PLANETS are tiny, insignificant things compared with their parent
stars鈥攍ittle more than moths fluttering around the stellar camp fire. You
might think that they could have no discernible effect on their stars and, until
recently, astronomers would have agreed with you.
But it is becoming clear that when old stars grow huge enough to swallow
their planets, they can become quite ill. 鈥淚t seems that planets can have a
profound effect on stellar evolution,鈥� says Mario Livio of the Space Telescope
Science Institute in Baltimore, Maryland. Planet gobbling could explain the
properties of many giant stars and the beautiful shapes of glowing clouds known
as planetary nebulae.
Any star that becomes a giant can swallow nearby planets, and almost all
stars eventually evolve into giants. When a star is in its so-called main
sequence phase鈥攖he phase our Sun is currently going through鈥攊t
鈥渂urns鈥� hydrogen nuclei in its core, fusing them and releasing energy. The
energy that pours out balances the great weight of the star and prevents it from
collapsing. However, when all the hydrogen fuel in the core is exhausted and
there is no longer any heat being generated, the core is crushed by the star鈥檚
weight, heats up tremendously and ignites unburnt hydrogen in a shell around it.
The surge of new heat inflates the outer parts of the star, swelling it into a
monstrous red giant often more than a hundred times the diameter of the Sun. The
red giant phase lasts about a tenth as long as the main sequence.
Advertisement
No one dreamt that planets could influence this process, because the only
planets likely to be swallowed are close-orbiting ones, and in our Solar System
the close planets are tiny. Mercury, Venus and the Earth would be little more
than a light snack for our Sun.
But in the past few years, astronomers have discovered about 20 planets
orbiting other stars. Most of these planets are giants at least as massive as
Jupiter. 鈥淚t now appears that at least 5 per cent of nearby stars have planetary
companions,鈥� says Livio. The big surprise was that in several of these systems,
the giant planets orbit searingly close to their stars鈥攎ost notably in the
case of 51 Pegasi B, which is a hundred times closer to its star than Jupiter is
to the Sun. 鈥淭his was a totally unexpected discovery,鈥� says Livio.
And it seems that such planets, if swallowed, could drastically change their
stars. In the June issue of Monthly Notices of the Royal Astronomical
Society, Livio and his colleague Lionel Siess reported computer simulations
of this catastrophe. Surprisingly, they found that a large planet continues to
orbit inside the star for thousands of years, only slowly being vaporised by the
heat. 鈥淵ou have to remember that these stars are super-tenuous gas balls with
their matter smeared over an absolutely huge volume,鈥� says Livio. 鈥淭heir outer
regions are as rarefied as what we would consider a good vacuum on Earth.鈥�
Despite the ethereal nature of the stellar envelope, it still applies a
frictional drag that slows anything passing through it, so the swallowed planet
gradually spirals down towards the core. How close it gets depends on its mass.
According to the simulations, a planet with the mass of Jupiter would evaporate
long before it reached anywhere near the core. But objects more massive than 20
Jupiters would survive almost all the way to the just outside the core, where
the temperature is about 2 million degrees. There, they would be dissipated by
the combined effects of the heat and the tidal forces exerted by the core, which
should stretch a planet and rip it apart.
Stardust
But the star doesn鈥檛 get off scot-free. Livio and Siess found that
gravitational energy dumped by the sinking planet should heat up the star,
causing it to puff off its cool outer layers as expanding shells of warm gas and
dust glowing with infrared light. And the rapidly orbiting planet should raise
the spin rate of the star鈥攔ed giants normally rotate only
sluggishly鈥攁nd contaminate the star with its heavy elements.
Clearly, these effects depend on the mass of the swallowed planet, with more
massive planets having a greater effect. And planets are not the only stellar
companions that can be swallowed by a star when it becomes a giant. In the past
three years, astronomers using the infrared sky surveys DENIS and 2MASS have
discovered a few dozen solitary brown dwarfs鈥攕mall failed stars which can
be between 10 and 80 times the mass of Jupiter. As these are bigger than
planets, they can have a more profound effect on a star. They may even merge
with the stellar core.
Whether they merge or not depends on the stage in the star鈥檚 life when they
are swallowed. Livio and Siess鈥檚 simulations show that if the star鈥檚 envelope is
thin, containing less than a hundredth of a solar mass of gas, the gravitational
energy unleashed by a sinking brown dwarf may be enough to eject the envelope,
leaving the dwarf in orbit around the stellar core. If the envelope is thicker,
the brown dwarf will either be dissipated or collide and merge with the
core.
Are there any signs of these processes in real stars? Fortunately, some of
the effects of swallowing a planet or brown dwarf, such as a high spin rate and
lithium abundance, should persist for hundreds of thousands of years. Livio and
Siess say that many red giants have all of these tell-tale traits鈥攖hey are
spinning very rapidly and emitting unexpectedly large amounts of infrared
radiation. And their light shows the spectral fingerprint of lithium, an element
that does not normally survive for long in a star, which suggests that it was
added by a planet fairly recently.
The two astronomers estimate that between 4 and 8 per cent of red giants show
evidence of planet swallowing. 鈥淭his agrees well with more direct estimates of
how common planets are,鈥� says Livio. Caty Pilachowski of the National Optical
Astronomy Observatory at Kitt Peak in Arizona agrees. 鈥淭he planet-swallowing
hypothesis is the best explanation I鈥檝e seen for the origin of these
lithium-rich giants.鈥�
Glowing nebula
Planet swallowing may also explain another puzzle. By the end of its giant
phase, a star鈥檚 radiation has completely blown away its outer layers, exposing
the core鈥攚hose eventual fate is to end up as a super-dense white dwarf.
Intense ultraviolet radiation from the core ionises the surrounding cloud of
ejected gas, causing it to fluoresce and glow. This is a planetary nebula,
so-called because of the glowing cloud鈥檚 resemblance to a planetary disc. There
seems no reason for the gas to be ejected more forcefully in one direction than
another, so you might think these nebulae would be spherical. But many are
bipolar, with material ejected along a preferred axis.
Twenty years ago, Livio suggested that this bipolar shape may arise when the
nebula has another star as a close companion. He proposed that the companion
becomes enveloped in the main star鈥檚 outer layers and spirals inwards, dumping
enough energy to eject these layers, albeit at a leisurely speed of about 20
kilometres per second. This material stays in the plane of the companion鈥檚 orbit
to form a doughnut-shaped cloud around the main star. Later, when the star鈥檚
blistering core is exposed, its radiation drives a fierce stellar wind of matter
away from the star at about 1000 kilometres a second. The wind blows in all
directions, but because it is impeded by the slow-moving gas in the doughnut, it
emerges perpendicularly as a bipolar outflow. 鈥淭he doughnut acts like a corset,
and the wind blows two bubbles perpendicular to the corset,鈥� says Livio.
The problem with Livio鈥檚 idea was that only a few planetary nebulae appeared
to have both a white dwarf and a spiralled-in companion in their hearts. But,
working with his student Noam Soker, who is now at the University of Haifa in
Israel, Livio showed that it鈥檚 no problem if the companion was a massive planet
or brown dwarf. 鈥淚t could have completely dissipated or merged with the white
dwarf,鈥� he says.
Young and hungry
All this applies to stars near the end of their lives. But it seems that
planets might be gobbled up by young stars too. There are high quantities of
heavy metals in the atmospheres of several stars with massive planets鈥攁
sign that they once ate one or more planets
(鈥淒eath stars鈥�, New 杏吧原创, 23 October, p 10).
These unfortunate bodies may have been
dragged inwards by tides in the disc of gas surrounding each young star.
Despite all the successes of the planet-gobbling idea, however, there are
still a few puzzles left to solve. Pilachowski points out that the story became
more complex with the recent observation of a lithium-rich giant in the globular
star cluster M3. A team led by Robert Kraft of the University of California at
Santa Cruz reported finding the star earlier this year. The astronomers argue
that it contains far more lithium than a planet could have carried鈥攊n fact
almost ten times as much lithium as any planet could have delivered to it. 鈥淭hey
argue that the lithium must have actually been made in the star by nuclear
reactions,鈥� says Pilachowski. 鈥淚 think the jury is still out. As usual, we need
more data.鈥�
But Pilachowski says this apparent discrepancy takes nothing away from Livio
and Siess鈥檚 work. 鈥淭heir work highlights an important weakness in our theory of
stellar evolution,鈥� she says. 鈥淣amely that planets can have a profound, and
until now unappreciated, effect on the evolution of stars.鈥�
As to whether the Earth will one day be consumed by the Sun, there is still
some doubt (see 鈥淓arth to ashes鈥�). 鈥淏ut whatever happens,鈥� says Livio, 鈥渋t won鈥檛
be for another 5 billion years. So we can all sleep safely in our beds
迟辞苍颈驳丑迟.鈥�
WHAT will become of the Earth in 5 billion years鈥� time, when the Sun鈥檚 core
runs out of hydrogen fuel and it swells to become a red giant? Many astronomy
books say that the Earth will be swallowed by the Sun, which, in its red giant
phase, could balloon out almost to the orbit of Mars. But in 1993 a team led by
Juliana Sackmann of Caltech in Pasadena pointed out that, although the Sun will
certainly expand past our current orbit, the Earth might no longer be there.
It鈥檚 all down to the stellar wind. Red giants lose material at a terrific
rate. The outer regions are only weakly bound by gravity because they are far
from the centre of the star, and so the pressure of the star鈥檚 light can drive
off gas at the surface. As the Sun loses mass, its gravity will weaken, and the
Earth will slip farther away. By this reasoning, when the Sun鈥檚 outer regions
reach Earth鈥檚 current orbit, it will have only 60 per cent of its present mass,
so the Earth will be 70 per cent farther away, and will probably escape.
Mario Livio of the Space Telescope Science Institute in Baltimore, Maryland,
and his colleagues have verified Sackmann鈥檚 calculations. But they point out a
competing effect, which throws our eventual fate into doubt. 鈥淭he Earth raises a
tidal bulge in the Sun, which it will try to drag around with it as it orbits,鈥�
says Livio. The upshot is that the Sun鈥檚 envelope will gradually spin faster,
while Earth will slow and move inward. 鈥淭he rate at which the Earth is sapped of
orbital energy depends crucially on the viscosity of the envelope, which is
uncertain,鈥� says Livio. 鈥淪o at the moment it is not possible to tell which of
the two effects will win and whether the Earth will be gobbled or not鈥� (The
Astrophysical Journal, vol 470, p 1187).
If the Earth is gobbled up, it will be vaporised before it gets very far
under the Sun鈥檚 skin. If it escapes it will continue to orbit, but from the
surface of the planet鈥攐r the blackened lump of slag it will have
become鈥攖he Sun will be a monstrous red glob filling almost all of the
sky.