In the 1950s, physicist Léon Brillouin published a landmark paper refuting the idea that adding a single diode, a one-way electrical gate, to a circuit is the solution to harvesting energy from Brownian motion. Knowing this, Thibado's group built their circuit with two diodes for converting AC into a direct current (DC). With the diodes in opposition allowing the current to flow both ways, they provide separate paths through the circuit, producing a pulsing DC current that performs work on a load resistor.

Additionally, they discovered that their design increased the amount of power delivered. "We also found that the on-off, switch-like behavior of the diodes actually amplifies the power delivered, rather than reducing it, as previously thought," said Thibado. "The rate of change in resistance provided by the diodes adds an extra factor to the power."

The team used a relatively new field of physics to prove the diodes increased the circuit's power. "In proving this power enhancement, we drew from the emergent field of stochastic thermodynamics and extended the nearly century-old, celebrated theory of Nyquist," said coauthor Pradeep Kumar, associate professor of physics and coauthor.

According to Kumar, the graphene and circuit share a symbiotic relationship. Though the thermal environment is performing work on the load resistor, the graphene and circuit are at the same temperature and heat does not flow between the two.

That's an important distinction, said Thibado, because a temperature difference between the graphene and circuit, in a circuit producing power, would contradict the second law of thermodynamics. "This means that the second law of thermodynamics is not violated, nor is there any need to argue that 'Maxwell's Demon' is separating hot and cold electrons," Thibado said.

I don't understand. It sounds like they're drawing useful work (I think?) from a system that's already at thermodynamic equilibrium. Doesn't that violate the 2nd law of thermodynamics?