A large magma furnace pours a mixture of silicate and metallic lava onto a flat surface, with a pool of semicooled lava flowing. The lava is black, orange, yellow, and silver.
Researchers explored what happens when silicate lava and metallic lava mix—or, as it turned out, don’t. Credit: Soldati et al., 2021, https://doi.org/10.1038/s41467-021-21582-w, CC BY 4.0

In 2018, designs for NASA’s Psyche mission to explore the metal asteroid of the same name were being finalized, awaiting approval to be built in a facility in California. Meanwhile, across the country, Arianna Soldati was conducting large-scale experiments that involved melting metal-rich basaltic rocks in a furnace, pouring the lava out, and studying how it flowed.

“During our last pour of the day, I saw this metal come out and I made the connection,” said Soldati, a volcanologist at North Carolina State University at Raleigh. Asteroid Psyche, thought to be a shard of a failed planet’s metallic core, was likely partially molten at one point.

“We’re trying to imagine what something that we’re not even sure exists could look like.”

“To be honest, I didn’t set out saying, ‘Oh, I’m going to study ferrovolcanism. Let me design an experiment to do that.’ It was more opportunistic than that,” Soldati said. That summer, “the Psyche mission…was definitely fresh on my mind.” What would those flows of molten metal have looked like, Soldati wondered, and what features would they have left behind for us to study?

“We’ve never seen a ferrovolcanic eruption. We’re not even sure if there is one,” Soldati said. “We’re trying to imagine what something that we’re not even sure exists could look like.”

From Out of the Crucible

“Ferrovolcanism is a very recent term—it’s only been around for a few years,” and it refers to volcanism that occurs on a metallic body, Soldati explained. The magma that erupts could be entirely metallic (type I in Soldati’s classification) or some combination of rock and metal (type II).

The first scenario describes what could have taken place on a mostly metallic asteroid like Psyche, but the second scenario might not be unknown on Earth: There’s an open debate among scientists whether the large iron deposit near El Laco volcano in Chile was the result of ferrovolcanism.

Soldati and her colleagues tested a ferrovolcanism scenario involving both silicate rock and metal. Using one of the furnaces at the Lava Project at Syracuse University in New York, the team melted metal-rich basaltic rock in a silicon carbide crucible. Under heat, the crucible degraded and released carbon which combined with oxygen to form carbon monoxide. The carbon monoxide then chemically reacted with the basaltic melt and separated it into a silicate melt and a denser metallic melt. When the crucible was poured out, the silicate melt flowed first, and the metallic one followed (see video below).

YouTube video

The scientists went into the experiment with few expectations. “There have been no previous experiments that we’re aware of with two materials that are so different and trying to get them to flow together,” Soldati said. “We weren’t sure what would happen.”

They found that the metallic flow was about 3 times denser and about 100 times less viscous and traveled about 10 times faster than the silicate flow at the same temperature—all in line with fluid dynamics theories. “But it was surprising how independent the two flows remained,” Soldati said. “The silicate flow started earlier, and then the metallic one followed, but it went underneath the rock flow and did its own thing. There was not a lot of interaction between the two. They remained fairly sharply separated.”

When the metallic flow reached the front edge of the rock flow, it burst out and started flowing freely. “It allowed us to study not only type II but also type I,” Soldati said. “For every experiment, we could get out information on two different types of ferrovolcanism.” On its own, the purely metallic flow was very turbulent: Thin streams separated and braided themselves together like a river delta, and ribbons and beads of metal broke off from the flow completely, only to be subsumed into the silicate flow. The results were published in March in Nature Communications.

A cooled flow of black silicate lava attached to a cooled flow of silver metallic lava with a clear boundary between them
Silicate lava (black) and metallic lava (silver) remained largely separate as they flowed from the lava furnace. Denser and faster flowing metallic lava traveled underneath slower moving and less dense silicate lava and remained unmixed as both cooled. Credit: Soldati et al., 2021, https://doi.org/10.1038/s41467-021-21582-w, CC BY 4.0

“The experiments demonstrate a mingling but inefficient mixing between silicate and metallic lavas,” explained planetary volcanologist Pranabendu Moitra. “It provides better insights to the flow behavior and morphology of dense, less viscous and turbulent metallic melts,” which behave very differently than silicate flow. Moitra, at the University of Arizona in Tucson, was not involved with the study.

Imagining the Unknown

If a ferrovolcano did exist, what would it look like? This experiment could help give a very basic idea. “Metal, because it is very dense and [of] very low viscosity, forms low-relief topography,” Soldati said. “There’s not going to be a tall ferrovolcano. There’s not going to be a Mount Fuji made of metal. The topography is going to be very shallow, with flows that will extend very far away from the vents. And these flows are probably going to be extremely braided, with many tiny channels.”

“These are gorgeous experiments! I would have loved to see them happen,” said Lindy Elkins-Tanton, a planetary scientist at Arizona State University in Tempe and principal investigator of the Psyche mission who was not involved with this research. But as this particular experiment involved both rock and metal, “I’m not confident they apply to Psyche; we can’t think of a circumstance when metal—really, it would likely be sulfide, FeS—and silicate magma would be erupting at the same time. Still, we’ll look for those textures.”

“The point of the study was not to image something. It was really to imagine what a certain landscape could look like, which, I think, is still an important part of science.”

Soldati and her team will be going back to the Syracuse Lava Project later this year to conduct more experiments, including some that will explore purely metallic lava flows with different types of metals and under different flow conditions. “The experimental results could further validate lava flow models,” Moitra said. “It will be interesting to explore the effects of various experimental parameters such as the slope, and the lava and ambient temperature, etc., on the speed and morphology of metallic lava flows as compared to the silicate ones.”

Experiments like these, said Soldati, offer volcanologists the rare opportunity to come up with a theory first and then go out and see whether they were right or wrong. “When the paper was in review, a comment from one of the reviewers corrected the spelling of the title, changing it from ‘Imagining Ferrovolcanism’ to ‘Imaging Ferrovolcanism.’ But the point of the study was not to image something. It was really to imagine what a certain landscape could look like, which, I think, is still an important part of science, especially in volcanology, which is such an observation-based scientific field.”

“We have to put in the imagination work first so we can compare with our observations later.”

—Kimberly M. S. Cartier (@AstroKimCartier), Staff Writer

Correction, 22 April 2021: This article was updated to more accurately define ferrovolcanism and describe the separation and flow of metallic lava.


Cartier, K. M. S. (2021), Imagining what a metal volcano would look like, Eos, 102, https://doi.org/10.1029/2021EO157385. Published on 21 April 2021.

Text © 2021. AGU. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.