Brother, Can You Spare An Electron?

The move into the quantum age of design will force designers to rethink how they deal with subatomic particles.

popularity

Every now and then in semiconductor design we come to a crossroads where we have to start thinking about problems differently. At 10nm we will be forced to do that again.

The problem—or opportunity, depending on your vantage point—this time involves electrons. While they’ve become increasingly difficult to manage in ever-thinner wires, where RC delay is producing unwanted heat, and in planar transistors, which is why transistors are now sprouting fins to control leakage, there is work under way to actually use those electrons for a higher purpose.

One of the developments necessary to make this happen is a better understanding of the electrons themselves. We’ve been able to photograph atoms using X-ray photography for the better part of the past decade. Researchers can now “sense” the electrons, too. That’s a critical step in managing them and in manipulating them.

Electrons are at the epicenter of spintronics research, which capitalize on the normal spin of electrons, and tunnel FETs, where electrons pass back and forth through a thin ferromagnetic barrier. But rather than viewing this as just one more headache, researchers are focusing on understanding how to harness them more effectively. At 10nm and beyond, when quantum effects become a regular consideration in designs, electrons need to be thought about differently. While most designs are being done at 40nm or higher these days, there are plenty of companies working on 14nm, too. And if they’re working on 14nm designs, they also need to be looking at 10nm, which is the next stop on the shrinking features roadmap.

10nm is the beginning of the shift toward quantum design, where things like memory no longer behave the same way because electrons don’t discharge evenly. At 8/7/6nm, these effects become a central part of designs, and understanding the behavior and uses of electrons moves well beyond the realm of the electrical engineering textbook and crosses into what used to be theoretical physics. It’s also the point where theoretical physics becomes practical physics.

While it’s a bit odd to think of electrons as your friends, they are at least the next frontier in future designs. If they can be manipulated, they can be useful in ways we haven’t considered in the past. And that’s exactly where current research is headed. How much of this makes its way into commercially viable chips isn’t known yet. But what is known is that we’ve only scratched the surface. Fasten your seat belts. We’re going in.

—Ed Sperling