Bottling the world’s coldest plasma

Laser-cooled plasma-in-a-bottle could answer questions about the sun, fusion power


Research News


HOUSTON – (March 1, 2021) – Rice University physicists have discovered a way to trap the world’s coldest plasma in a magnetic bottle, a technological achievement that could advance research into clean energy, space weather and astrophysics.

“To understand how the solar wind interacts with the Earth, or to generate clean energy from nuclear fusion, one has to understand how plasma — a soup of electrons and ions — behaves in a magnetic field,” said Rice Dean of Natural Sciences Tom Killian, the corresponding author of a published study about the work in Physical Review Letters.

Using laser-cooled strontium, Killian and graduate students Grant Gorman and MacKenzie Warrens made a plasma about 1 degree above absolute zero, or approximately -272 degrees Celsius, and trapped it briefly with forces from surrounding magnets. It is the first time an ultracold plasma has been magnetically confined, and Killian, who’s studied ultracold plasmas for more than two decades, said it opens the door for studying plasmas in many settings.

“This provides a clean and controllable testbed for studying neutral plasmas in far more complex locations, like the sun’s atmosphere or white dwarf stars,” said Killian, a professor of physics and astronomy. “It’s really helpful to have the plasma so cold and to have these very clean laboratory systems. Starting off with a simple, small, well-controlled, well-understood system allows you to strip away some of the clutter and really isolate the phenomenon you want to see.”

That’s important for study co-author Stephen Bradshaw, a Rice astrophysicist who specializes in studying plasma phenomena on the sun.

“Throughout the sun’s atomosphere, the (strong) magnetic field has the effect of altering everything relative to what you would expect without a magnetic field, but in very subtle and complicated ways that can really trip you up if you don’t have a really good understanding of it,” said Bradshaw, an associate professor of physics and astronomy.

Solar physicists rarely get a clear observation of specific features in the sun’s atmosphere because part of the atmosphere lies between the camera and those features, and unrelated phenomena in the intervening atmosphere obscures what they’d like to observe.

“Unfortunately, because of this line-of-sight problem, observational measurements of plasma properties are associated with quite a lot of uncertainty,” Bradshaw said. “But as we improve our understanding of the phenomena, and crucially, use the laboratory results to test and calibrate our numerical models, then hopefully we can reduce the uncertainty in these measurements.”

Plasma is one of four fundamental states of matter, but unlike solids, liquids and gases, plasmas aren’t generally part of everyday life because they tend to occur in very hot places like the sun, a lightning bolt or candle flame. Like those hot plasmas, Killian’s plasmas are soups of electrons and ions, but they’re made cold by laser-cooling, a technique developed a quarter century ago to trap and slow matter with light.

Killian said the quadrupole magnetic setup that was used to trap the plasma is a standard part of the ultracold setup that his lab and others use to make ultracold plasmas. But finding out how to trap plasma with the magnets was a thorny problem because the magnetic field plays havoc with the optical system that physicists use to look at ultracold plasmas.

“Our diagnostic is laser-induced fluorescence, where we shine a laser beam onto the ions in our plasma, and if the frequency of the beam is just right, the ions will scatter photons very effectively,” he said. “You can take a picture of them and see where the ions are, and you can even measure their velocity by looking at the Doppler shift, just like using a radar gun to see how fast a car is moving. But the magnetic fields actually shift around the resonant frequencies, and we have to disentangle the shifts in the spectrum that are coming from the magnetic field from the Doppler shifts we’re interested in observing.”

That complicates experiments significantly, and to make matters even more complicated, the magnetic fields change dramatically throughout the plasma.

“So we have to deal with not just a magnetic field, but a magnetic field that’s varying in space, in a reasonably complicated way, in order to understand the data and figure out what’s happening in the plasma,” Killian said. “We spent a year just trying to figure out what we were seeing once we got the data.”

The plasma behavior in the experiments is also made more complex by the magnetic field. Which is precisely why the trapping technique could be so useful.

“There is a lot of complexity as our plasma expands across these field lines and starts to feel the forces and get trapped,” Killian said. “This is a really common phenomenon, but it’s very complicated and something we really need to understand.”

One example from nature is the solar wind, streams of high-energy plasma from the sun that cause the aurora borealis, or northern lights. When plasma from the solar wind strikes Earth, it interacts with our planet’s magnetic field, and the details of those interactions are still unclear. Another example is fusion energy research, where physicists and engineers hope to recreate the conditions inside the sun to create a vast supply of clean energy.

Killian said the quadrupole magnetic setup that he, Gorman and Warrens used to bottle their ultracold plasmas is similar to designs that fusion energy researchers developed in the 1960s. The plasma for fusion needs to be about 150 million degrees Celsius, and magnetically containing it is a challenge, Bradshaw said, in part because of unanswered questions about how the plasma and magnetic fields interact and influence one another.

“One of the major problems is keeping the magnetic field stable enough for long enough to actually contain the reaction,” Bradshaw said. “As soon as there’s a small sort of perturbation in the magnetic field, it grows and ‘pfft,’ the nuclear reaction is ruined.

“For it to work well, you have to keep things really, really stable,” he said. “And there again, looking at things in a really nice, pristine laboratory plasma could help us better understand how particles interact with the field.”


The research was supported by the Air Force Office of Scientific Research (FA9550-17-1-0391) and the National Science Foundation Graduate Research Fellowship Program (1842494).


Links and resources:

The DOI of the Physical Review Letters paper is: 10.1103/PhysRevLett.126.085002

A copy of the paper is available at:

VIDEO is available at:

High-resolution IMAGES are available for download at:
CAPTION: Images produced by laser-induced fluorescence show how a rapidly expanding cloud of ultracold plasma (yellow and gold) behaves when confined by a quadrupole magnet. Ultracold plasmas are created in the center of the chamber (left) and expand rapidly, typically dissipating in a few thousandths of a second. Using strong magnetic fields (pink), Rice University physicists trapped and held ultracold plasmas for several hundredths of a second. By studying how plasmas interact with strong magnetic fields in such experiments, researchers hope to answer research questions related to clean fusion energy, solar physics, space weather and more. (Image courtesy of T. Killian/Rice University)
CAPTION: Rice University graduate student Grant Gorman at work in Rice’s Ultracold Atoms and Plasmas Lab. (Photo by Jeff Fitlow/Rice University)
CAPTION: Rice University physicists (from left) Grant Gorman, Tom Killian and MacKenzie Warrens discovered how to trap the world’s coldest plasma in a magnetic bottle, a technological achievement that could advance research into clean energy, space weather and solar physics. (Photo by Jeff Fitlow/Rice University)
CAPTION: Rice University graduate student MacKenzie Warrens adjusts a laser-cooling experiment in Rice’s Ultracold Atoms and Plasmas Lab. (Photo by Jeff Fitlow/Rice University)
CAPTION: Rice University plasma physicist Stephen Bradshaw studies solar flares, heating in the sun’s atmosphere, solar wind and other solar physics phenomena. (Photo by Jeff Fitlow/Rice University)

This release can be found online at

Follow Rice News and Media Relations via Twitter @RiceUNews.

Located on a 300-acre forested campus in Houston, Rice University is consistently ranked among the nation’s top 20 universities by U.S. News & World Report. Rice has highly respected schools of Architecture, Business, Continuing Studies, Engineering, Humanities, Music, Natural Sciences and Social Sciences and is home to the Baker Institute for Public Policy. With 3,978 undergraduates and 3,192 graduate students, Rice’s undergraduate student-to-faculty ratio is just under 6-to-1. Its residential college system builds close-knit communities and lifelong friendships, just one reason why Rice is ranked No. 1 for lots of race/class interaction and No. 1 for quality of life by the Princeton Review. Rice is also rated as a best value among private universities by Kiplinger’s Personal Finance.

From EurekAlert!

4.7 3 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
March 7, 2021 3:36 am


Reply to  Willem69
March 7, 2021 5:11 am


Reply to  Willem69
March 7, 2021 9:03 am

Fusion clean energy sequel goes on, its first episode inspired start of another TV great “I love Lucy”

March 7, 2021 5:27 am

Newton’s law of gravity is simple.
Maxwell’s equations of electrodynamics are not.

Reply to  BillR
March 7, 2021 9:34 am

Without Maxwell’s work Einstein would not get anywhere, and he said so himself :
 ‘I stand not on the shoulders of Newton, but on the shoulders of James Clerk Maxwell.’

Reply to  Vuk
March 7, 2021 10:27 am

Maxwell did a thought experiment: if change in the magnetic field produces changing electric field, and the changing electric field produces changing magnetic field and so on, kind of wave is generated by positive feedback of two changing interlinked fields, how fast such wave would propagate.
Using magnetic and electric permeability μ0  & ε0  of vacuum (or free space) and applying square law Maxwell set up this little less known equation v = 1/(μ0 ε0)^0.5.
When calculated it happen  to be ~ 299,792 km/s. This was Maxwell’s ‘eureka moment’, since result was same as the measured speed of light c.
Maxwell realised that light is ‘electromagnetic wave’ not a particle as commonly thought at the time.
If so the speed of light has to be constant regardless of frame of reference, i.e. inverse of a square root of two fundamental nature’s constants.
Einstein took over from there with his well known thought experiments.
 RF said: ‘formulate your hypothesis, do experiment etc….’
Maxwell and Einstein could not do required experiments at the time, but later experimental results proved their hypothesis to be correct and so become fundamental theories of contemporary science.

Reply to  Vuk
March 7, 2021 1:14 pm

Electromagnetic wave as envisaged by Maxwell, actual reality is more accurately shown by quantum mechanics four dimensional (x,y,z,t) fields theory representation.

EM wave.gif
Roger Taguchi
Reply to  Vuk
March 8, 2021 1:36 am

If the Poynting vector varies as ExB, shouldn’t the propagation vector in the diagram be in the opposite direction (using the Right Hand Rule, curl the fingers from E to B)? Or maybe switch labels of E and B?

Reply to  Roger Taguchi
March 8, 2021 6:39 am

Electromagnetic phenomenon is clearly exhibiting quantum properties, it might be more appropriate to use quantum mechanics fields concept.

Reply to  Roger Taguchi
March 8, 2021 7:52 am

You are assuming for the not-shown y-axis that +B is to left, but +B must be to the right. This would have been clear if the x/y axes were shown. Only the propagation (z) axis is shown.

Reply to  Vuk
March 8, 2021 5:11 am

That isn’t used anymore for a QM wave, it has to much classical representation embedded in it. If you want a quantum representation draw the field everywhere then draw a spin in the field same as you would draw a 3d wind formation like a cyclone etc

No photon would go in a circle but this gives the idea particles are simply patterns in the field.

Spacetime is defined as everywhere that the field extend to and it explains why entanglement works because no matter how far you take the particles apart the fields connect them.

Reply to  LdB
March 8, 2021 7:01 am

Ldb, I did say:”Electromagnetic wave as envisaged by Maxwell, ….. “, see above.

Reply to  Vuk
March 9, 2021 3:27 pm

So where is the magic 4th dimension “t” on your diagram?

Reply to  Vuk
March 9, 2021 4:12 pm

Actually, that formula is well known and appears in most (good) physics undergrad textbooks.
Scientists often wonder why is the speed of light what it is. This equation explains it although most physicists seem to have missed the implication. μ0 and ε0 (i.e. permeability of a magnetic field and permittivity of an electric field) are properties of the medium through which it is passing i.e. space itself. The nature of Space as an entity is studiously ignored by modern science.

As an aside,Maxwell’s laws were an unworkable hodge-podge until Oliver Heaviside tidied them up, reduced them from 17 equations to 4, and rewrote them in the vector form we know today.
In 1889, Heaviside came up with what is famously known today as the relativistic factor (1 – (v2/c2))*-0.5, which is the centrepiece of relativity. This was 16 years before the Special Theory of Relativity. In other words, this arises out of Electromagnetic Field Theory without any need to resort to relativity to explain it. Einstein failed to acknowledge this although he would doubtless have known about it.
Heaviside communicated this to the Irish mathematician Fitzgerald who, with Lorenz, then proceeded to work out the Fitzgerald-Lorenz contraction.
Furthermore, in 1891 Henri Poincare, the French mathematician and physicist, used this to work out that M=E/c2, years before Einstein did. Again, their was no recognition by Einstein of this fact and he claimed the credit.

And Maxwell could not have come up with his theory of this without the ground-breaking idea of FORCE FIELDS that Michael Faraday came up with. This was totally at odds with what everyone else thought at the time and was truly revolutionary but this is now forgotten. To add to the historic insult, Maxwell’s first equation is now called Gauss’s Law and the second is called Ampere’s Law.

March 7, 2021 6:02 am

At least the graduate student in the picture has his mask on so he won’t get the CV from the plasma.

March 7, 2021 6:34 am

I’m not familiar with this cold plasma that the article talks about. I thought plasmas were always hot.
Does anyone have a source that talks about cold plasmas using layman’s terms?

Reply to  MarkW
March 7, 2021 8:11 am

I had the same question. The answer seems to be that one takes liberties with the concept of temperature. As you know, a plasma is conductive because it has independently moving nuclei and electrons, which implies that if thesystem is in thermal equilibrium (i.e., if the conventonal definition of “temperature” applies) then the kinetic energies of the individual particles must be around the energy released when charges recombine, i.e., on the order of a few electron volts, which would correspond to temperatures of thousands of degrees. But these cold-plasma people are (as best I can tell from Wikipedia) talking about systems in which the “electron temperature” is considered separately from the “ion temperature”, the former being (necessarily) high enough to keep things ionized, while the latter can be lower — apparently much lower.

Reply to  Peter
March 7, 2021 9:59 am

Right, see “Nonthermal plasma” in the wikipedia (or in your favorite reference). This measure makes some sense, because ions are orders of magnitude more massive than electrons, and temperature–in the usual sense (when dealing with neutral atoms)–is based on the speed of atoms.
A minor quibble: a plasma need not be made of *nuclei* and electrons. A plasma can be partially ionized, meaning the ion retains one or more electrons. (A hydrogen plasma is an exception to this, since neutral hydrogen has only one electron to begin with. Hydrogen plasma is of course the kind of plasma relevant to nuclear fusion, although it may be rich in isotopes.)

Rich Davis
Reply to  MarkW
March 7, 2021 8:28 am

It’s EurekAlert!, so it should be rubbish.

Reply to  MarkW
March 7, 2021 9:15 am

“…. made a plasma about 1 degree above absolute zero, or approximately -272 degrees Celsius”
Superconductivity is a kind of ‘solid cold plasma’ except proper plasma is 4th state of matter, only interlinked chain of electrons (charged particles) moves about trough the positively charged ions of the molecular lattice without any resistance. Regretfully energy required to maintain the super-low temperatures is greater than the energy that can be retrieved from resistance free movement of electrons.

Michael S. Kelly
Reply to  MarkW
March 8, 2021 4:02 pm

In a helium-neon laser, the gas is at roughly 300 K (i.e. room temperature), and half of it is ionized. That’s a cold plasma. It’s also quite far from equilibrium, which makes it a laser medium.

D-T fusion requires a temperature of 150 million C. That sounds like a lot, but remember – it’s a dry heat.

Bruce Cobb
March 7, 2021 6:54 am

Now if they could just put time in a bottle…

March 7, 2021 8:59 am

I do not think a candle flame is hot enough to produce a plasma. Shouldn’t Bradshaw know that?

Reply to  joselori
March 7, 2021 10:08 am

Right about candles, see instructive article here: But I’m not sure what that has to do with Bradshaw. Cold (“nonthermal”) plasmas consist of cool ions plus “hot” electrons. The mercury vapor gas in an ordinary florescent lamp is a cold plasma, in the sense that the electrons are “hot” and the ions are not. That’s why you can touch a florescent bulb without burning your fingers.

March 7, 2021 10:11 am

There’s probably something to learn about the behavior of low-density and cold plasmas in deep space (extremely low density such that the ions can’t immediately recombine with electrons to make neutral atoms). However, the claim that this applies to fusion is pure BS. To produce power by fusion requires the plasma to be extremely hot, hot enough so the kinetic energy of the ions can overcome the electrostatic repulsion between ions so they can combine (fuse) to make heavier atoms thus releasing energy. The challenges associated with confining a hot, low atomic weight plasma (deuterium, tritium) are completely different from that for a cold, high atomic weight (strontium) plasma).

This false claim was made to help secure future research funding – fusion power is important, deep space astrophysics not so much. Where else have we heard false claims made just to secure future funding?

Reply to  Meab
March 7, 2021 3:21 pm

My impression is that they are trying to gain some control over plasma so they can,hopefully, do experiments and learn more about how it behaves. This “cold plasma” state is one where some control may be possible, starting with current understanding and current technology. These experiments may eventually lead to enough insight to be useful in dealing with other plasma (very hot) that might be useful for fusion energy production. That is not the same as saying these experiments are about producing fusion energy.

Jim Whelan
March 8, 2021 7:35 am

Not to denigrate this work but I seriously doubt it will make big difference in detailed knowledge about very hot plasma for the simple reason that very hot plasma tends to be chaotic. Kind of like what happens in fluid dynamics when flow reaches certain velocities. Parts of the equation tend to zero and the results cannot be calculated to any degree of accuracy.

James F. Evans
March 9, 2021 11:07 am

An excellent article.

But one quibble: the electric field needs to be discussed.

No electric field, no magnetic field… full stop.

But it is hard to observe & measure an electric field.

A stable electric field may be key to a stable magnetic field.

And further discoveries of physical nature.

And the technology that derives from our understanding.

%d bloggers like this:
Verified by MonsterInsights