100th anniversary The invention of the transistor will happen in 2047. What will transistors look like then? Will they even be the critical computing element that they are today? IEEE Spectrum Ask experts from around the world about their predictions.
What will transistors look like in 2047?
One expert predicted that transistors would be more versatile than they are now. Just as processors have evolved from CPUs to include GPUs, network processors, AI accelerators, and other specialized computing chips, transistors will evolve to suit a variety of purposes. “Hardware technology will become application domain-specific in the same way computing architecture has become application domain-specific,” he says H. -S. Philip Wongan IEEE Fellow, Professor of Electrical Engineering at Stanford University, and a former Vice President of the Corporate Research at TSMC.
Despite the variety, the basic operating principle — the field effect that turns transistors on and off — will likely remain the same, he suggests. Suman Dataan IEEE Fellow, Professor of Electricity and Computers at Georgia Tech, and Director of the Multi-University Nanotechnology Research Center rise of. Such a device would likely have minimum critical dimensions of 1 nanometer or less, allowing for a device density of 10 trillion per square centimeter, he says. Tzu Jae King LiuIEEE Fellow, Dean of the College of Engineering at the University of California, Berkeley, and a member of the University of California, Berkeley Intel Board of Directors.
“It is safe to assume that the transistor or switch designs of 2047 have already been demonstrated on a lab scale.” – Sri Samavidam
Experts seem to agree that the 2047 transistor will need new materials and possibly a stacked or 3D structure, with expansion of the planned complementary field effect transistor (CFET, or 3D stacked CMOS). [For more on the CFET, see “Taking Moore’s Law to New Heights.”] Data says the transistor’s channel, which now runs parallel to the silicon plane, may need to become perpendicular in order to continue increasing in density.
AMD Senior Fellow Richard Schultz, suggests that the main goal of developing these new devices will be energy. “The focus will be on reducing energy and the need for advanced cooling solutions,” he says. “A lot of focus is required on devices that operate at lower voltages.”
Will transistors still be the heart of most computing in 25 years?
It’s hard to imagine a world where computing isn’t done with transistors, but, of course, vacuum tubes were once the digital choice of choice. funding a quantum computing startup, which doesn’t directly rely on transistors, Reached $1.4 billion in 2021According to McKinsey & Co.
Experts in electronic devices say that advances in quantum computing won’t happen fast enough to challenge the transistor by 2047. “Transistors will remain the most important computing component,” he says. Saladin sword, an IEEE Fellow and Professor of Electrical Engineering and Computer Science at the University of California, Berkeley. “Currently, even with a perfect quantum computer, the potential areas of application appear to be somewhat limited compared to classical computers.”
Sri Samavidam, senior vice president of CMOS technologies at the European Chip Research and Development Center Imec, agrees. “Transistors will continue to be very important computing elements for the majority of general-purpose computing applications,” Samavedam says. “One cannot ignore the efficiencies achieved from decades of continuous improvement of transistors.”
Was the 2047 transistor actually invented?
Twenty-five years is a long time, but in the world of semiconductor research and development, it is not too long. In this industry, it usually takes about 20 years [demonstrating a concept] to an introduction to manufacturing,” says Samavedam. “It is safe to assume that the transistor or switch architecture of 2047 has already been demonstrated on a lab scale” even if the materials involved are not an exact match. King Liu, who demonstrated the modern FinFET program about 20 years ago, agrees. 25 years old with his colleagues at Berkeley.
But the idea that the transistor of 2047 is already in a lab somewhere isn’t universally shared. Saladin, for example, is not believed to have been invented yet. “But just like FinFET in the 1990s, it’s possible to make a reasonable prediction of the geometry” of future transistors, he says.
You can catch a glimpse of this architecture in proposed 3D-stacked devices made of 2D semiconductors or carbon semiconductors, AMD’s Schultz says. “Device materials that have not yet been invented could also be in range in this time frame,” he adds.
Will silicon still be the active part of most transistors in 2047?
Experts say that the heart of most devices, the channel area of the transistor, will still be silicon, or perhaps silicon-germanium – which is already beginning to advance – or germanium. But in the year 2047, many chips may use semiconductors that are exotic today. These can include semiconductor oxides such as indium gallium zinc oxide; 2D semiconductors, such as metal disulfide chalcogenide tungsten disulfide; and one-dimensional semiconductors, such as carbon nanotubes. Or even “others yet to be invented,” says Imec’s Samavedam.
“Transistors will remain the most important component of computing.” – Saif Salahuddin
Silicon-based chips may be combined in the same package with chips based on newer materials, just as today’s processor makers combine chips using different silicon manufacturing technologies in the same package, notes IEEE Fellow Gabriel Lu, a senior fellow at AMD.
The semiconductor material at the heart of the device may not be the central issue in 2047. “The choice of channel material will be primarily determined by the material that is most compatible with the many other materials that make up other parts of the device,” says Salaheddin. And we know a lot about combining materials with silicon.
In 2047, where will transistors be common where they are not today?
Everywhere. No seriously. Experts really predict that some measure of intelligence and sensing will creep into every aspect of our lives. This means that the devices will be attached to our bodies and implanted inside them; embedded in all types of infrastructure, including roads, walls, and homes; woven into our clothes. hold on to our food swaying in the breeze in the grain fields; Watching virtually every step of every supply chain; And do many other things in places no one has thought of yet.
Transistors will be “everywhere that needs computation, command and control, communications, data collection, storage and analysis, intelligence, sensing and actuation, interaction with humans, or a gateway into the world of virtual and mixed reality,” sums up Stanford Wong.
This article appears in the December 2022 print issue as “Transistor 2047”.