LEDs break the efficiency barrier

Sensemaking / LEDs break the efficiency barrier

MIT has produced an ultra-low-power diode that emits more light energy than it consumes in electrons. So how bright is its future?

By Roger East / 15 Aug 2012

MIT has produced an ultra-low-power diode that emits more light energy than it consumes in electrons. So how bright is its future?


In lighting, we tend to see heat as the enemy of efficiency – and LEDs as the coolest (and most efficient) yet. But researchers at MIT have seemingly taken this way off the scale of possibility. Their ultra-low-power diode emits more than twice as much light energy, in photons, as it consumes in electrons (of electrical power). So, what’s the future for the 230% LED? And, indeed, for the fundamental scientific principle that efficiency ratios beyond 1:1 belong firmly in the realms of fiction…?

Second things first: the laws of thermodynamics aren’t transgressed. What Parthiban Santhanam and his colleagues did was to push, to the extreme, a phenomenon that is actually a real problem for those who want bright LED lights – namely that diodes convert electricity most efficiently at very low power. (And we’re talking really very low power here – the amount you measure in picowatts, the millionth part of a millionth of a watt.)

Their method involved repeatedly halving the input voltage. This represented, they calculated, a fourfold drop in the input power – but each time the photon output fell only by half. Ultimately, they cut the input to 30 picowatts of electricity – and recorded an output of 69 picowatts of light. How so? Because the diode was also picking up and converting heat from its immediate environment. In technical terms, the light-emitting process causes the LED to cool, but gives rise to vibrations in its atomic lattice, due to entropy – and these vibrations create the heat it is ‘stealing’.

All fascinating stuff, but how useful in practice? Well, don’t try this at home: you can’t see anything by a 69 picowatt light. It’s conceivable that this research could help with making ultra-low power lights for specialist applications where it’s crucial that they generate no heat. The output levels are so tiny, however, that its future may not lie even in lighting at all, but perhaps in other ways of exploiting the diode’s heat pump effect – such as providing an instantly controllable cooling effect in heat-sensitive solid-state electronics. Either way, it’s an impressive breakthrough, albeit one whose real-world applications may take years to realise.
Roger East

Photo: istockphoto/thinkstock

What might the implications of this be? What related articles have you seen?

Please register or log in to comment.

Suggested