Crystal Fire #10

Despite his immense contributions, Shockley never became the millionaire he wanted to be. He recruited first-rate scientists and engineers, but many defected to start or join other more successful firms.

Neither Brattain nor Bardeen had anywhere near Shockley’s visionary appreciation of the transistor’s vast commercial potential. Both continued doing basic research – Brattain at Bell Labs and Bardeen on a variety of solid state physics topics, especially superconductivity, for which he won a second Nobel Prize.

Almost as important as the transistor’s invention are the techniques crystal growing and zone refining, which allow fabricating large single crystals of ultra-pure silicon and germanium. Without these crystals, the industry would not exist.

The transistor led to a startling transformation of technology and even culture and the nature of work  – computers (main-frame and then personal), modern televisions, the iPod, and cell phones.

The new Information Age comes with its own distinct challenges to human freedom and livelihood. The crystal fire has brought with it an intensity and immediacy of life in which many things become obsolete soon. Some people unable or unwilling to deal with the unceasing change widens the divisions between different peoples on a national and global scale. For as fire illuminates, it also consumes.

Some other challenges not mentioned in the book are privacy, cyber-crime on a global scale, and the proliferation of advertising.

This is my final post about Crystal Fire.

Crystal Fire #9

The use of transistors expanded rapidly. But a worrisome cloud loomed on the horizon. As the number of components – transistors, diodes, resistors, capacitors, and connections – grew, the tedious task of assembly resulted in more defects. A few visionary engineers began looking for ways to fashion circuits from a single block of material. Jack Kilby of Texas Instruments was a pioneer in making such “integrated circuits.” Kilby’s first successful attempt used germanium, but then silicon came to be recognized and used by Kilby and others -- its advantages outweighing its disadvantages. Photolithography was used to define the circuit elements. The monolithic idea also arose in a different form at Fairchild Semiconductor. Both companies got patents for their products.

Shockley Transistor also invented a diode. When manufactured in quantity, too many diodes being defective led to market failure. Another Shockley attempt at transistors didn’t fare well either. Observing Shockley Transistor’s hemorrhaging cash and Fairchild’s great success, Arnold Beckman finally sought to sell off his transistor division and did.

These miniature circuits were a new and revolutionary advance. They are the ultimate practical expression of the theoretical insights of Bardeen, Shockley, Noyes and others, filtered and amplified by hundreds of scientists following in their footsteps. Many of the insights came at Bell Labs, Bell Labs and Western Electric were slow to to appreciate the value of monolithic integration. The phone company did not have the same pressing need for miniaturization that prodded the computer and military markets. Other firms, especially Texas Instruments and Fairchild, led the way.

Crystal Fire #8

William Shockley was surpassed by others for promotion to head all research, so he left Bell Labs. Bell’s top managers felt he was most effective where he was. Other top physicists such as John Bardeen had complained of Shockley’s ham-handed management. Shockley lacked the broader organization skills for directing a wider variety of work. He sought positions at other businesses and as a university professor, but then decided to devote his efforts to start his own company in the semiconductor industry. He finally met with someone who was willing to back him for the $1 million he sought. This was Arnold Beckman, both a good scientist and a successful business man. He was the head of Beckman Instruments, a company that specialized in making analytical instruments, such as a pH meter, for controlling production processes.

Shockley flew Los Angeles and met with Beckman for a week to discuss and then form a business plan. Beckman wanted the new company to be in the LA area, but Shockley wanted it near San Francisco and Stanford University. The provost and dean of engineering at Stanford helped Shockley convince Beckman that being near Stanford would be the best place for recruiting employees and having contacts for getting business.

Shockley first tried to recruit people from Bell Labs, but wasn’t successful. He then sought recruits from other firms such as Motorola, Philco, Raytheon, and Sylvania. He also sought young PhDs at top schools like Berkeley, Cal Tech, and MIT. When Beckman announced the launch of Shockley Semiconductor Laboratory, there were only four Ph.D. scientists and engineers on board. As the facilities were being made for the laboratory, he recruited Robert Noyce and Gordon Moore. Noyce and Moore would later became the co-founder of Fairchild Semiconductor and Intel.

Beckman paid $25,000 to Western Electric to license patent rights in the transistor. Shockley’s contacts Bell Labs were helpful in gaining more info about production techniques. Still the going was tough. It was one thing to get methods working in the ultrahigh-technology environment of Bell Labs with its ample supply of first-rate scientists, engineers, and technicians and high-quality equipment. It was quite another to achieve the same results in the primitive surrounding of Shockley’s lab, even with the talent there.

In November, 1956 Shockley learned he was to be awarded the Nobel Prize in physics along with Bardeen and Brattain. They all met in Stockholm to receive their award in December 1956. As 1957 began, his company was not doing well. It had been in operation more than a year, but was still struggling to produce anything for sale. Employee defections began. Beckman began to realize that Shockley was a brilliant physicist but a lousy business manager. Shockley devoted his efforts to developing one kind of transistor while Noyce, Moore, and others thought it was a waste of time. Several months later Noyce, Moore and six other of Shockley’s brightest recruits resigned to start their own company. They got financing from Fairchild Camera and Instruments and named their company Fairchild Semiconductor.

Crystal Fire #7

When Bell Labs hosted a symposium in 1951 about the transistors invented at Bell, absent was mention about the technologies involved in fabricating the gadgets. Bell’s subsidiary Western Electric began licensing rights to transistors for a fee. The art of fabricating them could not remain secret much longer. It was a challenge because the conflicting demands by the military for secrecy yet multiple suppliers and commercial openness were difficult to meet simultaneously. Anyway, Bell Labs invited licensees to visit its Western Electric plant in 1952, then published a comprehensive description of the state of the art of manufacturing transistors. One of the technologies then revealed was a powerful new method of purifying germanium, which Bell Labs had kept under wraps for almost two years at the military’s request.

Military applications provided an immediate market for transistors where cost was not a concern. Transistors were much more expensive than vacuum tubes. From 1953-55 almost half the revenue for transistors came from the military.

Other demand came from hearing aid manufacturers. Keeping with Alexander Graham bell’s devotion to the deaf and hard of hearing, AT&T extended royalty-free licenses to hearing aid manufacturers. Transistors were especially helpful for hearing aids. Ones that relied on vacuum tubes were bulky with a battery-powered amplifying unit worn around the waist. Also, the batteries were expensive.

Other demand came for use in pocket-size transistor radios. Texas Instruments was a leader in those, licensing the technology to other manufacturers in exchange for royalties. A Japanese company formed a subsidiary called Sony that became very successful making such radios.

Later in the 1950’s new kinds of transistors were invented, including using silicon instead of germanium.

Crystal Fire #6

After World War II the people at Bell Labs could resume its focus on basic research and development. The middle chapters of Crystal Fire give many details about this period through the early 1950s. They learned many things about the electrical property of different elements, compounds, combinations, of solid state devices. They also encountered mysteries about these things that prompted further research. They also had to spend considerable time obtaining samples of materials absent undesired impurities or optimal amounts of desired impurities. Patents were filed and there were concerns about patents already in place, what exclusive rights the military might want, and what progress was being in other labs such as at Purdue University.

There were two types of transistors invented at Bell Labs during those years – the point contact invented by John Bardeen and Walter Brattain and the junction type spearheaded by William Shockley.

In June 1951 Bell Labs gave a press conference about transistors. The star of the show was the junction transistor. Perhaps its most remarkable feature was its extremely low power consumption, about one-millionth of conventional vacuum tubes. It also amplified signals with far greater efficiency than the point-contact transistor.

The point-contact transistor was still used in the Bell System. Practical use of it came earlier. It entered mass production at Western Electric (a subsidiary of AT&T), servicing complex switching equipment to permit direct dialing and bypassing traditional telephone operators.

But the point-contact transistor never made it to the commercial marketplace in a big way. Apart from transient usage in hearing aids and military equipment, the only important applications it ever found came in the Bell System. Other manufacturers were reluctant to put significant capital into its production, especially after the recent breakthrough by Shockley’s team. The future belonged to the junction transistor and its offspring.

Crystal Fire #5

Walter Brattain began working at Bell Labs in 1929. Mervin Kelly was a researcher at Bell Labs who became its director of research in 1936. He envisioned telephone switches being electronic rather than mechanical. William Shockley was Kelly’s first hire after a hiring freeze was lifted in 1936. Kelly hired Shockley for the latter’s knowledge of quantum mechanics. Shockley took Kelly’s vision as a guiding light. For the next several years, Bell Labs grew its research staff with specialists in quantum mechanics. In 1937 Clinton Davisson of Bell Labs won the Nobel Prize for his experiments that electrons behaved like waves.

Brattain and Joseph Becker studied the papers of Walter Shottky and Nevill Mott (neither at Bell Labs). The papers said that whenever a metal and a semiconductor come into contact, a double layer of charge crops up – positive on one side and negative on the other – because of the difference in work functions of the two materials. This leads to a kind of “hill” that electrons must surmount if they are to cruise from one side to the other. “Because this hill is asymmetric, with a steep cliff on the metal side and a shallow slope on the other, electrons move far more readily from semiconductor to metal than in the opposite direction." This finally provided a satisfactory explanation of rectification, which had mystified scientists for 65 years. Shockley, Brattain, and Becker attempted fabricating devices to make use of this effect. Not only was progress slow, it was interrupted with work on topics like radar, submarine detection, mines, and torpedoes due to World War II.

Crystal Fire #4

Bell Telephone Labs grew around the efforts of American Telephone & Telegraph to develop vacuum tubes for long distance communication. Through a variety of gimmicks, the Bell System enabled transmitting telephone calls over 1000 miles by 1900. Beyond that distance, transmission deteriorated. For transcontinental service to become a reality, the company needed a “repeater” device to replenish the electrical signals at points along the line. AT&T installed a transcontinental line by 1915. Repeaters in Salt Lake City, Omaha, and Pittsburgh boosted the electrical signals.

In the 1920s and 1930s the new discipline of quantum mechanics was used to discover the movement of electrons and the properties of conductors, insulators and semiconductors. New theories superseded older ones. Walter Brattain began work at Bell Labs in 1929. He and colleague Joseph Becker became interested in copper-oxide rectifiers. Copper-oxide belonged to a new class of substances called “semiconductors.” They had some unique properties different from both conductors and insulators. Unlike metals, the conductivity increases with increasing temperature. Copper-oxide rectifiers, dubbed varistors”, began to replace bulky vacuum diodes throughout the Bell system. The speed at which this happened was restrained at Bell Labs by a lack of physicists with enough understanding of quantum mechanics and a hiring freeze during the Great Depression.

Crystal Fire #3

Americans were latecomers to the quantum revolution. Europeans led the way in forming deeper and more fundamental insights about the quantum world. It was in the application of the new theoretical insights and experimental techniques that U.S. scientists excelled in the 1920s. They explored the internal structure and intrinsic properties of solids: color, hardness, conductivity, and so forth.

In the last quarter of the 19^th century, physicists became fascinated with cathode-ray tubes. In 1895 Roentgen discovered X-rays – so-called because he didn’t know what they were. X-rays excited the imagination, but much mystery about them remained. Were they waves or particles? Physicists searched for zebra-striped interference patterns like visible light produced. But if X-rays had wavelengths a thousand times shorter than visible light, any suitable diffraction grating would need spacing a thousand times smaller, less than a billionth of a meter. Nobody knew how to make such a fine grating. In 1912 two of Roentgen’s students found a wreath-like pattern of brights spots against a dark background directing X-ray beams onto a crystal of copper sulfate. Other crystals produced similar patterns. Max von Laue thought the interference patterns were caused by a three-dimensional lattice of objects within the crystals. William Henry Bragg and his son extended von Laue’s insights. All three received a Nobel Prize for their work.

Scientists could then peer inside crystals, where the atoms appeared to be arranged in layers, like eggs stacked in crates.

Physicists began to recognize the conductivity of metals had something to do with the availability of electrons within them. An excellent conductor like copper has plenty of free electrons. Good insulators, like glass or wood, have essentially none.

Chapter 3 has more about other discoveries in physics, such as Rutherford’s discoveries of the inner structure of atoms and Schröedinger’s wave equation. I will skip the details to stick to the main story of the book and avoid possible errors trying to summarize it.

Practical-minded Americans like Walter Brattain had little interest in the airy philosophical debates such as occurred between Niels Bohr and Albert Einstein. They were busy applying the new quantum tools to their study of matter. Plenty of unsolved problems about the nature and structure of atoms, molecules, metals, and crystals awaited exploration with the new methods.

Crystal Fire #2

A long strand of aluminum or copper wire tacked to the roof of a house or strung between trees served as an antenna to capture radio signals. Electrons in the wire oscillated back and forth as these waves passed, like corks bobbing on water. Another wire coiled around a cylinder provided a tuning device to select the specific radio frequency transmitted by a station and to eliminate unwanted signals. And a pair of earphones translated the tiny pulses of electric current back into the words or sounds that had been spoken or played into a microphone at the station.

A crystal detector converted the back-and-forth alternating current in the antenna and tuning circuit into one-way bursts of direct current required by the earphones. Exactly how crystals worked had been a mystery until the 1920s, despite their having been used. In 1874 Ferdinand Braun discovered that currents in crystals flowed more readily in one direction, and with a sharp wire tip pressed into a crystal face flowed in a single direction. This is called rectification.

Hearing about Marconi’s difficulties in sending signals long distances relying on a spark between two electrodes starting in the mid-1890s, Braun developed a new kind of sparkless transmitter, which eventually allowed transmitting voice and music, not mere Morse code.

Chapter 2 of Crystal Fire also gives personality and biographical sketches of Brattain, Bardeen, and Shockley.

Crystal Fire #1

Crystal Fire is a book about the invention of transistors, microchips, integrated circuits, and the birth of the information age.

On December 23, 1947 William Shockley arrived at his workplace, Bell Labs, eager for some news. Shockley was head of the solid-state physics group, and two people in his group had made an exciting discovery. The two were John Bardeen and Walter Brattain. Using little more than a slab of germanium, a thin plastic wedge, and a strip of gold foil, they had recently boosted an electrical signal almost a hundredfold. It was dubbed “transistor.” It was an archetypal moment, akin to 70 years earlier when Alexander Graham Bell said, “Mr. Watson, come here. I want you.”

Shockley had been seeking ways to fashion a solid-state device to replace the bulky, unreliable switches and amplifiers commonly used in phone equipment. By January, 1948 Shockley had figured out the important details of his own “junction” design – different from Bardeen’s and Brattain’s “point contact” design – which he believed would be more reliable and easier to mass-produce.

It took a couple more years to perfect the techniques to grow germanium crystals. It took a few more years to figure out how to improve and mass-produce the “junction” design to replace the “point contact” design well underway in manufacturing.

Bardeen departed Bell Labs in 1951. Shockley departed in 1955 for Silicon Valley to start a transistor-making company. They and Brattain met again in Stockholm in 1956 to receive the Nobel Prize for inventing the transistor.

By the mid-1950’s physicists and engineers began recognizing the transistor’s significance. A small innovative company, Texas Instruments, began producing small, portable transistor radios. A little-known Japanese company named Sony soon surpassed Texas Instruments. By 1961 transistors were the basis of a billion-dollar semiconductor industry. The majority of transistors in that era were used in radar and guided missile systems.

Another more technical source about the history of transistors is here.

The Innovators #3

Who invented the computer? With the criteria electronic, general purpose, and programmable (by plugging and unplugging cables), Isaacson's answer is ENIAC. It was completed in 1945 before transistors and microchips came into use. It was designed by two men, Presper Eckert and John Mauchly. "Mauchly and Eckert should be at the top of the list for inventing the computer, not because the ideas were all their own but because they had the ability to draw ideas from multiple sources, add their own innovations, execute their vision by building a competent team, and have the most influence on the course of subsequent development" (80-84). So ENIAC's creation supports his theme of collaboration.

As the microchip was being invented, different developers filed for patents for their invention. Getting a patent often took years. Jack Kilby's application was filed in January, 1959 but not granted until June, 1964.  Fairchild filed an application for Robert Noyce's invention in July 1959. But it was granted earlier, in April, 1961. "So who invented the microchip? As with the question of who invented the computer, the answer cannot be settled simply by reference to legal rulings. The nearly simultaneous advances made by Kilby and Noyce showed that the atmosphere of the time was primed for such an invention. Indeed, many others around the world ... had earlier proposed the possibility of an integrated circuit. What Noyce and Kilby did, in collaboration with teams at their companies, was figure out practical methods to produce such a device" (The Innovators, 179-80).

The Innovators #2

A top locus of collaborative invention was Bell Labs, especially in the 1940's. Its Wikipedia page lists by decades the many discoveries and developments there. Isaacson's The Innovators says nothing about many of them, but devotes many pages to some.

"Bell Labs ... was a haven for turning ideas into inventions. Abstract theories intersected with practical problems there, and in the corridors and cafeteria eccentric theorists mingled with hands-on engineers, gnarly mechanics, and businesslike problem-solvers, encouraging the cross-fertilization of theory with engineering. This made Bell Labs an archetype of one of the most important innovations of digital-age innovation" (48).

There Claude Shannon saw up close the wonderful power of the phone system's circuits, which used electrical switches to route calls and balance loads. In his mind, he began connecting the workings of these circuits to another subject he found fascinating, the system of logic formulated by George Boole. Boole revolutionized logic by expressing logical statement using symbols and equations.  Shannon figured out that electrical circuits could execute Boolean logical operations using an arrangement of on-off switches, making relays and logic gates (48).

Another milestone at Bell Labs was the invention of the transistor (Chapter 4). John Bardeen, Walter Brattain, and William Shockley were later jointly awarded the Nobel Prize in Physics for their achievement. The transistor provided the foundation for transistor radios, missile guidance systems and radar, and the invention of microprocessors, which came to be often called "integrated circuits" or "microchips." Microchips later became foundational for hand-held calculators, computers, and cell phones.

The Innovators #1

I have been reading The Innovators, a book written by Walter Isaacson. It is about the digital revolution, i.e. computers. Isaacson emphasizes that many innovations in the digital revolution were the result of collaborative efforts.

"This is the story of these pioneers, hackers, inventors and entrepreneurs -- who they were, how their minds worked, and what made them so creative. It's also a narrative of how they collaborated and why their ability to work together made them even more creative.
     The tale of their teamwork is important because we don't often focus on how central that skill is to innovation. There are thousands of books celebrating people we biographers portray, or mythologize, as lone inventors. ... But we have far fewer tales of collaborative creativity, which is actually more important in understanding how today's technology revolution was fashioned" (p. 1).

The names of some of the most famous collaborators are common knowledge, e.g. Bill Gates and Paul Allen, Steve Jobs and Steve Wozniak.  But there are several other collaborators, less commonly known, who also did a lot to make the digital revolution.  John Bardeen, Walter Brattain, and William Shockley at Bell Labs were awarded a Nobel Prize for inventing the transistor. The graphical user interface and mouse that Steve Jobs first exploited for Apple computers were invented by teams elsewhere. Robert Noyce and Jack Kilby invented the the first integrated circuit or microchip that helped launch the personal computer revolution. Noyce and Gordon Moore founded Intel, which mass-produced and improved the microchips that fueled the personal computer revolution. These collaborators plus more not mentioned here provide the content and evidence for Isaacson's story. He also gives examples of inventors who did great things, but with little collaboration, and that inhibited the wide-spread adoption of their ideas and the success of their ideas in the marketplace.

Transistors and the history of hearing aids

A book I am reading about the invention of transistors -- Crystal Fire -- includes the following about hearing aids. As a preface, Alexander Graham Bell was the founder of the Bell Telephone Company, and AT&T was initially a subsidiary of Bell Telephone (link).

In keeping with Alexander Graham Bell's devotion to helping the deaf and hearing-impaired, AT&T and its Bell Labs, where the transistor was invented, extended royalty-free licenses to hearing aid manufacturers. In 1952 Sonotone began selling hearing aids in which one of the three vacuum tubes was replaced by a transistor. A few days later the Maico Company came out with a model with three transistors and no vacuum tubes. A few months later Acousticon came out with a model with only one transistor and no vacuum tubes.

The principal limitation on vacuum tube hearing aids was the expense and encumbrance of the batteries needed to power the amplifying unit, generally worn around the waist. With the progress of miniaturization, solid-state circuits eventually allowed making hearing aids that could be worn entirely within the ear (p. 205).

The military was by far the biggest customer for transistors in the early 1950's. Their military use was in radar and guided missiles. Hearing aid manufacturers were probably the second largest customers. They were soon surpassed by the makers of transistor radios, which were introduced in October, 1954 by Texas Instruments, partnered with another company (Regency TR-1 radio).

For a little more on the history of hearing aids, Wikipedia has an article about it. Of course, hearing aids have become smaller along with microprocessors becoming smaller in the last 60 or so years