Early in its formation, Earth had no inner core. Over time, the inner core
has grown into a region of dense, almost pure solid iron that spans more than
2,400 kilometers. Seismology has enabled scientists to estimate its size and
some properties based on the way seismic waves travel from the core to the surface.
But, for more than 15 years, seismologists have struggled to understand a strange
property of P waves called seismic anisotropy: When an earthquake occurs, elastic
waves travel through Earth's iron inner core faster in the north-south direction
than in the east-west direction. To explain these large-scale seismic observations,
researchers are investigating iron on the molecular scale. Two studies in the
Sept. 6 Nature examine iron's high-temperature crystalline structure to shed
new light on anisotropy, and, in turn, the origin and formation of Earth's inner
Until now, scientists have been unable to study the properties of iron at very
high temperatures and pressures. Simultaneously creating these extreme conditions,
in the lab or in theory, has proved difficult, and researchers have been unable
to effectively study seismic anisotropy. Earlier attempts at simulating iron
elasticity ignored temperature effects entirely by setting the temperature to
A team of geophysicists has now found a method to theoretically model these
properties at high temperatures using a quantum-mechanical model of the electrons
in iron. The model is similar to the method used in past low-temperature studies,
but also takes into account the thermal vibrations of the atomic particles,
which are much more active at high temperatures than at absolute zero.
Gerd Steinle-Neumann and Lars Stixrude from the University of Michigan, Ronald
Cohen from the Geophysical Laboratory of the Carnegie Institution of Washington
and Oguz Gülseren from the National Institute of Standards and Technology
and the University of Pennsylvania used their new method to calculate the elastic
constants for hexagonal close-packed (h.c.p.) iron at temperatures between 4000
and 7000 K. Steinle-Neumann says the h.c.p., which makes up most of the inner
core, acts like "a Coke can with an ideal ratio. When you heat it, the
base shrinks and the sides slowly increase," changing the rigidity of the
crystalline prisms. As the sides lengthen and bases shrink, the hexagon is looser
and easier to disturb in the long direction, and more rigid and difficult to
disturb in width.
Using these newly calculated elastic constants, Steinle-Neumann and his co-workers
created a general model of inner core structure in which, under high temperatures,
the hexagonal bases preferentially align with Earth's rotational axis, matching
the observed anisotropy. "The changes were unexpected from a material physics
point of view, and with these new elastic constants, we can reconcile apparent
difficulties we've had in the past," Steinle-Neumann says.
Bruce Buffett at the University of British Columbia says "things completely
reversed" with elastic constants calculated at high pressure and temperature.
"It meant all the other explanations presented on crystal structure were
in question," he says. With the new elastic constants in hand, Buffett
and Hans-Rudolph Wenk at the University of California at Berkeley modeled the
texture of many iron crystals in the inner core. Texture is the collective evolution
of the crystals' orientation, and is either intrinsic to the crystal growth
process, or has outside forces at work.
In their Nature article, Buffett and Wenk propose a new forcing mechanism on
the crystals: electromagnetism acting on the inner core from the generation
of Earth's magnetic field in the outer core. They generated a physical and numerical
model to describe the crystals' deformation process, or flow under magnetic
stress. The model shows the step-by-step evolution of an aggregate of 500 crystals
growing from an initial random orientation.
Calculations of the average seismic wave speeds using the texture model, and
the new elastic constants of Steinle-Neumann and others, produce an anisotropic
difference that agrees with seismic observations. Buffett says the new modeling
"may be a first step" in understanding Earth's inner core dynamics
"We generally only get a snapshot of the present day inner core. On the
other hand, surface geologists can infer historic attributes from a rock,"
Buffett says. He says that if scientists unravel where anisotropy comes from,
they can begin answering questions about when the core formed and how fast it
is growing. "We are just trying to find a logical explanation for this
Lisa M. Pinsker