With his younger brother, Herb, a mechanical engineer, he established a small company called General Automation to research and develop energy and information technologies. Together they built, in 1959, a mechanical model of a nerve cell, a semiconductor switch they called the Ovitron, in the process pioneering the use of nanostructures. A year later, Stan and Iris opened the Energy Conversion Laboratory.
The Ovitron itself had no practical application, but in developing it, Ovshinsky made a breakthrough that would define his career, and make the “Ovshinsky effect” a science textbook phrase. He discovered that certain types of glassy thin films, known as amorphous or disordered materials, turn into semiconductors upon application of a low voltage. Semiconductors, the foundation of modern electronics, are materials that conduct an electrical charge but can be regulated, unlike common conductors such as copper.
At the time, in the early 1960s, scientists believed that semiconductors could be formed only from crystalline materials, such as purified silicon, in which all the atoms are arranged in a long-range order. Ovshinsky demonstrated that it was possible to form semiconductors from amorphous or disordered materials, like common glass or silicon alloyed with less-costly elements. Amorphous silicon made possible the production of devices that are now inexpensive and ubiquitous in computing and energy applications.
“He invented the field of disordered materials,” says Hellmut Fritzsche, former chairman of the physics department at the University of Chicago. “It was so revolutionary at the time that people at Bell Labs and other major research labs said, ‘This man is crazy.’ Stan’s contribution was to say that [crystalline material] is not necessary, and it is too restrictive. You can make semiconductor materials in many ways when they are not crystalline, when they are disordered. Then you have a great freedom to alter their properties by chemical modification.”
Soon a phalanx of physicists, chemists, and engineers were making a pilgrimage to the Ovshinskys’ modest Detroit lab, including a young Robert Noyce and Gordon Moore, who were then planning a company to produce computer memory products, the future Intel Corporation. Many who came to scrutinize Ovshinsky’s work stayed on to collaborate, captivated as much by Stan and Iris’s lively warmth as by the novelty of the science. Fritzsche and others who have known him over the years say he always exuded a remarkable confidence in his own abilities.
Ovshinsky made his major discovery while trying to develop an artificial neuron as the first step toward developing a cognitive computer, a working model of the cerebral cortex that he still dreams of completing one day. But he soon put the discovery to work. In September 1966, he filed the first patent on phase change technology, which enabled a new type of computer memory. The most common type of computer memory is dynamic random access memory, or DRAM, which replaced the magnetic core memory of the earliest digital computers. But DRAM chips lose their data when the power is switched off. Phase change memory, which Ovshinsky called ovonic unified memory, registers data by changing the physical characteristics of the semiconductor material, from amorphous to crystalline and back again, and that change remains in effect even without electrical current. When a cell phone user’s battery dies, but the phone retains her contact list, she has Ovshinsky’s invention to thank. The same basic technology underlies rewriteable optical discs, enabling consumers to download music onto CDs.
Driven by the joy of discovery and their stated intention to use science and technology to solve serious global and societal problems, Stan and Iris kept the company small and nimble by licensing their technologies to major manufacturers. Profits were poured back into research, and growth came almost despite the founders’ intentions. Energy Conversion Laboratory licensed its phase change technology to Intel and STMicroelectronics NV, both of which continue to develop and improve such chips.