The Tiny Switch That Almost Made It Big
In January 2015, while digging the foundation of what will one day be MIT’s nanotechnology research facility, construction workers unearthed a mysterious glass tube. It was a time capsule that had been buried in 1957 with instructions that it should remain sealed for 1,000 years. The capsule remained closed, but the MIT Museum’s records revealed its contents: keepsakes, documents detailing the state of MIT in the 1950s, a container of synthetic penicillin, and one tiny switch—the often forgotten genesis of a component that’s crucial to modern research on quantum computing. Unveiled at MIT’s Lincoln Laboratory earlier that year, the switch—better known as a cryotron—was seen as a promising new technology that could be the key to shrinking giant Cold War–era computers to a fraction of their size.
The cryotron was the brainchild of Dudley Allen Buck, SM ’52, ScD ’58. A scoutmaster and Methodist youth counselor, Buck entered MIT in 1950 after spending two years in the Navy’s cryptography unit. He began his MIT tenure as a research assistant on Project Whirlwind, which was developing a massive high-speed digital computer whose design eventually became the backbone of the American air defense system. Whirlwind was huge, both figuratively and literally. Taking up 2,500 square feet of space in MIT’s Barta Building (now Building N42), it was built before the transistor was in wide circulation; it initially performed computations using thousands of vacuum tubes arranged in complicated circuits.
Even before earning his master’s in 1952, Buck began thinking about ways to make smaller, faster logic switches that didn’t use vacuum tubes. He wanted to create circuitry so powerful that, as he later put it, “a large-scale digital computer can be made to occupy one cubic foot.” By December 1953, Buck had a plan for building a better switch. While vacuum tubes rely on heat to get electrons flowing, he envisioned a switch made from materials that can conduct current without electrical resistance when cooled to extremely low temperatures. These superconductors, he theorized, could be as thin as two single wires and would make it possible to build computers that were just as powerful as Whirlwind—and much more energy efficient—without using bulky tubes.
Buck spent the next two years at MIT’s Lincoln Lab building prototypes of his switch. He eventually settled on an easily fabricated design that used two superconductive metals: a straight, one-inch-long wire made of tantalum with a second wire made of niobium coiled around it. At or below 4.2 K—a temperature he achieved by dipping both wires in liquid helium—the tantalum conducted current without resistance. But pumping current through the coiled niobium generated a magnetic field that caused the tantalum wire to assume a normal rather than a superconductive state, so the researchers could turn the switch on or off using only electrical signals. By itself, a single cryotron had limited power, but a collection of switches “so small that 100 will fit into a thimble” could offer the same processing power as vacuum tubes that were orders of magnitude larger. Buck pictured cryotron computers, each with a custom refrigerator called a cryostat that could keep the switches cool enough to work.
Buck filed a patent for his “magnetically controlled gating element” in 1955 and shortly thereafter released information about his work to the public, conceding that the technology was “far slower than vacuum tubes and transistors” and that “the reliability of cryotron circuitry is not known.” Despite these issues, Buck’s research got the attention of both the private sector and the scientific community. Funded by the NSA, he worked on making the cryotron even smaller, experimenting with printing the technology on thin films. Meanwhile, engineers at A. D. Little, RCA, IBM, and GE all launched cryotron research programs.
By 1959, Buck had received a prize from the Institute of Radio Engineers, earned his ScD, and joined both the MIT faculty and the NSA’s Scientific Advisory Board. But in May 1959, less than a month after his 32nd birthday, he died from a sudden respiratory illness. “Dudley was not ambitious for himself, ” recalled one of his students, Charles Crawford ’59, SM ’60, ScD ’62. “He was ambitious for the human race.”
Interest in cryotron technology waned by the mid-1960s, when silicon microchips, which were cheaper and didn’t need refrigeration, became the standard for digital computers’ logic switches. But research on superconductive materials that started with Buck continues today. The Josephson junction, a modified version of the cryotron introduced in the 1960s, remains a mainstay of quantum computing research.
The inside story of how ChatGPT was built from the people who made it
Exclusive conversations that take us behind the scenes of a cultural phenomenon.
Sam Altman invested $180 million into a company trying to delay death
Can anti-aging breakthroughs add 10 healthy years to the human life span? The CEO of OpenAI is paying to find out.
ChatGPT is about to revolutionize the economy. We need to decide what that looks like.
New large language models will transform many jobs. Whether they will lead to widespread prosperity or not is up to us.
GPT-4 is bigger and better than ChatGPT—but OpenAI won’t say why
We got a first look at the much-anticipated big new language model from OpenAI. But this time how it works is even more deeply under wraps.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.