Third Law Of Thermodynamics Gets Proof

Two quantum physics professors have developed a proof of the third law of thermodynamics. As well as being academically satisfying, it could help make quantum computing more viable.

The wording of the third law varies, but a common version is “The entropy of a perfect crystal at absolute zero is exactly equal to zero.” That has a variety of practical effects, one of the most important being that it’s impossible to reduce temperature to absolute zero in a finite number of steps or in a finite time.

Until now, the law hadn’t been mathematically proved and there were even some practical steps involving quantum ‘shortcuts’ that appeared to violate it.

Jonathan Oppenheim and Lluis Masanes of University College London looked at the topic and mathematically derived the principle. It’s some hardcore stuff which I don’t mind admitting is largely over my head, but the upshot is that there’s a relationship between the energy used to cool a system and the speed at which it cools, even when using quantum shortcuts.

In very simplified terms, no finite speed will ever be fast enough to take the system all the way down to absolute zero within a finite time. In turn, infinite speed isn’t reachable without an infinite amount of energy.

The pair also used these relationships to calculate the theoretical maximum speed at which a system can be cooled within a given time. At the moment technological limitations mean such a speed isn’t practical, so it’s more a case of knowing the ultimate limit.

As well as filling in a missing gap in physics, the work could have some practical benefits for quantum computing, which takes advantage of the way quantum bits can be in multiple states simultaneously, unlike the binary on/off (or 0/1) of traditional computing.

While keeping processors from overheating is a big part of operating computers, temperature control and cooling are even more important with quantum computing where temperature changes can change the state of particles and potentially wreck the data.