A couple of weeks I ago I wound up with frostbitten fingers after an ill-advised cold-snap bike ride. It was minor in the grand scheme of frostbite, but I wound up with almost-alarming blue tinge and that characteristic thaw-pain that feels like all of your cells are about to burst open. Taking the actual wind chill together with the apparent wind chill of the moving bike, I'd say we were down to about 10 degrees Fahrenheit.
That's nothing, of course. The average temperature of the universe is around −454.76 degrees Fahrenheit, while the current world record for laboratory-based coldness is about 0.0000000001 of a kelvin (0 kelvins being absolute zero). That record doesn't mean very much in terms of my fingertips, however—it was achieved by cooling the nuclear spins of a piece of rhodium metal. So: subatomic cold.
Macroscale cold, or cold that we can see with our own eyes, is a different matter. At near-zero scales, it's much more difficult to achieve. Rather than dealing with the nucleus of a single atom, we're cooling many whole atoms together, which means restricting the innate motions—read: quantum fluctuations—of many atoms together. To this end, physicists have found a new way of cooling macroscale objects to below previously established limits via a technique known as "squeezed light," according to research published recently in Nature. It is cold that should not be.
The macroscopic object in question was an aluminum plate about 20 micrometers in diameter, or a bit less than half the width of a human hair. This isn't exactly fingertip-scale, but it's an entirely different realm compared to subatomic particles.
The challenge in cooling an object this large lies in what seems to be a fundamental barrier: that quantum backaction limit. This limit is a consequence of the uncertainty found in quantum systems, which are characterized by random fluctuations rather than the well-established positions and velocities we're used to in our everyday not-so-quantum world. The particles that make up our metal plate are always fidgeting, and this motion—this noise—is in defiance of true coldness, and, ultimately, absolute zero, where all motion ceases at every scale. Noise is heat.
The physicists behind the current paper were able to cool beyond the apparent limit of particle noise thanks to lasers. It seems counterintuitive—a laser should add energy to a system, and so it should add heat to the system.
The basic idea is to start with a microwave cavity, which is a small space within which light waves tune themselves to match the natural resonant frequency of the space itself. This cavity contains the aluminum plate. Apply a frequency of light to the microwave cavity that's below its natural resonance and higher frequency light particles that match the resonant frequency of the cavity start appearing. As the space fills with microwaves, these particles leak out. And every time a photon leaks away, it steals a bit of mechanical energy from the plate, cooling it.
This cavity setup has been used before, but what the current research adds is the aforementioned "squeezing." Like all waves, light waves have a property called phase, which is how the waves are scooted forward and backward in time. Phase is subject to quantum fluctuations just like any other measurable property of particles, but it's possible to create waves that are super-regular with respect to phase, and this offers a way of transferring the quantum fluctuations from other properties of the light particles to the fluctuation-stripped phase property. This bit of cheating is how the researchers got beyond the backaction limit.
"We are squeezing the light at a 'magic' level—in a very specific direction and amount—to make perfectly correlated photons with more stable intensity," offers NIST scientist John Teufel, who led the experiment, in a statement. "These photons are both fragile and powerful."
This isn't trivial research. Being able to cool larger and larger systems means being able to explore how quantum physics behaves at larger and larger scales, which is key to tasks like quantum information processing.