Our understanding of the interior structure and composition of the earth and its thermal characteristics is limited by the fact that we can’t access most of it directly, and so most of what we think we know about the subject is inferred and subject to revision. The deepest drill hole only penetrates about 10 km which leaves us thousands of kilometers short of the center. Due to its inaccessibility we must of necessity rely upon mainly indirect methods to discern its properties, and this leaves plenty of room for speculation and reinterpretation of the narrative in the light of new information, …of which there is plenty. Matters are further complicated by trying to distinguish between processes that occurred during the development of the planet over time to bring it to its present state and what is currently happening down there. What follows is a slightly haphazard review of what we think we know about the world beneath our feet.
Taking the earth overall, it is thought that 90% of the Earth’s mass is composed of iron, oxygen, silicon and magnesium.
The two main regions of the earth’s interior are the central innermost region, the core, and an outer layer called the bulk silicate earth (BSE), The BSE includes both the mantle (99%) and the crust (1%). Both core and mantle are, to varying degree, high temperature/pressure environments where strange elemental forms and associations can exist. The core has a radius of about 3850 km, which although more than half of the total radius, constitutes just 15 percent of the Earth’s volume. The radius of the whole earth is about 6350 km, with the added 2500 km from core/mantle boundary to the surface making up 85% of the volume.
Our understanding of the interior is based on various forms of remote sensing and inference including density considerations, corresponding meteorite composition, seismic data, electromagnetic and geological observations together with high temperature/pressure experiments on various elements and compounds.
Due mainly to the magnitude of the bulk earth density, 5.5 g/cm3 , …about twice the density of the crust and mantle, and the composition of some iron-rich meteorites, the central core is thought to consist mainly of metallic iron (Fe), together with some nickel (Ni), silicon (Si) and sulfur (S). Recent suggested additions include hydrogen, oxygen and oxides of magnesium or silicon or even hydrides of the heavier elements, (see below).
Although even stony meteorites contain 10-30% iron (rare lunar meteorites would be an exception), about 5% of located meteorites have a much higher nickel-iron alloy content. Contrast this proportion with the estimated earth’s core volume of 15% which suggests that core type material may be underrepresented in the meteorite population.
The core is in turn divided into an inner and outer region, the inner core being solid while the outer core is thought to be mobile with convection currents radiating away from the center. These convection currents in the outer core (which are powered by exothermic or heat producing activity such as radioactive decay or even oxidation) are thought to be the main source of that feature of the core with which we are most familiar, i.e. the magnetic field that extends well beyond the surface of the earth and without which the surface would be much less habitable.
From a chemical point of view, the core elements are generally present in their reduced form which means that they have neither a shortage or excess of electrons. This implies that the elements coexist as a metal alloy which is a mixture of the metallic components. In a metal the electrons are shared between many atoms and these electrons can undergo something called gapless excitations (in a gapless system the energy state of the electron can change continuously, …like kinetic energy or even electrical energy) and by which means the metal will act as a conductor. This contrasts with elements in the form of chemical compounds where some electrons are shared between the elements to form the molecules that make up the compound. These compounds are generally gapped systems where the electrons are constrained to to discrete energy levels that can store or release chemical energy.
Note that it has recently been suggested that the reduced metallic alloy model may not be the whole story as some metal oxides, either silicon dioxide (quartz) or magnesium oxide may be present and that the ongoing production of the oxides may add to the thermal budget. Just how this oxidation reaction might proceed in the presumed strongly reducing environment of the outer core remains to be explained. Others speculate that some of the metals exist as sulphides (e.g pyrite, FeS2), silicides (e.g. nickel mono-silicide, NiSi) or, if hydrogen in significant amounts, hydrides. This idea seems to be a better fit to the existing chemical model.
As one would expect given the proposed presence of fluid metal, the core is a high energy region, it’s hot. The science is far from settled but popular estimates range from about 3,000 °C for the outer core to around 6,000 °C or more for the inner core. We base these estimates of the core temperatures on the melting behaviour of iron at very high pressures. Core pressure estimates vary from +1 million to around 3.5 million atmospheres (one atmosphere of pressure is also called 1 bar and is around 15 psi) which roughly translates to anywhere between 15 and 55 million psi. At these pressures, iron melts somewhere between about 4,200 °C to about 7,200 °C.
The source of this energy/heat is thought to consist of residual primordial accretion energy, gravitation derived friction generated by descending heavy elements and a current exothermic or energy releasing process, the main contender being radioactive decay, and possibly accompanied by an exothermic chemical process like oxidation. Estimates of the energy contribution of radioactive decay vary widely as there is uncertainty about the amount of radioactive material present in the core. The convection process may also result in electrostatic charge separation similar to charge buildup in a thunder cloud.
As mentioned above, the core, or more specifically, the state and motion of the outer core is thought to be the principal source of the earth’s magnetic field which has two main poles (North and South) like a permanent bar magnet (note that the mantle, crust, oceans, ionosphere and magnetosphere also contribute to the overall field) . But temperature levels in the core would seem to rule out a permanent magnet as the source, …molten metal is generally not magnetic (we believe that for a substance to be magnetic many of the the particles must be aligned, …a difficult feat to achieve for metal in a mobile liquid state). Permanent magnets generally lose their magnetic field at about 500°C (for the technically minded :-), the temperature at which some materials lose their permanent magnetic properties or their ability to be attracted by a magnet is known as the Curie point).
We know of only three ways to create a magnetic field, firstly via these natural or man made permanent magnets, next, and this is the favoured dynamo model, by utilizing an electrical circuit where the current, in addition to creating an electric field, creates a magnetic or B field. The relative stability of the earth’s magnetic field suggests a similarly stable, solenoid type circuit. The third possibility involves a state of matter known as plasma where atoms experience some degree of charge separation which creates both unbound electrons and ions. This plasma state produces its own static? magnetic field which will induce currents in molten material. In addition although the magnetic field is stable over shorter time scales, we observe that the poles wander over time and sometimes reverse polarity. This dynamic nature of the field suggests that it is current/circuit related. We also find the field to be asymmetric. However this characteristic doesn’t readily lend themselves to a rotating or convecting liquid outer core model.
As discussed above, it is likely that some proportion of the elements of the alloy are in an ionized, plasma-like state, where one or more electrons are separated from their atoms. These relatively unbound electrons and the corresponding electron deficient particle result in dynamic charge separation of the moving atoms which creates energy gradients/transmissions, which in their turn produce a dynamic magnetic field, which then induces eddy currents in the surrounding material, (and the possibility of superconducting conditions at some scale cannot be ruled out, metallic hydrogen is a superconductor). A picture emerges of a chaotic electrical milieu. This is of more than academic interest as the small scale charge separated plasma component of the alloy (and the resulting electromagnetic/electrostatic mismash) is, as mentioned above, thought to generate the large scale magnetic field that pervades the solid earth and near space. Somehow or other, a small scale charge separation process is able to generate a global field. As mentioned above, this core related magnetic field and the associated magnetosphere is one of the components that makes the surface of the earth habitable as it deflects most of the harmful high energy cosmic rays that arrive from the sun and the wider cosmos. It is somewhat ironic that, far from being a somewhat distant and irrelevant hot blob :-), life on the surface depends in part on the behaviour of the distant and only indirectly accessible core.
Other more remote possibilities for the source of the field include, a. the presence of a permanently magnetic section of the core that is a remnant of past events or b. present day electrical activity above or below the earth’s surface or c. the convecting/moving fluid may induce electrostatic charge separation similar to charge buildup in a thunder cloud. Static charge accumulation does not generate a magnetic field but dielectric failure will create a circuit of plasma which will then discharge and in the process generate a magnetic field.