Personal tools
News from ICTP 92 - What's New
Italian researchers have determined that the Earth's core
may be much cooler than previous estimates suggested, possibly
confirming what Jules Verne anticipated nearly 150 years ago in
his Journey to the Centre of the Earth.
Core Values
While we know with great accuracy the temperature of the Sun and, in fact, of many celestial bodies several light years away, we have yet to accurately determine the temperature of the Earth's core, a 'bare' 3,000 kilometres (km) beneath our feet.
With remarkable foresight, Jules Verne, in Journey to the Centre
of the Earth (1864), maintained that the Earth's core is solid--an
opinion that contrasted sharply with those of most scientists
in his day. The latter contended that the Earth's core is comprised
of hot volcanic liquid.
In the early twentieth century, more than 50 years after Verne
published his classic science fiction travel story, seismic data
gathered by scientists indicated that the Earth's nucleus consists
mainly of iron in two states--a liquid layer about 2,200 km thick,
surrounded by a solid core of about 1,200 km in radius. This solid
core, foreshadowed by Verne, guarantees that the core's inner
temperature cannot exceed iron's melting temperature.
The melting temperature of iron under ordinary conditions is well
known--about 1,500°C. But the temperature at pressures found
in the Earth's core, of the order of 3 million atmospheres, is
much more open to question.
To reproduce in the laboratory such extreme pressure and temperature,
researchers at Lawrence Livermore National Laboratory, in California,
created impacts with special ultra-high-speed 'gunshots' containing
bullets travelling at 10 km per second. Others, including Soviet
scientists in the 1960s, conducted experiments in the immediate
vicinity of nuclear explosions. These violent methods led to estimates
that the core's melting temperature was most likely below 7,000°C.
But the results were never free of uncertainties.
In an effort to remove these uncertainties, Alessandro Laio, Guido
L. Chiarotti, Sandro Scandolo and Erio Tosatti, of the International
School for Advanced Studies (SISSA), Italian National Institute
for the Physics of Matter (INFM), and Abdus Salam International
Centre for Theoretical Physics (ICTP), and Stéphane Bernard
of the French Commisariat à l'Energie Atomique,
chose an alternative approach that is perhaps more elegant and
surely less dangerous.
Instead of trying to reproduce for real the Earth's core conditions
in the laboratory, they sought to reproduce the conditions--or,
more precisely, simulate them--in the computer by solving in real
time the fundamental equations governing the motion of individual
iron atoms. The simulations have provided surprising results,
as the authors of the research revealed in the 11 February edition
of Science.
The conclusions suggest that the Earth's core is somewhat cooler
than previous estimates--a 'mere' 4500-5000°C or 2000°C
cooler than prevailing projections. The heat liberated during
crystallisation is also more modest, about one-half the previously
accepted value.
This conclusion implies, among other things, an increase in the
speed at which the Earth's solid core grows at the expense of
the liquid layer above it. That, in turn, may reduce the time
expected for the entire core to solidify and begin to cool down.
Moreover, the simulations reveal that the theoretical speed of
transverse sound in solid iron at Earth's core is anomalously
low, and essentially identical to that accurately measured for
real seismic waves.
As a result, we could hypothesise, as Ronald Cohen of the Geophysical
Laboratory in Washington, D.C., and Lars Stixrude of the Georgia
Institute of Technology in Atlanta, Georgia, have done, that the
Earth's solid core may be compared to a single iron crystal of
cyclopic dimensions, a unique gigantic gem confined forever by
geological history to the deepest recesses of our planet.
Erio Tosatti
Erio Tosatti is a professor of physics at the International School
for Advanced Studies (SISSA) and a long-time consultant with ICTP's
Condensed Matter Group.
For more detailed information, see Physics of Iron at
Earth's Core Conditions in Science 287 (11 February
2000), pp. 1027-1030.