How a single electron can break a transistor bond at seven electronvolts

A lone electron, provided it carries the right amount of energy, can snap a minute chemical bond that helps keep crucial parts of a computer chip intact.

These results challenge the long-standing assumption that such harm accumulates slowly and steadily over time.

Instead, the work reframes electronic ageing as a targeted quantum incident, clarifying precisely where-and when-breakdown first starts inside today’s devices.

Inside the transistor’s weakest link

At the silicon–oxide interface within a transistor-where the stress of switching is concentrated-a particularly delicate bond is left vulnerable to passing charge.

At the University of California, Santa Barbara (UCSB), Woncheol Lee showed that the bond’s failure can be traced to single electrons arriving within a tightly defined energy band.

Rather than being worn down by many repeated impacts, the bond gives way when one electron briefly occupies a concealed state that makes the bond unstable.

That well-defined trigger helps to account for why performance loss can seem to arrive abruptly, with little or no warning.

The result also strengthens the case for working out why some bonds endure while others fail when conditions change.

Hydrogen remains static

During fabrication, chip manufacturers introduce hydrogen because it terminates incomplete silicon bonds-the tiny links between silicon atoms that underpin the chip’s structure.

This passivation step is essential and must happen before those sites can act as electrical trouble spots.

By keeping silicon–hydrogen bonds capped, unwanted defects are prevented from disrupting charge flow through the transistor.

If hydrogen departs, the now-exposed bond begins to capture charge, gradually pushing the device away from its intended behaviour.

A serious reliability issue can therefore begin with a microscopic chemical “patch” loosening in precisely the wrong place.

Earlier clues lingered

Well before the new evidence was established, engineers had one recurring hint. One report noted that deuterium-hydrogen’s heavier isotope with an extra neutron-often allowed transistors under stress to keep working for longer.

Initial device studies indicated lifetime gains of roughly ten to 50 times after deuterium processing. Because deuterium has more mass, the outcome suggested that the nucleus itself mattered.

What remained unresolved was the mechanism: how the added mass produced the benefit, and which electrons were actually responsible for the damage.

A looming, dominating energy

The hazardous window clustered around seven electronvolts (a standard unit of particle energy), rather than being spread across a wide range.

At that energy, an electron can briefly enter a state that actively drives the bond apart instead of helping to hold it together.

Because that state persists only fleetingly, one suitably timed electron can be more destructive than many of its counterparts.

This, in turn, helps make sense of why bond breaking was seen to spike there, while still occurring at lower levels just beneath the threshold.

Quantum motion breaks bonds

After excitation, hydrogen does not simply shoot away like a tiny ball tracing a neat, classical trajectory. Instead, its motion disperses into a wave packet-a quantum description of where the atom may be as time evolves.

The bond breaks once enough of that spread extends beyond a safe separation, even if the atom is not wholly located there.

With that rule included, the model accounted for previously puzzling measurements from real devices more clearly than before.

Beyond heat-driven damage

For years, hot-carrier degradation-chip wear driven by unusually energetic charges-was treated largely as a complicated heating problem.

Under ordinary thermal damage, higher temperatures should accelerate failure because atoms vibrate more strongly and cross barriers more readily.

In this framework, however, a brief electronic “kick” emerges first and has to be addressed before conventional thermal effects can be applied.

“This process doesn’t fit into the usual picture of heating-induced damage; it’s a short-lived quantum event that we can now model without needing to fit it to an experiment,” said Lee.

An isotope becomes a design tool

When the researchers replaced hydrogen with deuterium (hydrogen containing an extra neutron), the bond-breaking process slowed by about 100 times.

The heavier nucleus alters the quantum dynamics, making it more difficult for the wave packet to spread far enough.

That means isotope choice becomes part of what is practical to design for when reliability testing is the priority.

Small defects, big consequences

The group then carried the idea beyond a single break, examining how it unfolds inside an operating transistor.

Under intense operation, high-energy electrons traverse a thin barrier, expose hidden weak points, and progressively reduce the device’s reliability.

Every severed bond leaves a minute defect, and the accumulation of many such defects can slow the chip’s ability to switch on and off.

Previous accounts struggled to connect these subtle, local changes to the larger-scale failures engineers encounter in real hardware.

A new lens on material failure

Bond rupture triggered by electrons is not exclusive to silicon, which is why the result has significance beyond everyday computer chips.

The paper links the same underlying physics to radiation damage, light-driven chemistry, and hydrogen-related defects in other semiconductors.

Applied outside silicon, the wider framework could help materials scientists identify vulnerable bonds before they turn into costly reliability issues.

That, in turn, could convert a hidden failure mechanism into something engineers can screen for ahead of mass production.

Designing chips around quantum limit

The pieces now align: the seven-electronvolt threshold, the lack of temperature dependence, and the protective effect of deuterium all indicate the same quantum trigger.

Manufacturers still need to verify how broadly the model applies across modern chip stacks, but they now have a clear target for what designs must accommodate.