[April 4, 2021] – Carver Mead was already a legend in VLSI when I started to cover the area of electronic design automation in the 1990s. Together with his Cal Tech student David Johannsen and others Mead formed Silicon Compilers Inc which was in the forefront of VLSI methods that used software-based Hardware Description Language to markedly transform semiconductor, design and production.
I also got to follow Mead’s less known work on neural network hardware – work that was overshadowed during a long entrenchment that saw general-purpose CPUs and then GPUs used for AI. As well, he played a role in the rise of the fabless semiconductor company, which has become a news thing again in financial journals as Intel weighs its future plans.
What I did not know about was his efforts to produce a practical MESFET for communications, as well as groundbreaking work on electron tunneling and hot-electron transport. Did not know, for that matter, his narrative capability as a lecturer on electrical engineering. The latter is what brings me to this blog post today.
Mead in a 2019 lecture discussed lessons from the early days of semiconductors. It is quite a story.
His YouTube presentation is as much about the taming and commercialization of electricity – the development of electrochemistry – as it is about the birth of semiconductors, beginning as it does with Volta’s invention of chunked battery storage stacks in the very late 18th Century, and the subsequent work of Faraday.
Before this, it wasn’t electricity, it was “sparks,” said Mead. The phenomena needed to be understood, not veiled as magic.
The presentation came with a bit of Western twang and bemusement, natural as Mead hailed originally from California’s Central Valley. Understated but more telling for that.
Mead’s technology history is populated with people of high curiosity asking questions of how electric current moves in a variety of circumstances. The story proceeds mostly in terms of resistance, capacitance and conductance in metals and semiconductors.
From the early days, before the discovery of the electron, people had to deal with diverse behavior of electricity as it passed through different metals. The modern story could start with the work of Greenleaf W. Pickard who “came along in 1906” to explore contacts between different kinds of wire and different kinds of minerals.
Pickard’s style of silicon crystal point contact detector found use as a radio receiver on ships at sea under US Navy backing in World War I. Mead has a congenial way of connecting the device to its larger meaning – “A lot of times bad things happen to ships and you can’t see them from anywhere and it will be really, really good to know where they were and what had happened.” Often, it is world affairs that drive innovation, no doubt.
On the way to the detector, Pickard had gone through about 30,000 combinations of minerals, and discovered the usefulness of silicon, although germanium was – much later – to be the basis for the first transistor.
The story from Schottky to Shockley. The evolution of the semiconductor finds electrical engineering, chemistry and physics deeply intertwined. It has been about observing and understanding. Metals and metalloids with distinct patterns of conductivity were of special interest.
The figures that emerge include Schottky, he of the Schottky barrier diode. He was a German physicist and materials researcher, who spent a great portion of his time, including the span of World War II, working at Siemens, where the research for radar, important to the war effort on both sides, involved some innovative circuit development.
“He was a dogged fellow … trying to understand the metal semiconductor ever since [Karl Ferdinand] Braun had shown it,” Mead said. “He kept coming up with theories. And they kept being wrong … shown to be wrong by experiments.”
In 1942, in the midst of war, he published a paper describing the use of selenium as a semiconductor, discovering that the impurities that were in selenium comprised atoms that had an extra positive charge that made for what is now commonly called a p-type semiconductor. With this guidance, engineers could plot out capacitance, and move forward.
“That turned out to be a key thing,” Mead said. People working on radar at Bell Labs realized that this betokened a semiconductor device that would amplify signals. That brings us to Bardeen, Bratton and Shockley, eventual recipients of the 1956 Nobel prize in Physics. They built, observed and understood the workings of semiconductors and discovered the transistor effect.
Mead’s history was completely compelling, so I perused the transcript – sure it held an explorative path for general technology prediction. Alas, the song of the technology cowboy from the Central Valley seemed to boil down to this: ‘Smart guys ask small, pointed questions of troublingly, nuanced phenomena by doing experiments – sometimes thousands.’
This was the answer to my prayers. It wasn’t the answer I wanted, but it was the answer. The path to understanding “magic “to the point where you can do something with it requires work, work, work.
Along the way during this lively lecture I learned a lot – some of that got fleshed out as the quotes below.
*“Any phenomena gets turned into an instrument and that’s a good thing.”
*“The galvanometer became the thing that allowed people to do electricity quantitatively.”
*“Electricity was Sparks since antiquity and sparks were fun, and you could make magic with them and do all kinds of things, but there wasn’t much quantitative you could do.”
*”Our story is about controlled electrical current and that all started when Volta came up with a voltaic cell. Faraday was able to do his work because of the invention of voltaic core.”
*“Before the discovery of the electron, electricity was just a fluid – still a pretty good way of reasoning about it.”
There is a lot more to learn by sitting in on this lecture. What is so prominent you may agree is Mead’s capability as a storyteller – weaving the narration from Franklin, Volta and Faraday to Moore’s law, DRAMs and CMOS. It’s a story of error and trial … one in which “the people that just put their head down to do it beat out all the people that were trying to be fancy.” That’s nitty gritty analysis from a silicon hero, and worth making note of. — Jack Vaughan
Lessons from the Early Days of Semiconductors – Carver Mead – 4/24/2019 – YouTube
Note on the history of the transistor – Wikipedia
Note on history of silicon compilation – HistoryofInformation.com