COLD CASE: An artist's depiction of ultracold potassium-rubidium molecules, whose chemical interactions are noticeably governed by quantum mechanics.
Image: G. Kuebler/JILA
-
Gravity's Engines
We’ve long understood black holes to be the points at which the universe as we know it comes to an end. Often billions of times more massive than the Sun, they...
Read More »
A new study shows that molecules cooled to have near-negligible collisional motion can still react chemically with one another. At just a few hundred nanokelvins above absolute zero, the researchers could even change the speed of the chemical reaction by tweaking the molecules' quantum states, paving the way for highly controlled chemistry using the tools of physics. (A nanokelvin is one billionth of a kelvin.)
The study appears in the February 12 issue of Science, authored by scientists from two institutes affiliated with the National Institute of Standards and Technology (NIST): JILA, run jointly by NIST and the University of Colorado at Boulder; and the Joint Quantum Institute, a partnership between NIST and the University of Maryland, College Park.
In 2008 a group including many of the same researchers announced the creation of a dense gas of potassium–rubidium (KRb) molecules at a few hundred nanokelvins. Now that ultracold gas has been shown to decay through a heat-releasing chemical reaction as its molecules interact through the phenomenon of quantum tunneling, in which particles skip over classical barriers. In this case the blockade is a so-called momentum barrier between two identical molecules experiencing mutual repulsion.
At ultracold temperatures, classical physical conceptions of molecules become less useful than quantum-mechanical ones, says Jun Ye, a JILA physicist and study co-author. "They're so cold that you can no longer think of them as a ping-pong–ball kind of object," Ye says. "They're really quantum-mechanical waves." Those waves can overlap at relatively great distances, leading to remote interactions between the molecules. "As soon as they feel each other's presence, when their wave function starts to overlap, very interesting things start to happen," Ye says. In this case the gas molecules swap atoms to form K2 and Rb2 molecules, which then escape.
Shifting the starting conditions in the potassium–rubidium gas highlighted the quantum-mechanical nature of the chemical reactions. When all the molecules were set to the same initial quantum state, the gas decayed over several seconds. But when the molecules were prepared in different states—specifically, in a heterogeneous mix of spins—the reaction proceeded 10 to 100 times faster. The difference is a logical extension of the Pauli exclusion principle: Identical molecules repel one another to avoid occupying the same place simultaneously.
"In order for them to approach each other, they would have to overcome this momentum barrier," Ye says. "Tunneling is really what is happening; they're tunneling through this angular-momentum barrier." Distinguishable molecules—those with different spins—are freer to cozy up.
Jeremy Hutson, a chemistry professor at Durham University in England who wrote an accompanying commentary for Science on the research, says that the fine-scale physical manipulation possible in the ultracold regime provides an accompanying level of chemical control. "I think it's the selectivity that's remarkable under these circumstances," Hutson says. "You can make a change as tiny as flipping a singular nuclear spin and completely change the course of the reaction."
Hutson says that it may soon be possible to act on a whole population of atoms or molecules at once. "The strength of this general field, this ultracold chemistry, if you like, is that in the future it is likely to be possible to carry out chemical reactions on entire samples of molecules in a coherent and controlled way," he says.
But now that the JILA and Joint Quantum Institute researchers have shown how readily chemistry can proceed at ultracold temperatures, Ye wants to figure out how to nip it in the bud to extend the lifetime of the potassium–rubidium gas, which at present begins to chemically degrade in about a second. "The goal for us is to be able to suppress these reactions," Ye says. "Now that we can understand and control them, the next step is to really learn how to eliminate them."
See what we're tweeting about
15 Comments
Add Comment"Now that ultracold gas has been shown to decay through a heat-releasing chemical reaction"
Reply | Report Abuse | Link to thisI was under the impression that when things get that near zero K there isn't any heat to loose. If you can remove heat from something that cold maybe this will get us a step closer to absolute 0 Kelvin.
I'd also be interested in knowing if this would have any orthologs in the E8 theory of everything...
Temperature is a measure of the kinetic energy of random motion in a system. There can still be potential energy, and what happens in the collisions here is that some of the potential energy (stored in chemical bonds) gets transferred into kinetic energy. The kinetic energy released doesn't really become "heat" unless and until it is randomized in subsequent collisions.
Reply | Report Abuse | Link to thisIt's hard to imagine any substance in gaseous state in that sort of temperature...
Reply | Report Abuse | Link to thisI was taught that the probability of a tunneling event varies as e^1/mass. If that's true, then the probability of a rubidium atom tunneling is something like e^156,000 times less likely than an electron tunneling (I assume this means 'tunneling through the same potential barrier').
Reply | Report Abuse | Link to thisHave I been misinformed, or have I misunderstood, or is something odd going on at NIST?
@Danthrax: It's actually exp[-sqrt(mass*barrier/hbar^2)]. But you're right that it's enormously slower for a heavy particle than for (say) an electron, and it's the whole KRb molecule that's doing the tunneling here (not just an atom). But the barrier here is absolutely tiny: it arises from the centrifugal potential due to rotation of the KRb molecules about one another, and is measured in microKelvin. So the net effect is that the need to tunnel slows the reaction down a lot but doesn't stop it completely.
Reply | Report Abuse | Link to thisJeremy Hutson
Jeremy, think of it in this way: Anything that has even the smallest amount of heat is able to give off that heat in the right environment.
Reply | Report Abuse | Link to thisNow if we are looking at nanoscale temperatures and you want to get to absolute zero--I, in the past, have expounded on the proposition that absolute zero will be found in a perfect vacuum--then I would hypothesize that the smallest temperature that can be removed to provide absolute zero would also be the amount of energy of the graviton.
Jeremy, think of it in this way: Anything that has even the smallest amount of heat is able to give off that heat in the right environment.
Reply | Report Abuse | Link to thisNow if we are looking at nanoscale temperatures and you want to get to absolute zero--I, in the past, have expounded on the proposition that absolute zero will be found in a perfect vacuum--then I would hypothesize that the smallest temperature that can be removed to provide absolute zero would also be the amount of energy of the graviton.
Does any of this improve the warmth of my socks in winter?
Reply | Report Abuse | Link to thisMirror images of molecules exist. With a modified spin, there is some similarity to the story about supercooled water freezing when it is heated.
Reply | Report Abuse | Link to thisIf you can control the reactions that precisely, could this then be used as a computer; or perhaps an information storage device?
Reply | Report Abuse | Link to thisIt seems a bit redundant to state that the temperature is "nanokelvins above absolute zero" since kelvin is an absolute scale of temperature. I would not say that an object is 1m longer than zero.
Reply | Report Abuse | Link to thisThe explanations fall flat. It's hard to believe that at zero K temperatures, with linear and vibrational motions nearly stopped, molecules can affect each other at greater distances than at warmer temperatures. This is similar to claiming that a cold, dead sun exerts more gravitational force than a burning sun of identical mass. The claim that very cold molecules act like waves is equally silly. Do they have regular, periodic movements of fixed amplitude? Do two molecules exhibit constructive and destructive interference when they move together? What's their frequency, Kenneth? Oh, never mind, they're "quantum-mechanical waves" so they can act in any way that's necessary to "explain" the outcomes.
Reply | Report Abuse | Link to thisI really tire of scientists (especially theoretical physicists and climatologists) who invent nonsense to explain results they do not understand.
@Dr.T, there was never a claim that the molecules affect each other at greater distances when they are cold. Instead, the point is that the quantum mechanical effects are no longer swamped out by thermal ones.
Reply | Report Abuse | Link to thisThink of trying to detect whether two tiny tuning forks were in tune or not, standing next to the speakers at a rock concert. No, the tuning forks themselves will become driven oscillators in that environment. You need to get into a very quiet room to do your experiment on the tiny tuning forks.
And you can only detect some quantum mechanical effects if you get the thermal noise down to vanishingly low levels.
As to whether molecules act like waves, yes, it's been established experimentally beyond really any doubt. Constructive and destructive interference in a beam of gold ions (much more massive than a KRb molecule) has been conclusively demonstrated.
Louis DeBroglie suggested this in his PhD thesis in 1924, and it was first confirmed experimentally in 1927. DeBroglie received the Nobel Prize in physics for this. Look up the Davisson-Germer experiment for starters.
While I myself am skeptical about current doomsday claims by some climatologists, its in part because their computer models are not able to describe real weather--let alone climate--systems at all. Quite to the contrary, particle physicists are not relying on dubious computer models. In fact, the computer models they use can agree with experiment to 9 decimal places (read QED by Richard Feynman, for starters).
If someone can consistently throw a dart 100 miles in any direction and always hit where he said he would, within the width of a human hair, you ought to admit that he is not "inventing nonsense," but rather might know a thing or two about dart throwing.
@Dr.T
Reply | Report Abuse | Link to thisI forgot to add that, yes, the KRb molecules do have a frequency and they do have a wavelength. Look up DeBroglie wavelength and you will find the formula for determining the wavelength of any particle based on its mass. Here's a handy calculator:
http://hyperphysics.phy-astr.gsu.edu/HBASE/quantum/debrog2.html
The wavelength, of course, varies with the velocity of the particle and at high velocities there are relativistic effects. At the site above are calculators for both relativistic and classical velocities. The DeBroglie frequency of an electron (if my math is right) is something like 5 X 10^16 per second. A KRb molecule will be much, much higher.
@Robert Schmidt: Yes, there are certainly proposals to use ultracold molecules in optical lattices for quantum computing. See for example David DeMille's paper "Quantum computation with trapped polar molecules" in Physical Review Letters 88, 067901 (2002). Although the proposal seemed quite futuristic when it was published, in fact most of the experimental obstacles have now been overcome. This is an idea that might be realized quite soon!
Reply | Report Abuse | Link to this