News Release

Evidence for potassium as misisng heat source in planetary cores

Peer-Reviewed Publication

University of Minnesota

MINNEAPOLIS / ST. PAUL--There's a small problem with Earth's magnetic field: It should not have existed, as Earth's rock record indicates it has, for the past 3.5 billion years. Motions in the Earth's molten iron core generate convection currents--similar to boiling water--which produce the field. Many sources of heat drive these currents, but the known sources seem inadequate to maintain the field this long. In 1971 University of Minnesota geology and geophysics professor Rama Murthy theorized that radioactive potassium in the core could supply additional heat, but researchers investigating that claim have been unable to obtain reliable experimental data. In a paper to be published Thursday (May 8) in Nature, Murthy presents experimental evidence for his idea and shows why other researchers have been unable to corroborate it.

The work helps explain how Earth has maintained its magnetic field, which shields the planet from harmful cosmic rays and the constant stream of charged particles from the sun known as the solar wind.

"Earth is losing energy from its surface at a rate of about 44 trillion watts," Murthy said. "About 75 percent is heat from the mantle [the middle layer, composed of rock], and 20 to 25 percent is heat from the core. Measurements of cooling at the core-mantle boundary show too much loss for a core to maintain heat and a magnetic field for 3.5 billion years." But if radioactive elements such as potassium, and perhaps uranium and thorium, also exist in the core, the heat from their radioactivity could keep the core hot enough to move and maintain the magnetic field, he said.

Earth's core is believed to consist of metallic iron and iron sulfide. Soon after the planet coalesced from a cloud of hot gas and dust 4.5 billion years ago, the core was liquid; since then it has cooled to the point where about 10 percent is solid. According to Murthy, a radioactive isotope of the element known as potassium-40 could have been incorporated into Earth's core as it formed. Some scientists have doubted this because potassium is not found in any metal ores at the surface. However, said Murthy, the core is not pure iron. The presence of sulfur (as iron sulfide) could have allowed potassium to dissolve in the original core material. Potassium-40, with its half-life of 1.3 billion years, could have supplied enough radioactive heat to keep the core hot enough to maintain the magnetic field for billions of years.

A hot core will dissipate its heat one way or another. If it's only a little hot, the heat will be conducted in a smooth manner into the surrounding mantle, just as heat from warm water will smoothly pass to the container and the air. But if it's too hot, the heat energy--as in a pot of boiling water--drives convection currents in the core, turning it into a gigantic magnetic generator.

No one can directly study the Earth's core, but scientists can subject samples of core- and mantle-like material to tremendous heat and pressure in the laboratory. When Murthy and his colleagues subjected samples of iron, iron sulfide and potassium-bearing silicate rock to temperatures and pressures similar to those found at some depth in Earth's mantle, they found that a significant amount of potassium moved from the silicate "mantle" into the metallic iron-iron sulfide "core." Extending the results to temperatures and pressures existing at the actual Earth's core indicated that sufficient potassium could end up in the core to supply the missing heat.

When previous researchers measured movements of potassium into "cores," their data was too scattered to draw any conclusions. The problem, said Murthy, is that standard procedure calls for the sample material to be polished with oil for analyzing its potassium content. But the presence of oil, he found, causes a rapid loss of potassium from the sample. Instead, Murthy polished his samples dry, using boron nitride powder. The potassium stayed in the samples and produced reliable data.

Because it is white, the boron nitride powder caused a minor difficulty. "I was afraid to travel with it, because I obtained it from a company the same week the anthrax scare happened," said Murthy. "So I shipped the powder to our Washington laboratory." Murthy and his colleagues performed the work at the Geophysical Laboratory of the Carnegie Institution of Washington, D.C.

The next step, says Murthy, is to study how potassium moves into core material at higher temperatures and pressures, closer to conditions deep in Earth's mantle. He said similar experiments should be done to see whether uranium and thorium could also have moved into the core during the planet's formation. As the sources of Earth's inner heat get sorted out, Murthy said the new knowledge will refine ideas about how continents drift. Continents are part of Earth's crust, or thin outer layer. They are thought to move around by convection currents in the mantle, but whether those currents exist in the whole mantle or just the upper part is still an open question, Murthy said.

"We can also simulate the temperature and pressure in the core of Mars because the planet is much smaller than Earth," said Murthy. "Mars has more sulfur in its core than Earth, so it ought to have collected more potassium." The Red Planet once had a magnetic field, but it was lost. The probable culprit is the small size of Mars, which caused it to lose too much heat to maintain convection in its core, said Murthy. He said Mars' core may be entirely liquid now, losing heat just by conduction, rather than the convection needed to produce a magnetic field.

###

Murthy's colleagues were Wim van Westrenen and Yingwei Fei of the Carnegie Institution. Van Westrenen is also affiliated with the Institute for Mineralogy and Petrography in Zurich, Switzerland. The work was supported by the University of Minnesota, NASA, the Carnegie Institution and the Swiss National Science Foundation.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.