In the Nova episode "Origins: Earth is Born", Neil DeGrasse Tyson states, "But Mars is just a fraction the size of the Earth, so it cooled more rapidly. And as it cooled, its molten iron core hardened. As a result, Mars stopped generating its magnetic shield." (February 28, 2004)

see: http://www.pbs.org/wgbh/nova/space/origins-series-overview.html#origins-earth-born

This comment from Tyson is consistent with what I have long believed. But I wonder.

The melting point of iron (Fe) is around 1538 °C.

see: https://en.wikipedia.org/wiki/Melting_points_of_the_elements_(data_page)

The the geothermal gradient for Mars is estimated to be around 6 °C per km.

see: What is the temperature 55 km beneath the surface of Mars?

If the estimate of 6 °C per km is the ballpark, then I would expect that iron is molten as a mere 256 km of depth. At the core of Mars (over 3000 km in depth) I would expect the temperatures to be far higher than 1538 °C.

It seems to me that it is impossible for iron to be in solid form at Mars' core.

  • 1
    $\begingroup$ @zephyr This is a topic that is not trivially researchable by a layperson: adsabs.harvard.edu/abs/2015EGUGA..1715240D $\endgroup$
    – called2voyage
    Commented Dec 14, 2016 at 20:41
  • 3
    $\begingroup$ Your analysis makes the fatal mistake of applying a linear extrapolation well beyond the realm where that linear model is valid. That said, there are solid reasons based on how satellites orbit Mars to think that Mars does not have a completely solid core. $\endgroup$ Commented Dec 15, 2016 at 8:03

4 Answers 4


Mars core may or may not be solid, I'm of the opinion that it's not solid.

See recent NASA press release here

and a much earlier study here

It has been known since 2003 that at least part of Mars’ interior is molten, based on how easily the Sun’s gravity distorts the planet’s shape, but no one knew whether it is completely liquid, or whether there is a solid inner core like Earth’s.

and Nasa's 2003 announcement here

on Neil deGrasse Tyson, link here, but I'm not sure it's from him, so . . I might delete later.

I've heard (but should provide a source) it said that Mars core has solidified. I remember NdGT saying that, but if he says it on a show, it's really the show saying, not him. Anyway, I'll try to get that point worked out.

But, these seemingly conflicting statements might still have some relation (I think). A basic explanation is that the difference between liquid and solid under the pressures you get inside a planet are somewhat inexact. The other thing is, to generate a magnetic field, you need more than just a liquid metal around a solid metal, but you need dynamic spiraling. See here and here.

Venus rotates very slowly, this rotation means that Venus has essentially zero Coriolis effect, which might help it generate a permanent magnetic field. Earth's relatively rapid rotation and Coriolis effect, and perhaps it's very gradually slowing rotation due to the Moon, help Earth maintain it's magnetic field. A very slow variation in the rate of rotation of the solid core and outer solid mantle could generate enough spin to maintain the magnetic field.

See here on the Coriolis force and here on the moon's possible role on maintaining the Earth's magnetic field.

The point of all this, planetary magnetic fields are complicated but they likely require the right type of permanent flow within the liquid part of the core, probably around a solid iron core. Earth maintains that perhaps in part due to higher core temperatures which enable more easy flowing as well as the release of heat through condensation as the outer core solidifies onto the inner core and perhaps also, due to the Moon. Mars at one time in it's past probably had a magnetic field similar to or perhaps even stronger than Earth, evidenced by some powerful crustal fields that remain. See here.

I don't know why he uses the word "solidified" when other research says liquid. That might be a good question to ask him, but, Neil deGrass Tyson may be right that the cooling of the smaller Mars may have caused it to lose it's magnetic field, but based on article's I've read, he's incorrect to imply it solidified, but the liquid core may have become solid enough that the dynamics that could have once caused a magnetic field could have stopped, at least, I would speculate that this is possible but I don't think anyone knows for 100% certain. The history of planetary magnetic fields as well as the precise makeup of planetary cores and what turns on or off a magnetic field inside a planet isn't well understood.

and Zepher's answer is also important that the temperature of a liquid iron core at pressures inside of a planet are very different than the melting temperature at 1 ATM. Earth's inner core is hotter than it's outer core but the greater pressure keeps it's inner core solid while it's outer core is liquid.

  • 1
    $\begingroup$ Re I assume this is a direct quote, see here. I would assume the exact opposite. That blog appears to be written by Greg Bentall rather than Tyson, Einstein, Drake, or Sagan (also tagged in that blog post). Bentall appears to be mixing and matching what's he's read, possibly with a layperson's misunderstanding, and possibly some now out-dated stuff. Note well: given other stuff I've seen from Tyson, I truly doubt he's a proponent of the rare earth hypothesis. (Compare with the bulk of that blog post.) $\endgroup$ Commented Dec 15, 2016 at 7:49
  • $\begingroup$ @DavidHammen quote removed. I left the link there for now. Will try to clean up a bit later. $\endgroup$
    – userLTK
    Commented Dec 15, 2016 at 13:16

You're mostly missing an important point to do with the geothermal gradient.

The geothermal gradient used in that question, and applied to a $55\:\mathrm{km}$ depth is not consistent all the way through Mar's interior. They cite the Earth's geothermal gradient (see image below) and from that page you can see how the temperature profile changes with depth. There is a much stronger gradient towards the surface. Thus you can't naively apply the surface geothermal gradient through to the core. Rather, to get the core temperature, you'd need to integrate the gradient function.

Earth's Temperature Profile Earth's Temperature Profile


The absence of a magnetic field is not sufficient evidence to conclude that a planet has a solid core. It may be that the interior of Mars is liquid but lacks convective currents vigorous enough to generate a magnetic field.

see: http://adsabs.harvard.edu/abs/2015EGUGA..1715240D

NASA has scheduled a mission to Mars (for May of 2018) which is designed to give us new information about the deep interior.

see: http://insight.jpl.nasa.gov/overview.cfm

Also, in conceptualizing a geothermal temperature gradient for Mars (or any other planetary body), the relationship between depth and temperature is not linear. The curve will also have variances according to the composition of the layers, which is currently not well known. Furthermore, eutectic properties of the Martian lithosphere could result in unexpected melting points.

Therefore, knowing the radius of a planetary body is also not necessarily sufficient evidence to conclude that it has a molten core.


Scientists cannot theorize clearly if minerals and elements are arranged in plumes and strata inside mars, and the diameters that make up the martian condensed/fluid inner/outer strata and plumes, they don't know if there is something fluid and metallic [mobile electrically charged plumes of (probably) heavy elements (perhaps including iron, the 4th most common element on our world's surface), and heat from radioactive decay].

They will know more if there is any thermally driven geological movement inside the planet, including molten iron, if they can find any small temporal fluctuations in the magnetic field.

For the moment, a molten core only gives way to theories on "paleomagnetism, a latent magnetic crust of 10-125Km" deep stating that all the magnetism is left over from 4 billion years ago, and that certain zones have been de-magnetized by impact collision.

Meteorites de-magnetize the crust and it may be possible to date the chronology of the weakening molten core of mars from billions of years ago, to know how long ago the molten core was magnetic, and how fast it ceased being as fluid as it was.

Yt is not possible to determine detailed vertical variations of the magnetization of the Martian crust on the basis of magnetic data analysis alone, it is possible to estimate the thickness of the magnetic part of the Martian crust using other independent observations. The magnetic layer is bounded at the bottom by the depth to Curie isotherm of its major magnetic minerals, and at the top by the depth to the base of a near surface zone that has been demagnetized by impact-induced shock waves.

Some studies have demonstrated that the secondary magnetization acquired by the lower crust in the absence of the core dynamo has little contribution to the observed magnetic anomalies.

The measurements of the magnetic crust are the only witness of the inner core's activity, and of it's spherical geometry: Irregular magnetic field on mars

Difficult measures are needed: 3D maps of the magnetic field, seismic data from multiple stations and meteorites, Magnetic basalt/igneous maps of 1/2/3 billion years chronology.

enter image description here enter image description here Some elements like Tg are twice as heavy as iron, melt at 2.5 times higher temps, and are conductive, so they would be much more likely than iron to be solid and to sink inwards if they are not melted into core alloys of metals. It's difficult to know the alloys and stratification of molten/solid materials that happen inside a planet.

The Maximum pressure at the center of mars is about <50 GigaPascals, because earth's core pressure is 330-360Gpa maximum, and mars is 10 times lighter. Here are images where you will see that H20 at the center of mars would need over 800'C to change from ice to water.

Iron needs to be over 3000K to melt inside mars at the very center , so we have no idea, all we know is that the magnetic field of earth is 40 times stronger than that of mars, which is a fairly good measure of the quantity and fluidity of metals inside mars.

Scientists have more questions than answers about earth and mars's core physics, even after many years of seismology.

For our planet, they have been able to determine that the center of the earth is solid iron/nickel, surrounded by a viscous and fluid outer core of similar material. If there is a lot of friction at the core of our planet, in a conductive medium, it is channeled in a N/S arrangement precisely matched with the rotation of the Earth. The Martian core is completely different, because it's magnetosphere is very irregular and lop-sided.

Here are some interesting facts about martian geology: http://www.planetary.org/blogs/emily-lakdawalla/2008/1710.html?referrer=https://www.google.fr/

enter image description here

  • 1
    $\begingroup$ The statement "although they do clearly contain a bit of something fluid and metallic which is able to generate a magnetic field 1/40th as strong as ours" is incorrect. Mars currently has no dynamo, period. That weak magnetic field is the result of a dynamo, but it ceased operating long ago (~ 4 billion years ago). You linked to Emily Lakdawalla's blog, where she clearly writes about the dynamo having shut down billions of years ago. $\endgroup$ Commented Dec 17, 2016 at 23:19
  • $\begingroup$ In addition, the image of water's phase diagram (the last image in the answer) is completely irrelevant. $\endgroup$ Commented Dec 17, 2016 at 23:26
  • 2
    $\begingroup$ @comprehensible I don't think it's correct that Mars has a magnetic field 1/40th of Earths. Mercury's has a weak magnetic field, listed as listed as 0.006 Earth's or about 1/160. Mars is generally considered to have none. It has to be on the scale of thousands or tens of thousands of times weaker than Earth's because it's weaker than Mercury's. I've never seen it specified in a precise ratio. $\endgroup$
    – userLTK
    Commented Dec 18, 2016 at 7:28
  • $\begingroup$ Sorry, am trying to save face on this erroneous answer by confused editing and changes of subject, but i don't know if anyone will be fooled. let's hope that the pretty pictures distract them. $\endgroup$ Commented Dec 18, 2016 at 7:37

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .