in the boron gas a p-type layer forms on the surface silicon with two layers P and n is then placed into an oxidation furnace and sealed when water vapour is injected the silicon inside the furnace joins with the oxygen in the air and the surface becomes covered with a strong oxide film this method was discovered by carl frosh and lincoln Derek the thin purple layer is the oxide film beneath that is the p-type layer the layer below is an n-type layer a window is opened in the oxide film and an n-type material is spread in again using the gas phase diffusion method the method for opening windows is based on photographic technology and adjacent window for later use is also open when a metallic film is added the two windows become electrodes the unneeded portions are then cut off and a Mesa shape similar to a plateau is completed now using gas control it becomes possible to finally build the p-type layer and thus the transistors characteristics improve dramatically Fairchild achieved rapid growth after the company started mass-producing transistors for military use but an unforeseen development threatened the company's future Mesa transistors that were shipped started to fail in great numbers and it was traced fairly readily to the fact that when we were putting the can on it we had a welder which was splashing some material all inside the can and naturally some of this material in the form of we called it UFO an unidentified flying object would fall on the material and one would fall on the junction on that Junction and make a short Fairchild then came up with the plane our transistor in which the silicon was completely covered with an oxide film and fortified against contamination this is the plane our transistor developed by Jean here knew nothing for truths from the silicon crystal surface except the necessary electrodes making it contamination proof it was the point our technology called selective diffusion that combines oxide film photo etching and the gas phase diffusion of impurities the planar transistor conquered the market bringing Fairchild vast profits Fairchild which had almost been driven to the brink of bankruptcy by contamination that plagued the Mesa transistor made rapid progress towards becoming a giant company to make a plain our transistor an oxide film is placed on top of n-type silicon along with a photosensitive agent a diagram of a window that is to be opened in the oxide film is placed on top and exposed to light a window thus opens in the photosensitive film and the oxide film appears when chemically treated the oxide film melts and washes away and the silicon appears it is then placed in a furnace to diffuse impurities a p-type boron compound circulates inside as for the n-type silicon in the furnace only the inside of the window changes into a p-type the grey n-type layer is gradually transformed into a blue p-type layer another oxide film is added in a photosensitive agent applied on top in order to open a small window a diagram is placed on the top and exposed a small window thus has opened on top of the p-type layer gas phase diffusion is again used to apply an n-type material as for the p-type layer only the inside of the window changes into an n-type everything is completely covered again with an oxide film a photosensitive agent is applied a diagram placed on top and exposed when treated an n-type layer protrude from the middle as does a p-type layer an overall aluminium film is then added along with a photosensitive agent and diagrams of electrodes it is then exposed when treated the unneeded portions rinse away and only the electrode portions remain only the necessary electrodes protrude while everything else is covered with oxide film this technology eventually led to the integrated circuit Jack Kilby is 69 years old in September of 1958 he created an integrated circuit as you looked at it it became apparent that you could make any kind of components that you were needed with one exception of silk and you could certainly make transistors they can make diodes resistors capacitors other circuit element this is mister Kilby's first integrated circuit which has been donated to the Smithsonian Institution this is an integrated circuit that Robert Noyce invented and mass-produced the white lines seen on the surface are wires for connecting parts these are connected to transistors and resistors which were made one layer down and are encased in oxide film thus a logic circuit for computer use was built into small silicon crystals this was the debut of the Orthodox integrated circuit two years before on October 4th 1957 the Soviet Union launched Sputnik the world's first man-made satellite the United States desperately tried to catch up but there was a clear difference in rocket technology the Soviet Union leaving behind it as stunned America then launched a man into outer space cosmonaut Yuri Gagarin orbited the earth I believe that this nation should commit itself to achieving the goal before this decade is out of landing a man on the moon and returning him safely to the earth ten months behind the Soviet Union the United States also placed an astronaut in orbit the rocket couldn't match the Vostok of the Soviet Union challenged by the technology of Sputnik the West aimed at miniaturizing electronic equipment GWA dumar of the British royal radar establishment built small-scale circuits into silicon in 1957 the US Army Signal research and development laboratory and RCA technicians conceived micro modules this was an idea designed to make the overall size of equipment smaller by making the major circuits that comprised the equipment into thin blocks which would then be piled upon each other westinghouse technicians came up with the concept of molecular electronics although there was something resembling a model it was not clear how to make it and with what structure I thought I he thought you Monday in those days we didn't say ICS we call them solid circuits we heard that they had been perfected but the details were very sketchy all we heard was that some circuit had been made inside a solid and almost no other information a report saying that the first IC had been made for the first time appeared in the magazine electronics to tell you the truth when I first read it I didn't understand what they were talking about we were going to treat the matter lightheartedly and overlooked the whole thing but mr. Kikuchi remarked that's really odd could they be talking about some kind of new thing this made us think twice and we started to tackle research on integrated circuits in the electro technical laboratory the two researchers took on the test manufacturing of a solid-state circuit using independent technology this is a technical memo written by Mitsubishi Electric semiconductor division in 1960 it mentions the molecular electronics conceived by Westinghouse Mitsubishi Electric's president who was excited by molecular electronics issued in order for the independent development of the same technology which was named the millech Tron project however at the time Mitsubishi did not have plain art technology about the only thing it had was rudimentary technology for creating resistance in silicon using line etching technicians spread paraffin on the surface of silicon made into two layers PN and then scratched it with a pencil when they later soaked it in etching fluid the portion where the paraffin had come off displayed resistance the resistance value being determined by the length of the scratch in January of 1961 the electro technical lab team of taro II and zdenda completed a solid-state circuit with which they had been struggling they built onto a germanium crystal chip an alloy transistor a resistor and a condenser which were connected with wires and fixed with resin that was the first Japanese integrated circuit following the work at the electro technical lab Mitsubishi Electric unveiled several different tests manufactured electrons we announced eleven kind the silicon was silicon but at the time we couldn't use it for transistor technology therefore in order to do so we opened a hole in the silicon to insert a germanium transistor and buried it in there in other words the integrated circuit hadn't been formed molecularly it was merely pasted in place it was a fake this led to problems for we were asked by people like mr. zdenda how was your electron made we were embarrassed and kept silent rather than reply this is the patent for the planar transistor this was also the basic technology for integrated circuits with this patent as a weapon Robert Noyce of Fairchild asked the Ministry of International Trade and Industry to authorize his production of semiconductors in Japan methi said no on hearing this noise came to my place and angrily said nobody's as thick-headed as those guys at Japan's me t so then he decided to ask Nippon Electric to buy the planner patenting Kato Honda osa funa and noise became friends on meeting at an academic conference mr. Asif una brought noises conditions directly to the company's top officials and they immediately bought the exclusive rights to the plane our patent school we took out an exclusive license by doing that it became impossible for even Fairchild to export to Japan we were resented by all the other manufacturers in particular I think Itachi was the most upset at the time Hitachi was struggling with measures to prevent the degradation of Mesa transistors when broken open it's possible to see these silicon crystals inside the hitachi staff tried to eliminate contamination by treating the surface of the completed transistor crystals later NASA ethanol Oh you'll install a human the technology that we used was to put a thin oxide film such as quartz onto the surface of the Mesa transistor later and in this way we try to protect the surface from water vapor and other impurities so that it might reach it let's get on a little even idea the heat treatment after the transistors were completed had to be done at low temperature organic material of silicon oxide eventually turned into a gas it was then circulated in the furnace and the wafers were placed inside to see what would happen the surface became covered with a silicon oxide film but making a film by dissolving an organic compound at such a low temperature is very different from making a film such as quartz at a high temperature it becomes very rough and full of holes so that water vapor ends up completely permeating everything through those holes since this method wasn't serving its purpose we came to the point where we had to try something else this time they added a metallic film to the top of the silicon oxide and placed it in a furnace to utilize the phenomenon in which two materials occasionally fuse when heated together at low temperature they tried innumerable metal in silicon oxide combinations but they couldn't create the desired metal epoxy no on all candle but a young person who was working on my experiment with me partly out of mischief placed a few lead particles on a quartz plate and I'm above a burner now my you know someone else after this name barn I'm already what's that I when he did that at a very low temperature the lead formed into drops and holes formed on the quartz plate elicited through which the lead trickled out stylist here's a reenactment of the experiment a ball of lead is placed on a small dish of quartz glass when placed over a flame the leg gives off a white smoke and begins to burn before long the quartz glass also begins to burn with the lead and the bottom of the vessel melts like candy and falls out demonstrating that when an orderly oxide film is made at low temperature with metal it changes into a high density film it was factory technicians who first thought of using this method as a means of countering the plane our patent they decided to remove the oxide film and apply a new film to create a transistor based on low temperature surface passivation technology however when it went into mass production unimaginable difficulties awaited the technology which had gone well during the test manufacturing stage you can you tell you in comparison to our problems a stomachache was nothing we couldn't make even one the customers have been given a delivery date and although we said we would supply 100,000 pieces per month we just couldn't do it this problem wasn't with only one company many companies had placed orders and we were being criticized everywhere we had pursued an independent path and had taken orders but if you take orders and can't deliver it's like having the whole world against you nobody supported us but during this process technicians were still reared business requires racking your brains over a problem and then somehow overcoming it when such confidence is applied to the next difficulty the good businessman will definitely see it through I think that the determination and hard work of the people who assumed leadership in the factories in those days is what gave birth to Japan's semiconductor industry in other words they exhibited tenacity in regard to perfecting domestic technology on July 16th 1969 Apollo 11 took off destination the moon four days later the astronauts climbed down and stood on the moon's surface it was a brilliant achievement just eight years after President Kennedy announced the goal of sending man to the moon the greatest achievement was that the astronauts piloted Apollo 11 themselves on this round trip between the earth and the moon okay rocket wink tranquility we can't be on the ground you got a bunch of guys about to turn blue we're breathing again thanks a lot Neil Armstrong reporting there are no difficulty adapting to the one-sixth gravity of the morning what made this all possible was an ultra small guidance computer the system was ordered by NASA from the Massachusetts Institute of Technology in line with President Kennedy's objectives extremely low power consumption extremely high reliability and a system smaller than one cubic foot were the requirements this is Apollo 11s ultra-small guidance computer it is comprised of various modules separated according to function warning flight control sensors and logic the surfaces of all of these are densely filled with ICS there are 60 ICS per side on a module 120 in all a total of 550 transistors and 2826 ICS were used in all of the devices the power consumption was a mere 55 watts in 1960 at the beginning of development one IC cost $1,000 four years later the cost had dropped to $25 and 12 years later for Apollo 17 the price of an IC had become just $1 due to American space development and national defense projects the cost of ICS dropped dramatically while performance improved markedly meanwhile it was Japan that applied these results to civilian products Japan developed consumer goods that incorporated large numbers of semiconductors such as radios televisions and tape recorders then from the late 60s through the 70s the fierce calculator Wars broke out the competition prompted demand for vast quantities of ICS spurring Japan's semiconductor industry to make even greater technological progress you

Channel: RC286

Views: 213,334

Keywords: Japan (Country), Industry (Organization Sector), Electronic, Documentary, History, Experimental, Culture, Technology (Professional Field), Invention (Award Discipline), Integrated Circuit (Invention), Transistor (Invention), Electrical Engineering (Industry), electricity, vintage, Electronic Circuit (Taxonomy Subject)

Id: uGRNXmWng3M

Length: 38min 32sec (2312 seconds)

Published: Thu Nov 14 2013