The researchers, led by Jason Campbell, came across the problem while studying the fluctuations between on-off states in progressively smaller transistors.

They claim that a widely accepted model explaining errors caused by electronic 'noise' in switches does not fit the facts. For decades, they say, the engineering community has largely accepted a theoretical model that identifies these defects and helps guide efforts to mitigate them.

The theory — or the elastic tunnelling model — predicts that as transistors shrink the noise frequency should get higher.

But Campbell and his colleagues at NIST, together with scientists from the University of Maryland College Park and Rutgers University, have now shown that even in nanometer sized transistors, the noise frequency remains the same.

"This implies that the theory explaining the effect must be wrong," Campbell said. "The model was a good working theory when transistors were large, but our observations clearly indicate that it's incorrect at the smaller nanoscale regimes where industry is headed," he added.

The problem has particular implications for low-power transistors, the researchers stress, as the fluctuations they have observed grow even more pronounced as the power decreased.

"This is a real bottleneck in our development of transistors for low-power applications. We have to understand the problem before we can fix itand troublingly, we dont know what is actually happening."

Campbell credits NIST colleague K.P. Cheung for first noticing the possibility of trouble with the theory.