Geology Reference
In-Depth Information
volume. First, a giant pair of metal plates squeezes from above and below with awesome
force that can reach thousands of tons. Those massive plates exert a viselike grip on the
clever second stage, which consists of six curved, interlocking steel anvils—three above
and three below—which in turn press uniformly from all sides against a third stage with
a cube-shaped cluster of eight tungsten carbide anvils. The sample of powdered mineral
plus water must be tightly encased inside the innermost fourth stage, often with a gold or
platinum linersothatthereactants don'tsquirtoutthesides.Asifgenerating pressureisn't
hard enough, the sample must also be baked with electrical heaters buried deep inside the
sampleholder,andthetemperaturehastobemeasuredcontinuouslywithadelicateloopof
special wire called a thermocouple.
Another popular experimental approach to simulating Earth's deep interior is the dia-
mond anvil cell, which generates extreme pressures by squeezing two diamonds with flat
tips together. First take two typical half-carat brilliant-cut gem diamonds, just like the
stonesinoldweddingrings,andpolishdownthesharppointsonthebottomtoflat,circular
surfaces a twentieth of an inch across—what will become the anvil faces. Then mount the
diamonds in a precisely aligned metal vise, and between them place a thin piece of met-
al with a tiny hole punched out. Center the hole over the opposed diamond anvils, load it
withwaterandmineralpowder,andsqueeze.Amodestforceonthediamondscreatesatre-
mendous pressure because the anvils are so small and thus concentrate the force. Diamond
anvil cells have sustained record high pressures equal to the three million atmospheres
found in Earth's inner core. The beauty of the diamond anvil cell is that you can see your
pressurized sample by looking through the transparent gems. A battery of analytical spec-
troscopic techniques can be brought to bear, and it's easy to heat the sample to mantle con-
ditions with a high-powered laser, which can also shine through the transparent diamond
anvils.
Ifeverythinggoeswell—ifthedesiredpressuresandtemperaturesarereachedandmain-
tained, if the thermocouple doesn't break, if the sample doesn't leak—then the tricky ana-
lyticaltasksbegin.Somewater-bearingmineralslikeclaysandmicasareeasytorecognize,
but how do you measure a few parts per million of water in an otherwise dry sample? The
ion probe is one option; its high sensitivity and spatial resolution led to Erik Hauri's dis-
covery of trace amounts of water in lunar volcanic glass. Infrared spectroscopy, which can
reveal characteristic bonds between oxygen and hydrogen, is another useful tool. Newly
formed bonds between hydrogen and oxygen alter the way that infrared radiation interacts
with a crystal—changes that can reveal water entering the mineral structure. Nevertheless,
cautious colleagues (and wary rivals loath to be scooped) will always raise the possibility
that experiments are flawed or analytical techniques too insensitive. A single fluid inclu-
sion—a minute pocket of water too small to see with a microscope—can yield a false sig-
nal in such finicky measurements.
