快猫短视频

JUST OVER 4.5 billion years ago, the newly formed Earth suffered a traumatic
blow. Another planet, the size of Mars, came careening through space and slammed
into it with an impact that melted our planet鈥檚 young crust and threw dust and
debris sky-high. For almost an hour, the white-hot crash site shone brighter
than the Sun.

Miraculously, the Earth survived. But the smaller planet wasn鈥檛 so fortunate.
Its remnants flew into orbit, where they collected to form a huge Moon, one of
the largest in the Solar System.

Planetary researchers are pretty confident that this is how our Moon was
created. But there are many things about the collision that they don鈥檛
understand. Why did it produce a satellite at all? Why didn鈥檛 the debris simply
fall back to Earth? And why just one Moon rather than several? It is only
recently that researchers have started to answer these questions, by combining a
new look at the chemistry of Moon rock with computer simulations of the impact.
At last we are getting a clearer picture of how it happened. Our Moon, it seems,
was born from vapour.

Before the Apollo astronauts visited the Moon, nobody had much of an idea of
where it had come from. NASA鈥檚 stated science objective for the Apollo missions
was to determine which of three possible explanations for the Moon鈥檚 origin was
right. Had it formed at the same time as the Earth? Was it a large asteroid that
the Earth鈥檚 gravity had caught from the inner Solar System? Or was it a boulder
somehow thrown out from a furiously spinning young Earth?

Within a few months of the first samples of lunar rocks returning to Earth
with the astronauts, it was clear that none of these ideas worked. The Moon鈥檚
composition was too similar to Earth鈥檚 for it to be some captured asteroid. Yet
it was too different from the Earth to be a straightforward chip off the old
block. In particular, there were no traces of water. On Earth, even the driest
desert rocks contain some water bound up chemically. Also, the Moon was too
light to contain nickel and iron, which form the Earth鈥檚 core. At the 1970 Lunar
Science Conference in Houston, researchers were baffled. 鈥淭here was utter
confusion,鈥� says Jay Melosh of the University of Arizona, who was at the
meeting. 鈥淭here were historians of science there to record science in the making
after such a crucial experiment,鈥� he recalls. Yet it ended up as a huge
anticlimax.

Soon afterwards, Bill Hartman and Don Davis at the Planetary Science
Institute in Tucson, Arizona, suggested the impact idea. Perhaps, they said, the
Moon was a mixture of Earth-like and captured material. That would explain the
confusing composition.

Al Cameron of the Harvard-Smithsonian Center for Astrophysics and William
Ward of the Jet Propulsion Laboratory in Pasadena, California, seized on the
idea and immediately tried to see how it measures up against what we know about
the Earth鈥檚 spin. The Moon is moving outwards from Earth鈥� currently at
about the rate that fingernails grow鈥攁nd we can calculate that 4.5 billion
years ago the Moon was much closer, a mere 1/20th as far away as it is today. An
isolated system like the Earth and Moon combined can鈥檛 gain or lose angular
momentum鈥攚hich depends on how fast the masses are rotating and how far
they are from the centre of rotation. So when the Moon first formed, it was
orbiting much faster than it is today, and the Earth was also spinning much
faster鈥攕o fast that a day lasted only four hours.

Cameron and Ward wondered what kind of collision could have set the
Earth-Moon system spinning so rapidly. They calculated that the object could be
anything from a Moon-sized chunk hitting at a glancing blow to a larger,
Mars-sized planet hitting at about 45 degrees from head-on. Both of these seemed
eminently possible in the pinball world of the early Solar System, where there
was plenty of flying rubble left over from building the planets.

It was a neat idea, but not a popular one. 鈥溈烀ǘ淌悠祍 don鈥檛 like
catastrophes because they reek of miracles,鈥� says Melosh. And from the outset
there were doubts about the physics. Most people assumed that the force of the
collision would be able to throw enough debris into orbit to form the Moon. But
Melosh realised this flew in the face of simple physics. Why, he wondered,
wouldn鈥檛 the stuff simply fall back down to Earth?

When a solid object is launched from the surface of a planet, it can follow
one of two possible paths. If its launch speed is less than the planet鈥檚 escape
velocity, it will arc through space in a parabola and then crash back to the
surface. If it is travelling at more than the escape velocity it will fly
outwards on a hyperbolic path. Neither of these paths leads to the sort of
stable orbit now occupied by our Moon. 鈥淚f you only ejected solid material
during a collision, almost nothing would get into a closed orbit,鈥� says Melosh.
He realised that for the Moon to form where it did, there must have been more
than the initial impulse from a collision. Something else must have provided the
debris with the equivalent of a modern satellite launcher鈥檚 final burn.

Falling down

Cameron hit the same snag when he tried to simulate the collision. Working
with Willy Benz, now at the University of Berne in Switzerland, he had developed
a new way to simulate celestial mechanics. For the purposes of their
calculations, Cameron and Benz split the young Earth and the colliding planet
into many small lumps, and worked out the gravitational interaction between
them. They also used computer code originally developed to simulate the stresses
on the metal casings of nuclear weapons at Los Alamos National Laboratory. This
made sense because the core of both the Earth and, presumably, the sister planet
that hit it, were made of iron. But after running a simulation that tracked the
motion of 10,000 particles, Benz and Cameron found it hard to get their debris
into the right orbit. Instead, it kept falling back down to Earth.

The picture changed if the Earth was deformed enough to make its
gravitational field uneven. Benz and Cameron discovered that a glancing blow
from a body three times as large as Mars could squeeze the Earth into a shape
that would keep the debris in space. That seemed to be one problem solved鈥�
except that it led straight to another. A collision like this would leave the
Earth and Moon with far too much angular momentum.

Another possibility was that the Earth was not yet fully spherical because it
was still only partially formed. However, in that case much of the material that
makes up the Earth and Moon today would have accumulated from the Solar Systems
dust disc for millions of years after the collision, leaving the Moon with much
more iron than it actually contains. 鈥淭he Moon is so light that H. G. Wells
thought it was maybe full of holes,鈥� says Melosh, 鈥渂ut it鈥檚 rather that it
doesn鈥檛 have a lot of iron.鈥�

Last year, Erik Asphaug at the University of California, Santa Cruz, and
Robin Canup at the Southwest Research Institute in Boulder, Colorado, performed
simulations like Cameron鈥檚, but with 10 times as many particles. They also
incorporated a clever idea from Melosh about how to get around the problem of
the 鈥渇inal burn鈥�. Melosh realised that vapour generated in the collision could
play a key role in getting material into orbit. The first kick would be provided
by the energy from the collision itself, but a second more prolonged push could
come from the pressure of gas created by vaporising the two planets鈥� crusts. Gas
expanding out from the impact could provide the equivalent of a final boost to
put solid particles into orbit.

In Asphaug and Canup鈥檚 simulations, a Mars-sized planet clips the Earth鈥檚
crust and instantly vaporises many times as much material as the Moon is made up
of today. It also releases shards of solid debris, and enough heat to melt the
crusts of both bodies. As the rock vapour plume rises, it dumps tiny particles
of solid debris in orbit, and eventually begins to condense there itself.
Volatile components like water escape further out into space before they cool
and condense, and their vapour isn鈥檛 dense enough to boost solid particles out
that far. That, Asphaug and Canup reckon, is why the Moon contains so little
water. The simulation ends with plenty of rubble in tight circular orbits, with
just the right amount of angular momentum to create the Moon.

Asphaug and Canup aren鈥檛 sure why their simulation succeeds where others have
failed. They suspect that Melosh鈥檚 insight about vapour is the key, but neither
they nor anyone else knows how to work out how much rock an impact would
vaporise. Instead they have had to make an educated guess of how much rock
vapour there would be, and then add its effect in. Because of this, they can鈥檛
tell whether all or even most of the Moon came from the vapour phase. What鈥檚
really needed is the characteristic 鈥渆quation of state鈥� for planetary crusts.
This equation determines how much of a material will be solid liquid or gas, and
what the pressure will be, given a certain temperature. Melosh has now developed
an equation of state for rock, and Asphaug and Canup plan to include this
equation in their simulations, to find out just how significant the role of the
vapour really is.

Cameron is cautious about these new simulations. He says he won鈥檛 be entirely
convinced until he manages to reproduce their results by entering Melosh鈥檚
equation of state into his own simulation. 鈥淚鈥檓 not saying it鈥檚 wrong,鈥� he says,
鈥渂ut as far as I鈥檓 concerned, it鈥檚 an unfinished problem.鈥�

Canup agrees more work is needed, but she鈥檚 pretty sure the overall picture
is right. In particular, she reckons that her simulations solve another lunar
puzzle that has bothered planetary scientists. Most other planets in our Solar
System have several small moons rather than one big one, and they tend to be
further from their planet. Our own Moon鈥檚 large size and proximity are thought
to be crucial for the evolution of life, as they have helped prevent Earth
suffering the sort of chaotic changes in tilt that have dogged Mars. Also, the
strong tidal forces from the Moon have helped to create many large, shallow
coastal regions鈥攋ust the sort of environment where life is thought to have
originated. Planetary scientists have long wondered why we were uniquely blessed
with a single, stabilising satellite.

Asphaug and Canup鈥檚 simulation could provide the answer. It indicates that
much of the impact debris was unavailable for forming satellites. Instead, the
debris is left orbiting the Earth around a radius known as the Roche limit. This
close in, tidal forces from the planet are so strong that they tear apart any
satellites as fast as they can form. But there is just enough material outside
this limit to form a single, large Moon. 鈥淚f two smaller moons formed, they
would be so close they would be pulled together,鈥� says Canup. But once formed,
the Moon is so large relative to the Earth that it pulls material within the
Roche limit into unstable orbits that decay back to Earth.

Knowing how our Moon formed could help us understand other celestial bodies.
Alan Stern, also at the Southwest Research Institute in Boulder, is one of a
number of scientists who pointed out in 1999 that Pluto鈥檚 moon Charon has many
of the same features as our own Moon, and could have formed from an impact in
the same way. Stern thinks there are probably other examples even further out in
the Solar System.

And if giant collisions really are a normal part of a planet鈥檚 adolescence,
why stop at the boundaries of our own system? Stern calculates that any
Earth-sized planet that recently underwent a giant collision would shine as
brightly as a white dwarf star for thousands of years. If so, it would appear
bright enough in the sky to be spotted by a large telescope here on Earth. As
most Earth-like planets are swamped by the light from their host star, looking
for mighty, Moon-making collisions in space might prove to be the best way for
us to see a planet like ours in the making.

To confirm whether the Moon really did form from gas, researchers need to get
their hands on more Moon rocks. The samples brought back to Earth by the Apollo
missions were collected from just a few sites on the Moon鈥檚 surface. And while
they provide some evidence consistent with the impact theory of the Moon鈥檚
formation, they only offer a limited view of lunar geology.

Many Apollo samples contained high levels of titanium, which turns out to be
just what you鈥檇 expect if much of the early Moon was once molten. But that
doesn鈥檛 tell you whether the whole Moon was molten, or just a thin layer near
the surface. Last year, Clive Neal, a planetary scientist at the University of
Notre Dame in Indiana, found mixtures of trace elements in the Apollo samples
that are typical of a high-pressure mineral called garnet, which could only be
stable below about 400 kilometres into the Moon鈥檚 interior. This implies that
some of the rocks now on the surface originated deep inside the Moon. Levels of
other elements in the same rocks show they could not have been molten, meaning
the interior of the Moon probably never even melted, Neal believes. 鈥淭he
interior core is unprocessed debris.鈥� Drilling cores deep into the crust to see
if the signatures of garnet and primitive material get stronger could show
whether he is right and pin down the composition of the primitive Moon more
precisely.

Robin Canup at the Southwest Research Institute in Boulder, Colorado, says
new samples could also confirm one important prediction of her simulation of an
Earth-planet collision: about 80 per cent of Moon rock should come from the
colliding planet, and about 20 per cent from the Earth鈥檚 crust. 鈥淲ith current
samples, it鈥檚 not yet clear,鈥� she says.