Spinning the Core | Science News

Love Science? Welcome Home.

Support Amazing Science Journalism.

Create the New Science Generation.


Spinning the Core

Laboratory dynamos attempt to generate magnetic fields the way planets and stars do

3:20pm, May 2, 2013
dynamo experiment

The dynamo experiment at the University of Maryland is the biggest ever made with  whirling molten sodium.

Sponsor Message

Daniel Lathrop spent seven years and $2 million building the stainless steel sphere in his laboratory. It’s two spheres, actually — nestled one within the other like a pair of Russian dolls. Only these dolls contain 12 tons of molten metal and spin independently at astonishing speeds.

With his contraption, Lathrop, a physicist at the University of Maryland in College Park, hopes to re-create the Earth’s spinning metal heart. As the planet rotates on its axis, electrically conducting liquid iron churns thousands of kilometers down in the outer core. The iron’s sloshing motion, in a process called a dynamo, creates and sustains Earth’s magnetic field.

Given the crucial role the planet’s magnetic field plays in guiding navigators and protecting Earth from solar storms, scientists know surprisingly little about it. Geophysicists don’t know exactly how the magnetic field got started billions of years ago or how it has managed to sustain itself for so long. It’s even a mystery why Earth has a magnetic field in the first place. Not all planets do. Tiny Mercury has one, for instance, while Mars has none. Stars like the sun generate powerful internal dynamos as well, and laboratory models filled with superhot metal might be able to re-create them.

Lathrop’s goal is to provide some hard science about where Earth’s magnetism comes from. His Russian dolls live in a laboratory at the university, where they stand in as a miniaturized version of the Earth. Liquid sodium filling the space between the inner and outer spheres replaces the planet’s liquid iron outer core. Lathrop hopes that the swirling sodium will create its own dynamo and generate a self-sustaining magnetic field.

If the experiment works, Lathrop’s team will be able to study the forces that drive Earth’s dynamo and determine what might happen to our magnetic field in the future. It has flipped quasi-regularly in Earth’s past, so that magnetic north becomes magnetic south, and vice versa. Some scientists think the planet is due for another flip. “We could be headed for a reversal right now, but that’s just a hunch,” says Lathrop. “We’re stuck between hunches and science.”

Since last spring, when his device was first filled with sodium, Lathrop has turned it on about once a month. Flip a switch on Monday morning, and by midday Tuesday all the sodium — which is solid at room temperature — has gotten hot enough to melt. Flip another switch and the spheres begin whirling like dervishes, churning the liquid sodium between them.

Instruments around the device gather information about how the liquid is flowing and whether it is generating any magnetism. No dynamo yet, but the device allows the team to generate huge amounts of data very quickly. “Every one second in our experiment mimics 5,000 years of Earth’s history,” says Lathrop. “In a few hours, I can deliver millions of years of high-quality data.”

Building a better dynamo

“What’s interesting from an experimental point of view is that dynamos are a threshold phenomenon — you either get one or you don’t get one,” says Peter Olson, a geodynamicist at Johns Hopkins University. “You have two options for making one: You can start with the most Earthlike configuration and try to work up to that threshold, or you can start with a less realistic configuration, make a dynamo and then start removing the unrealistic bits.”

The latter approach is how three groups have already achieved dynamos in a lab. The first two were reported in 2000, in Riga, Latvia, and in Karlsruhe, Germany. Both forced liquid sodium in cylindrical tanks to flow in a helical pattern — a twisty motion that was enough for the fluid to generate a dynamo. 

Building on that, physicists put together a third sodium experiment in Cadarache, France. “It’s modeled after a French washing machine — a great agitating device,” says Olson. It uses a copper cylinder filled with liquid sodium stirred by a disk at each end. The disks can rotate in the same or opposite directions, essentially pushing or pulling the sodium and setting up all sorts of chaotic flows.

In 2006, the Cadarache experiment generated a dynamo. The device shows a much richer variety of magnetic behavior than the earlier two. For example, the direction of the magnetic field in the device reverses direction every so often, as Earth’s does.

But the machine works only because it contains some of those unrealistic bits Olson refers to. In particular, it makes a dynamo only if one or both of the stirring disks are made of iron. That introduces an extra magnetic force that helps the dynamo get going. Take out the iron, and the Cadarache machine no longer crosses the threshold.

The latest suite of sodium experiments uses spheres, not cylinders, to rotate the fluid in a more planetlike scenario. Earth is, after all, round. None of these experiments has yet achieved a dynamo, but they have contributed to some important discoveries that may help Lathrop create his. In particular, researchers have learned a lot about turbulence, the unpredictable changes in direction that a flowing liquid sometimes takes.

Imagine a fast-flowing river in which eddies carry the water from the center current to the stationary banks. Those eddies — the turbulence — suck speed from the middle of the river and move it to where it rapidly decays. Turbulence of the same sort normally plays havoc with an experimental dynamo, says Cary Forest, a physicist at the University of Wisconsin–Madison.

Forest and his colleagues have been working with a sodium experiment smaller than Maryland’s. In 2006, they reported that turbulence within the sodium flow in their device generates its own weak magnetic field. That, in turn, lowers the conductivity of the sodium, making it hard to get enough electrical charges flowing fast enough to set off a true dynamo.

“That’s a big killer,” says Forest. “You have to spin your system five times as fast to get it up to the point where you thought you had to be.” Nevertheless, the discovery helped explain why the new generation of sodium experiments hasn’t been able to generate dynamos yet.

On the other hand, if you get a dynamo going in the first place, turbulence may not be so much of a problem. In an experiment in Grenoble, France, scientists have forced a strong magnetic field onto the flowing sodium. Because of that they can essentially suppress much of the turbulence that usually roils the liquid, says team leader Henri-Claude Nataf of the University of Grenoble.

That, Nataf says, indicates what is happening in planetary cores. Once a planet like Earth starts spinning and generating its own magnetic field, that magnetism tamps down turbulence. The scientists at Grenoble can now study how that happens inside the flowing sodium in their experiment.

Meanwhile, back in Maryland

Lathrop’s spinning spheres at Maryland are without a doubt the big daddy of the sodium experiments, and while the sodium portion is just getting under way, the team tested the experiment with water several years ago to be sure all the mechanical parts worked before tanking up with a liquid metal that can give off a highly flammable gas.

Even then, the scientists began discovering unexpected things. The water showed flows forced by Earth’s precession, the wobbling of the planet’s rotational axis in space. That observation, Lathrop says, supports the idea that similar flows exist in Earth’s core.

Starting in late 2011, the Maryland scientists drained the water from between the spheres to make way for sodium. The metal is commercially available for making indigo dye for blue jeans, and Lathrop’s team ordered 62 barrels of the stuff. Just heating it up enough to liquefy it and then loading it all in took almost five months.

“I wouldn’t want to do that again,” says Lathrop. He and local fire safety officers had to get creative because liquid sodium is so dangerous. They invented a new way to put out laboratory fires in case of any accidents. And for safety’s sake the experiment initially ran at two revolutions per second, which is half of the fastest speed it can achieve.

Even at that speed, and before achieving a dynamo, the Maryland machine is hinting at new discoveries. The team has documented 15 different flow states. Like weather patterns in Earth’s atmosphere, each flow comes with its own complications. “We are sailing out into uncharted territory,” Lathrop says.

He thinks the machine will have no problem generating a dynamo once it powers up to full speed, probably later this year. Among other things, Lathrop will be looking for magnetic field reversals like those seen on Earth. 

Ever since scientists generated the first global model of Earth’s magnetic field nearly 180 years ago, its strength has decreased by some 10 percent. That might indicate that the planet is heading into a reversal right now (the last one happened 780,000 years ago; they generally take about several thousand years from start to finish). If Lathrop’s machine can generate a dynamo and then start flipping direction, scientists might have more insight into what triggers such changes on Earth — and how likely it is that we are headed for another one.

Beyond sodium

Not all scientists are content with sodium experiments, even very big ones. At Wisconsin, Forest is trying to take the idea of a dynamo up a notch. Quite a few notches, actually — with the superheated state of matter known as plasma.

Liquid metals are a good generalization for studying Earth’s core, Forest says. But most dynamos in the universe, those within stars, are entirely different beasts. They operate in magnetic regimes far beyond Earth’s.

Scientists measure the strength of a dynamo with something called the magnetic Reynolds number. A low magnetic Reynolds number means that the dynamo is weak and could soon dissipate. A high number means that the dynamo is powerful. Earth’s magnetic Reynolds number is on the order of 1000. The sun’s is on the order of 100 million. And plasma streaming between galaxies can have a magnetic Reynolds number more like a million billion.

Forest designed his plasma experiment to mirror magnetic regimes far beyond Earth’s. If successful, it could give researchers an unprecedented glimpse of what happens around black holes and within the hearts of stars. “There’s so much to learn,” he says.

The problem is that plasma is difficult to contain. In research machines such as fusion reactors, scientists use strong magnetic fields to confine plasma, but those fields interfere with seeing what might happen during a natural dynamo. “It’s almost impossible to study how magnetic fields come into being using plasma because you need a magnetic field there to begin with,” says Forest. “It violates the rules of the dynamo game.”

Forest figured out a work-around by putting the whole machine in a sort of magnetic bucket and then attaching 3,000 strong magnets to the surface of the outer sphere. The surface magnets clear out the plasma in the outermost portion of the device and stir the plasma remaining within to create turbulent flows for study.

At 3 meters in diameter, the Madison plasma experiment is the same size as the Maryland sodium one. Forest’s team can pump in a little helium or argon gas, add voltage and create a plasma at some 50,000° to 100,000° Celsius. “It looks even more cool than Dan’s,” Forest says. A clear window on the side of the outer sphere offers a view of the glowing plasma flickering within, like the ethereal dance of the northern lights.

Forest and his colleagues created plasma for the first time in the device last fall, and since then have been measuring its density, temperature and other properties. Some of the flows zip along at nearly 10 kilometers per second, allowing them to achieve very high magnetic Reynolds numbers.

Already, the team is seeing quirky viscosity in the flows. Forest thinks the machine is on the verge of mimicking astrophysical phenomena such as accretion disks of gas and dust swirling into a black hole.

So the race is now on to see which team might achieve a dynamo first: the Madison plasma experiment or the Maryland sodium one. Both are so huge that they may succeed out of sheer size. If so, then physicists are going to be busy for a very long time, says Forest: “Nobody’s ever built anything like this.”


P. Olson. Experimental dynamos and dynamics of planetary cores. Ann. Rev. Earth Planetary Sci. Vol. 41, 2013. doi:10.1146/annurev-earth-0502120124033. [Go to]

S.A. Triana, D.S. Zimmerman and D.P. Lathrop. Precessional states in a laboratory model of the Earth’s core. J. Geophys. Res. Vol. 117, 2012, B04103. doi:10.1029/2011JB009014. [Go to]

D.P. Lathrop and C.B. Forest. Magnetic dynamos in the lab. Physics Today. Vol. 64, July 2011, p. 40. doi:10.1063/PT.3.1166. [Go to]

E.J. Spence et al. Observation of a turbulence-induced large scale magnetic field. Phys. Rev. Lett. Vol. 96, February 10, 2006, 05502. doi:10.1103/PhysRevLett.96.055002. [Go to]

Further Reading

A. Witze. Geomagnetic field flip-flops in a flash. Science News. Vol. 178, September 25, 2010, p. 10. [Go to]

L. Grossman. Geophysicists push age of Earth’s magnetic field back 250 million years. Science News. Vol. 177, March 27, 2010, p. 12. [Go to]

S. Perkins. Eddies in the deep Earth. Science News Online, May 18, 2008. [Go to]

Get Science News headlines by e-mail.

More from this issue of Science News