Geoscience Reference
In-Depth Information
If the inner core froze out of the outer core,
then the light alloying element may have been
excluded from the inner core during the freezing
or sedimentation process. An inner core grow-
ing over time could therefore cause convection in
the outer core and may be an important energy
source for maintaining the dynamo. The possi-
bility that the outer core is below the liquidus,
with iron in suspension, presents an interesting
dynamic problem. The iron particles will tend to
settle out unless held in suspension by turbu-
lent convection. If the composition of the core is
such that it is always on the iron-rich side of the
eutectic composition, the iron will settle to
the inner-core--outer-core boundary and increase
the size of the solid inner core. Otherwise it will
melt at a certain depth in the core. The end
result may be an outer core that is chemically
inhomogenous and on the liquidus throughout.
The effect of pressure on the liquidus and the
eutectic composition may, however, be such that
solid iron particles can form in the upper part of
the core and melt as they sink. In such a situa-
tion the core may oscillate from a nearly chem-
ically homogenous adiabatic state to a nearly
chemically stratified unstable state. Such com-
plex behavior is well known in other nonlin-
ear systems. The apparently erratic behavior of
the Earth's magnetic field may be an example of
chaos in the core, oscillations being
controlled by nonlinear chemistry and
dynamics
.
Since the outer core is a good thermal con-
ductor and is convecting, the lateral tempera-
ture gradients are expected to be quite small. The
mantle, however, with which the outer core is
in contact, is a poor conductor and is convect-
ing much less rapidly. Seismic data for the low-
ermost mantle indicate large lateral changes in
velocity and, possibly, a chemically distinct layer
of variable thickness. Heat can only flow across
the core-mantle boundary by conduction. A ther-
mal boundary layer, a layer of high temperature
gradient, is therefore established at the base of
the colder parts of the mantle. That in turn
can cause small-scale convection in this layer if
the thermal gradient and viscosity combine to
give an adequately high Rayleigh number. It is
even possible for material to break out of the
thermal boundary layer, even if it is also a chem-
ical boundary, and ascend into the lower mantle
above D
. The lateral temperature gradient near
the base of the mantle also affects convection
in the core. This may result in an asymmetric
growth of the inner core. Hot upwellings in the
outer core will deform and possibly erode or dis-
solve the inner core. Iron precipitation in cold
downwellings could serve to increase inner-core
growth rates in these areas. These considerations
suggest that the inner-core boundary might not
be a simple surface in rotational equilibrium.
The orientation of the Earth's spin axis is con-
trolled by the mass distribution in the mantle.
The most favorable orientation of the mantle
places the warmest regions around the equator
and the coldest regions at the poles. Insofar as
temperatures in the mantle control the temper-
atures in the core, the polar regions of the core
will also be the coldest regions. Precipitation of
solidironisthereforemostlikelyintheaxial
cylinder containing the inner core.
Formation of the inner core
There are two processes that could create a solid
inner core. (1) Core material was never completely
molten and the solid material coalesced into the
solid inner core, and (2) the inner core solid-
ified due to gradual cooling, increase of pres-
sure as the Earth grew, and the increase of melt-
ing temperature with pressure. It is possible that
both of these processes occurred; that is, there
was an initial inner core due to inhomogenous
accretion, incomplete melting or pressure freez-
ing and, over geologic time, there has been some
addition of solid precipitate. The details are obvi-
ously dependent on the early thermal history, the
abundance of aluminum-26 and the redistribu-
tion of potential energy. The second process is
controlled by the thermal gradient and the melt-
ing gradient. The inner core is presently 5% of the
mass of the core, and it could either have grown
or eroded with time, depending on the balance
between heating and cooling. Whether or not the
core is thermally stable depends on the distribu-
tion of heat sources and the state of the mantle.
If all the uranium and thorium is removed with
the refractories to the lowermost mantle, then
the only energy sources in the core are cooling,
