Potassium in the Earth's core

The Earth's magnetic field is generated in its liquid outer core: a shell of liquid iron mixed with a small percentage of nickel and of a number of light impurities like oxygen, silicon and sulphur, which extends between ~ 1200 km and ~ 3500 km from the Earth's centre. The field protects the Earth's atmosphere from the solar wind, and has been present on the planet for over 4 billion years. Its intensity fluctuates, and from measurements taken since 1840 we know that it has decreased by ~ 10%. It is also known from the study of magnetisation in minerals found in ancient clay pots that the magnetic field in Roman times was about two times stronger as it is now. This drop in intensity has mounted speculations that we may be heading towards a reversal, a stage in which the intensity of the magnetic field is usually very low, increasing the chances of magnetic storms. Modelling the Earth' magnetic field with the goal of predicting its evolution is one of the most exciting problems in the Earth sciences, although still elusive.
The mechanism responsible for the generation of the field is known as the geo-dynamo, and runs on thermal and compositional convective currents in the liquid outer core. Compositional convection is due to expulsion of light impurities, either from the inner core as it freezes (the same mechanism by which sea icebergs are made of fresh water), or possibly by exsolution of magnesium or silicon dioxide as their solubility drops on core cooling, as recently proposed. Thermal convection relies on a temperature gradient between the bottom and the top, and is sustained by the excess energy that is not lost by conduction. It could be viewed as a thermal engine, whose efficiency depends on the thermal conductivity. The total energy provided to the Earth's liquid core comes from various sources, including the freezing of the inner core with release of latent heat, secular (primordial) cooling, gravitational energy due to contraction of the Earth's radius on cooling, and radiogenic heating, believed to be mainly due to potassium, although contribution from uranium and/or thorium cannot be discounted. The main difficulties that we face at present to model the Earth's magnetic field and its thermal history are related with our poor understanding of these energy sources available in the core.
The main aim of this project is to constrain the likely amount of potassium in the Earth's core, to determine if this element could be a significant source of energy for the core. This will be done by providing solubility data for potassium in liquid iron mixtures representative of Earth's core composition and in equilibrium with the mantle, both at the present day and in the early Earth, when most of the prospective amount of potassium in the core would have dissolved. The new data will be based on highly accurate quantum mechanics based simulation methods in conjunction with statistical mechanics techniques, with the help of high performance computers. Solubility data will be established by computing the chemical potential of potassium in liquid iron mixtures and silicate solids and melts. The methods that we will develop and use are completely general, and may be applied to other fields, including for example material design and the study of chemical equilibrium between mixtures.

The direct outcome of the research will benefit the geophysics community by establishing better estimates of the likely abundance of potassium in the core, which will allow the development of more accurate model for the thermal history of the Earth and for more realistic simulations of the geo-dynamo. The recent launch of the "Swarm" mission from the European Space Agency to accurately map the Earth's magnetic field is testimony of the importance and timeliness of this problem.

The newly developed computational techniques will benefit the condensed matter and material science community, which will be able to port these techniques into material design and nanotechnology projects.

Our involvement with the Numerical Group Algorithm Ltd will contribute to the optimisation of computer codes on national facilities.

Our collaboration with the industrial company AWE will export the research to the real economy.

We will give the general public a better description of what the interior of our planet looks like and what are the energy sources that keep its dynamic processes running.

More generally, I believe that our work has the potential to inspire the younger generations, and attract them to science and the solution of scientific problems using high performance computing, an area where the U.K. is at the leading edge.