Cornell Center for Materials Research

Ask A Scientist!

Previous Week's Question Published: 20 June, 2001 Next Week's Question
Quakes reveal the earth's core
Question
How hot is it at the earth's core? Does the heat affect our temperature?

Question
You ask a difficult question! We can't go down to the core (it is a depth greater than the distance from Ithaca to Dallas), so we have to be clever. Let's start with a fact: the core is metallic iron--a reasonable choice, as iron is dense, can generate a magnetic field, and is an abundant element. How does this help us determine how hot the core is? Let's use a second fact: the outer core, from about 2885 to 5154 kilometers is molten while the inner core from about 5154 to 6371 kilometers is solid. We know this because of the special behavior of two types of seismic waves (sound waves in the earth). P waves pass through both solids and liquids, but S waves pass only through solids not liquids. When there is an earthquake on one side of the earth, the seismic waves reaching the other side can be used to "see" what the interior of the earth is like. One of the most striking discoveries was that S waves do not pass through the outer core, meaning that it is a liquid. But weak S waves generated at the surface of the inner core when the P waves hit do travel through the inner core. Thus, the inner core is solid.

Because we can calculate the core's pressure, the temperature question is answered by determining the melting point of metallic iron at the pressure of the inner-outer core boundary. To do this, we must worry about the purity of the iron as impurities lower the melting temperature. We know that there is a little nickel--iron meteorites, which come from the cores of fragmented planets in the asteroid belt have about 10%. We suspect that the other impurities in the outer core are mostly sulfur and oxygen.

How can we determine the melting temperatures of these metals at core pressures? One of the most useful devices for doing this is a diamond anvil cell. In it, we can squeeze a small sample between two diamonds up to pressures as high as the earth's core and then heat the sample by shining a powerful laser beam right through one of the diamonds to see if the sample melts. Doing it isn't so hard. What is hard is knowing the temperature accurately and knowing for sure that what you see is really melting. It has taken years to perfect this technique but now it is able to give us some pretty good melting temperatures of the materials that we believe make up the inner core, outer core, and mantle, the rocky layer over the core that makes up most of the earth. When we combine all of this information, we come up with a temperature of about 3550C (6400F) at the core-mantle boundary and 4500C (7200F) at the inner-outer core boundary. This is not the final answer - geologists are still working on the problem - but it seems to be the best answer we have right now.

The heat from the earth's interior is escaping through the surface of the earth. It is small compared to the heat from the sun's rays but it can be measured and it is consistent with other measurements like those described above.

 
Edited on: 19 June 2007 2:37 pm