Victor Lopez-Richard and his team at the Federal University of São Carlos, Brazil, have just shattered one of computing's most stubborn myths: that transistors have hit an immovable thermodynamic wall. Working in collaboration with researchers at the University of Würzburg and the University of Richmond, they've developed a theoretical framework showing that "memtransistors"—semiconductor devices with built-in memory—can actually bypass the Boltzmann limit, a fundamental constraint on how efficiently transistors can switch electrical currents on and off.
This matters because transistor efficiency is the backbone of modern computing. Since the 1940s, transistors have gotten smaller, faster, and more numerous—so much so that computing power has roughly doubled every two years, a trend known as Moore's law. But that exponential progress has been slowing, and the culprit is the Boltzmann limit: a thermodynamic law that sets a minimum energy cost for switching any electronic device. For decades, engineers assumed this barrier was insurmountable, and many have proposed scrapping the basic transistor design entirely in favor of exotic alternatives like ferroelectric materials or quantum tunneling mechanisms.
Lopez-Richard's insight was simpler and more elegant. He noticed that previous theoretical descriptions of transistor behavior hadn't captured something important: the intrinsic memory effects that already exist in many nanoscale devices. When electrons flow through certain semiconductor materials, they sometimes get trapped in the crystal structure before being released later—a phenomenon that creates a natural "memory" of whether the transistor was recently switched on or off. This memory effect, once thought irrelevant, turns out to be the key.
By unifying quantum transport theory—which describes how electrons move through nanoscale structures—with the physics of charge trapping, the researchers created a new analytical framework for understanding these memory-preserving transistors. The results, published in Physical Review Applied, demonstrate that memory dynamics can naturally improve a transistor's switching efficiency beyond the conventional Boltzmann limit. "Unlike previous approaches, our model explains sub-thermal switching without relying on ferroelectric materials or tunneling mechanisms," Lopez-Richard explains. More importantly, the team's theoretical work provides practical design rules for optimizing memtransistors in real devices.
What makes this discovery genuinely exciting is its accessibility. Rather than requiring a complete reimagining of transistor technology, the team's findings suggest that the Boltzmann barrier could be overcome using principles already present in existing nanoscale devices. This could accelerate the development of energy-efficient computing architectures, particularly neuromorphic and in-memory computing systems—technologies that mimic the brain's efficiency and promise dramatic reductions in power consumption.
For an industry that has long feared the end of Moore's law and the limits of classical transistor design, Lopez-Richard's work offers a refreshing perspective: sometimes the breakthrough you need isn't a revolutionary new technology, but a deeper understanding of the elegant physics already at play. The path forward isn't necessarily exotic—it's just been hiding in plain sight.