Stars, Planets and Life
   Joerg Rhiemeier

And pray that there's intelligent life somewhere up in space,
because there's bugger all down here on Earth.
-- Monty Python, Galaxy Song (from The Meaning of Life)

Stars

Big stars, small stars, red stars, blue stars

Stars come in different colours and sizes. There are bright stars and faint
stars, there are blue, white, yellow and red stars. While brightness is a
measurable quantity, colour is a subjective impression. However, the
astronomers have found a way to make the fuzzy term `colour' into a measurable
quantity. This is the spectral type. They classify stars by characteristics of
their visual spectrum, and as colour is the visible outcome of the spectrum,
each spectral type corresponds to a colour. The spectral types are O (blue), B
(blue-white), A (white), F (yellow-white), G (yellow), K (orange) and M (red).
(Note that these apparent colours are merely a delusion of the eye. Against the
blue straylight at night, the stars appear much redder than they are. Actually,
the `reddest' stars are no redder than a light bulb!) Each of these types is
subdivided into 10 subtypes denoted by digits 0 to 9. However, not all
colour/brightness combinations are equally likely. If you draw a diagram
showing the relation of colour and luminosity (the Hertzsprung Russell Diagram,
HRD -- fig.1), you will see that most stars are aligned on a narrow band
stretching from bright blue to faint red stars. This is called the main
sequence; among these, the faint red stars are the most common. (An typical
area of 100x100x100 light-years contains a few hundred G-type stars, but
several thousand M-type stars.) Stars above the main sequence (brighter than
main sequence stars of their colour) are called giants. Most areas below the
main sequence are virtually empty; most stars below the main sequence fall into
the class of the very faint white dwarfs, but there is another group of stars
below the main sequence, the subdwarfs. Despite their name, these are not
fainter than white dwarfs, but only half an order of magnitude fainter than
main sequence stars of the same colour. Subdwarfs are very rare (except in
galactic centre, globular clusters and elliptical galaxies); they only occur in
the spectral types F, G, K and M (the O, B and A subdwarfs all have died long
ago -- see section Luminosity classes and populations).

The colour of a star depends on its surface temperature: blue stars are the
hottest, red ones the coolest. The hotter a star is, the more light it
irradiates per surface area. This means that of two equally-sized stars, the
more blueish one will be the brighter, of two different-coloured,
equally-bright stars, the more reddish one will be larger. This means that
bright red supergiants are HUGE, with diameters of up to several thousand
million kilometers, while white dwarfs are only planet-sized. Along the main
sequence, the differences in diameter are within one order of magnitude or two
(the blue stars are bigger than the red ones).

However, the stellar masses differ much less than diameter and luminosity. The
biggest blue stars have about 50 solar masses, the smallest red ones about one
tenth of the mass of our good old sun. Giants usually are less massive than
equally-bright main sequence stars; white dwarfs, though planet-sized, have
masses comparable to that of our sun. For sun-like main sequence stars, the
luminosity approximately grows proportionally to the fourth power of the star's
mass.

This means that stellar densities span a wide range. Of the main sequence
stars, the cool red ones are the densest. These reach a density of about 5
grams per cubic centimetre, while blue main sequence stars are about 0.05
g/cm^3. However, red giant stars are very tenuous; a cubic metre of a red giant
contains only a few milligrams. This is much less than Earth's atmosphere!
Astrophysicists often speak of a `red vacuum'. On the other hand, white dwarfs
are incredibly dense: a cubic centimetre taken out of a white dwarf weighs
about a ton. The mass of an entire car compressed into a thimble!

Luminosity classes and populations

Besides spectral types, two further classifications are common. The first one
are the luminosity classes. These go as follows:

   *  Ia bright supergiants (>10000 solar luminosities)
   *  Ib supergiants (5000-10000 s.l.)
   *  II bright giants (200-5000 s.l.)
   *  III normal giants (50-200 s.l.)
   *  IV subgiants (5-50 s.l.)
   *  V main sequence stars
   *  VI subdwarfs
   *  VII white dwarfs

These luminosity classes are appended to the spectral type, e.g. M0Ib, K2III,
F5IV, G2V, K3VI. VII is rarely used; it's much more common to denote white
dwarfs by a prefix D (e.g. DA5).

The stellar populations are defined by the age of the stars; they differ mainly
in chemical composition. Population I stars contain much more heavy elements
than Population II stars. The reason is that Population II stars are
first-generation stars; they are as old as the galaxy (about 15000 million
years) and were formed of pre-stellar matter, while Population I stars are
younger and contain reasonable amounts of the ashes of stars that died before.
All main sequence stars belong to Population I, as well as supergiants. The
Population II stars are subdwarfs or some giants and subgiants. While
Population I stars are concentrated in the galactic disk, Population II stars
are found in the central region of the galaxy and in the galactic halo.
Globular clusters consist solely of Population II stars, as well as elliptical
galaxies. The difference in composition strongly affects the probability of
planets and life: Population II stars have at best hydrogen giant planets
without mineral cores. These are gloomy, pale and colourless globes because
there is nothing that could form coloured clouds, just hydrogen and a small
proportion of helium (primarily in their core) -- and no life!

Stellar evolution

Stars form when interstellar gas clouds collapse until the core becomes dense
and hot enough to ignite nuclear fusion. The young star is now a T Tauri or
protostar -- an unsteadily flickering youngster struggling for a balanced
state. At this time, planets begin to take shape. In the HRD, these young stars
are to find some way above the main sequence. This T Tauri stage lasts a few
dozen million years until the star reaches the main sequence.

The main sequence stage is the glory days of a star's life. How long this
lasts, depends on the mass of the star: our sun has been on the main sequence
for 4500 million years now and is expected to do so for further 5000 million
years. Small red stars are likely to stay up to 50000 million years on the main
sequence, while bright blue stars have a main-sequence lifetime of only a few
million years. This means that only medium and small stars stay long enough on
the main sequence to give life on their planets enough time to evolve.

But finally, the end is near. When the star has used up all the hydrogen in its
core, nuclear fusion comes to a halt. The core cools down and collapses, and,
by that, heats itself up again until the centre is hot and dense enough to
allow fusion of helium into carbon. The core temperature now exceeds 100
million kelvins (main sequence stars have core temperatures between 5 and 30
million kelvins). Hydrogen fusion ignites in a shell around the core, and the
immense heat inflates the star. It grows up to a few hundred times in size. Now
the star is a giant star, one or two orders of magnitudes brighter, but also
much redder than before: being far away from the core, the surface is now
cooler than it was during main sequence times. The star inflates until it
reaches a new stable state. But it will never be as stable as the main sequence
stage. Many giants pulsate (e.g. Delta Cephei and Mira stars), and last only
some million, if not a few thousand years. Again, the smaller the star, the
longer it remains stable. However, the giant stage is far too short and far too
unstable to let life evolve on outer planets which now have comfortable
temperature (while the inner planets are calcinated or even evaporated).

What happens next, depends on the star's mass. When all the helium is used up,
the core collapses again. A final upheaval of nuclear fusion blows off the
star's outer hull, creating a planetary nebula. If the star has less than 1.5
solar masses, its core shrinks and becomes a white dwarf. A white dwarf is a
dead star without nuclear fusion going on in the inside, eradiating only its
remaining heat until it finally glows out and becomes a dark, compact object, a
black dwarf. At a mass of more than 1.5 solar masses, the white dwarf undergoes
a further collapse. This ignites further nuclear reactions. Carbon turns to
oxygen, neon and up the periodic table until silicon. These reactions run
faster and faster until, in a final flash of glory, silicon gets fused to iron.
This releases so much energy that the star blows up: a supernova. The remaining
core is pressed furtherly together until the electrons crash into the nuclei
and the nuclei merge together. This is a neutron star, effectively a giant
atomic nucleus of stellar mass and 20 kilometers diameter! However, the most
massive neutron stars collapse furthermore until the escape velocity at the
surface exceeds the speed of light. This means that nothing can escape from
this kind of object (except, perhaps, an FTL starship, though this is very
likely to be destroyed by tidal stretch long before it reaches the `event
horizon'). The star is now a black hole.

(However, a black hole might irradiate energy and lose mass, though. Stephen
Hawking has explained how: in the vicinity of the black hole, the high energy
density leads to the frequent formation of virtual particle/antiparticle pairs.
Now, one of the two particles might cross the event horizon, while the other
evades this trap. Now we have a particle with mass and kinetic energy coming
out of nowhere. But this energy must have come from somewhere, of course!
According to Hawking, it is drawn from the black hole, which loses the
corresponding amount of mass.)

Planets and moons

Which stars have planets at all?

Not all stars have planets. There are three classes of stars which are unlikely
to have planets:

   *  Members of close binary or multiple systems. In this case, the stars
     would disturb each other's planetary orbits. As a rule of thumb, no orbits
     with radii of more than 1/3 of the minimum distance between the two stars
     are stable. For planets bearing life, the orbit radius must be even
     smaller compared to the minimum star distance to allow stable climatic
     condotions. Alpha Centauri is probably the closest where something like
     Earth could exist. However, even if the stars are far enough apart to
     allow for stable orbits of planets, their tidal forces might preclude the
     formation of planets, just like the tidal forces of Jupiter might be the
     reason why the asteroids between Mars and Jupiter did not combine into a
     compact planet.
   *  Fast-rotating stars. These have an angular momentum similar to that of
     the entire solar system. The most probable reason why they have retained
     such a high angular momentum is simply that they have no planets to share
     their angular momentum with, except perhaps some small far-out
     terrestrials. Most O and B stars and many A stars fall into this category.
     These are the big ones who are too short-lived to let higher life-forms on
     their planets evolve anyway. The smaller F, G, K and M main-sequence stars
     rotate slowly, as our sun does, and probably have planets.
   *  Population II stars. When they were formed, elements heavier than helium
     were very scarce. Thus, they don't have planets, except perhaps
     hydrogen-helium giants which would be, because of the absence of heavier
     elements, colourless and sterile.

There is a possible explanation why large stars seem to be planetless. Within
each solar system, there is an inner limit within which planets cannot form
because it is simply too hot. Above about 1000 or 2000 degrees C, not even
metals and silicates condense any more. In our solar system, this limit is
quite close to the sun, far within Mercury's orbit. But a huge O star is so hot
that even the outer fringes of its protoplanetary disk won't condense.

This leaves us with the medium to small population I stars. Most of these stars
are still on the main sequence (the sun has a main-sequence lifetime of 10000
million years, thus a red giant of solar mass is at least 10000 million years
old, and that was the time when the last population II stars were born).

Planet types

In our solar system, there are two basic types of major planets:

   *  Terrestrial planets. These are composed of rock, with a metallic core.
     Volatile substances, if any, only exist in comparatively small amounts at
     the surface. Terrestrial planets of our sun are Mercury, Venus, Earth and
     Mars. Of the larger moons, the Earth's moon, Io and Europa (which has only
     a comparatively thin ice crust above an ocean heated by tidal friction)
     fall into this class.
   *  Jovian planets, a.k.a. gas giants. Besides having a small (perhaps
     Earth-sized) rock/metal core, these are composed mainly of hydrogen and
     other volatiles. A jovian planet has no solid surface, but rather a deep
     atmosphere which becomes denser and denser towards the planet's core until
     it turns into a supercritical, liquid-like state. Between this massive
     layer of liquid hydrogen and the rocky core lies a layer composed of water
     and other hydrogen compounds. The jovian planets of our solar system can
     be divided into two distinct types:
        o  Type I: Jupiter, Saturn. The pressure within these is so high that
          the inner part of their hydrogen bodies has turned into a metallic
          state.
        o  Type II: Uranus, Neptune. These are smaller and contain a much
          higher fraction of water and other hydrogen compounds. The hydrogen
          layer is much thinner; the pressure does not suffice to create
          metallic hydrogen.

Pluto falls in neither of these categories. Recent observations seem to mandate
the conclusion that Pluto is just a very large comet rather than a planet. (The
assumption that Pluto is a former satellite of Neptune has been dropped.)

It is clearly noticeable that, while terrestrial planets are confined to the
inner part of the solar system, the jovian planets are to be found in the outer
parts. But is this necessarily so, or is this merely coincidental? The truth
probably lies somewhere in between -- it might be possible that jovian planets
form in the inner area of a planetary system, but it is less likely. The `seed'
of a jovian planet is always a large solid body, massive enough to grab large
amounts of hydrogen and other gases from the solar nebula.

Outside the 170-kelvin (the evaporation point of water ice in vacuum) limit (in
our solar system, this is at about 3 A.U. from the sun) there is just much more
material available to build such an object (there is much more water vapour in
the solar nebula than silicates or metals!), thus, out there it is more likely
that a juvenile planet gains enough mass to grow into a gas giant.

In the inner solar system, the initial solid body must be entirely built from
rocks and metals. Then it can start attracting water vapor, carbon dioxide,
nitrogen and other heavier volatiles first, and then hydrogen and helium.
Besides, higher temperatures and solar wind constitute further difficulties.
The gases are just harder to hold. This makes the formation of a jovian planet
on a close orbit less likely. But if there are such objects in other solar
systems, they probably contain a higher proportion of metals, silicates and
heavy volatiles (water, nitrogen, etc.) than Jupiter. They will probably differ
enough from Jupiter and Saturn to justify being classified as a third type
besides the Jupiter/Saturn and Uranus/Neptune types. (The reason why, in our
solar system, the smaller type II giants occur farther outward than the big
type I planets is probably because there was less material available out there
-- the solar nebula doubtlessly was thinner at its edges. The temperature
differences might also have played a role.)

The next question is: how big can a jovian planet be, before igniting itself
and becoming a star? It is assumed that the smallest stars have about 8% of the
Sun's mass. Smaller objects don't reach the core temperature necessary to
ignite nuclear fusion. They might gain enough heat from contraction to emit a
faint reddish glow, but such a brown dwarf is not really a star. On a related
issue, there are probably no planets which exceed Jupiter's radius by very
much. Huge gas giants are simply denser, but not that much thicker. Jupiter
isn't that much bigger than Saturn (142000 km as compared to 120000 km), though
it has more than three times its mass, but it is denser. A planet three times
as massive as Jupiter might have a diameter of 150000 or perhaps 160000 km, but
probably not more. The smallest red dwarf stars aren't much larger, either.

The asteroid belt is considered to consist of left-overs from the formation of
the solar system; it is assumed that Jupiter precluded the formation of a
planet here. (The old, popular theory which states that the asteroids are the
remnants of a destroyed planet, is now dead meat.) An asteroid belt between the
terrestrial planets in the inner and the giants in the outer part of the system
might well be a typical case.

Thus we get the gross picture: terrestrial planets (and only occasionally a
jovian with terrestrial moons) in the inner solar system, and jovian planets
with icy satellites in the outer solar system, with an asteroid belt bounding
the two realms from each other, and a huge cloud of comets around.

The aforementioned reasons why jovian planets are less likely to form inside
the 170-kelvin limit might also be the reason why massive stars rotate faster.
Being brighter and hotter, they push their 170-kelvin limit towards the fringes
of their solar nebula (and heat up the inner areas so much that not even iron
and silicates condense), such that only a few smaller planets form (if any).
Without planets, the star retains its original rotation momentum, while smaller
stars share most of it with their planets and therefore rotate slowly.

Moons

Some planets have moons, others don't. It looks like that, in general, large
planets have more and larger moons.

It may be assumed that Jupiter, Saturn and Uranus are typical cases concerning
their moons. They all have satellite systems which look like miniature
planetary systems. And that is, basically, what they are. The moons formed in
their orbits around their planets just like the planets orbiting the sun. (Note
that their orbits, except for the tiny outer moonlets of Jupiter and Saturn's
Phoebe, which are most likely captured asteroids or cometary cores, lie within
the planet's equatorial plane, even in case of grossly-tilted Uranus.) Ring
systems complete with `shepherd' moonlets might also be standard case for gas
giants (all jovian planets in our solar system have them!).

Neptune is different. The system looks as if it has been through a great deal.
The ring system with its accompanying moonlets is intact, but the rest looks as
if something had wreaked havoc upon it some time ago. One moon, Triton, orbits
on a retrograde orbit, the other, Nereid, on a highly eccentric one. This
system seems to have been disturbed by something. The most popular theory is
that something bad happened which not only messed up Triton's and Nereid's
orbits, but also threw out another large moon entirely -- the one we now know
as Pluto. However, this assumption is now considered obsolete. The
state-of-the-art theory, however, claims that Triton originally was an
independent body, a giant comet just like Pluto, which was captured by Neptune
-- and thereby messed up Nereid's orbit. If this was true, it would mean that
there might be quite many more Pluto-sized objects around at the outer fringe
of the solar system.

Now consider the terrestrial planets. Mercury and Venus have no moons (it is
now pretty certain that Neith, that `ghost moon' of Venus which used to fool
astronomers for over 200 years, is not really there), Earth has one, Mars has
two. Mars's two moons are tiny, while Earth's is a really large one -- it would
make a dignous companion of a large jovian planet, or a small but respectable
planet (definitely much more than an asteroid!) in itself. It is therefore
sometimes said that Earth and Moon are a binary planet.

It is almost certain that Earth's moon is a rare, exceptional case. Terrestrial
planets normally just don't have moons that big. The currently best accepted
theory is that Earth once collided with another big terrestrial body. Much
material was thrown out of Earth's body during that event, and from this formed
the Moon. Note that such a collision is a fairly unlikely event, hence
Earth/Moon might be a real freak of nature. The normal case seems to be that a
terrestrial planet has a few tiny moons, or none at all.

Regarding Mars's moons, it is not certain where they have come from. They might
have formed from debris in orbit around Mars which was left over from the
formation of the planet, or (considered more likely) they might be captured
asteroids.

Finally, let's have a look at the Pluto/Charon system. Pluto is accompanied by
a moon which has half of Pluto's diameter and perhaps 1/10 or 1/8 of its mass.
This means that Pluto and Charon are much more clearly a binary planet (or, for
that matter, a huge binary comet) than Earth and Moon (imagine Earth being
orbited by Mars instead of the Moon!), especially if one considers two
properties of this system that Earth and Moon do not share:

   *  Pluto and Charon are tidally locked to each other. Most moons always turn
     the same side towards their planet. Our moon does, for example, does this.
     This is also the case with Charon, but also with Pluto: its rotation
     period is equal to the revolution period of Charon. Thus, on Pluto the
     moon never rises, nor sets. It just stays put in the sky as seen from one
     hemisphere of Pluto; from the other, it can never be seen.
   *  The mass centre lies in open space between the two bodies, while in the
     case of Earth and Moon it lies, though noticeably displaced from Earth's
     centre, well within Earth.

So the rule seems to be that terrestrial planets usually have few tiny moonlets
or no moons at all and jovian planets have miniature planetary systems and
rings. But there are always ones which break the rule.

Prospects for life on planetary surfaces

The place where one might look for life first are terrestrial planets. But, as
we see in our solar system, not every terrestrial world offers good prospects
for life. We have five in our solar system (if we count the Moon as well), and
only one of them -- Earth -- abounds with life in manyfold forms, including
sentient beings. There might be primitive microscopic life on Mars, but it is
as good as certain that there is no life at all on Mercury, Venus and the Moon.

These worlds have all evolved differently. The evolutionary path a planet takes
depends on two factors: surface temperature and surface gravity. Earth has
exactly the right temperature, and it has enough gravity to hold an atmosphere
dense enough to allow for liquid water. The Moon lies in the same optimal
temperature zone -- but it is too small. With one sixth of Earth's gravity, it
could not hold any significant atmosphere.

Mars is farther away from the sun than Earth, and it is colder. But the
question remains if its problem really is its distance from the sun. It rather
seems to be its size, similar to the case of the Moon. Regarding size, Mars
lies in the middle between Earth and Moon. And this impression of Mars standing
halfways between Earth and Moon is backed up by the comparison of other
properties of the three bodies as well. It has an atmosphere, but it is
tenuous: too thin to allow for significant amounts of liquid water. But it
seems to have had a denser atmosphere in prehistoric times. Old river beds have
been found on Mars, and this means that there must have been liquid water once.
But the planet was too small to retain the dense atmosphere needed for that for
long.

Now look at Venus. Venus is not too small -- it is only a bit smaller than
Earth, the difference shouldn't matter. Indeed, it has a very dense atmosphere.
The problem with Venus is that it is too close to the sun, and therefore too
hot. It might have had liquid water once, but it was very warm nevertheless,
and a run-away greenhouse effect set in, spoiling the thing.

The Carbonate Circuit Process

The most prominent volatile substance on terrestrial worlds is water, followed
by carbon dioxide and then nitrogen. Water becomes liquid if the planet is cool
enough to allow for this; carbon dioxide reacts with basic minerals (metal
oxides and silicates) forming carbonates, if the planet is not too hot.
Therefore, Earth's atmosphere is mainly composed of the remaining gas:
nitrogen. Water is mostly in the oceans, carbon dioxide bound as carbonates.
Now, both liquid water and carbonates coexist with water vapour resp. gaseous
carbon dioxide in a dynamic balance, a circuit process. The water circuit is
simple: water evaporates from oceans, lakes and rivers, and rains down again.
But there is also a carbonate circuit. If carbonates are carried into deeper
layers of the Earth's crust, they are broken up, and carbon dioxide is released
into the atmosphere through volcanoes. But it does not accumulate there
indefinitely. It gets washed out by rain, and forms carbonates again when the
dissolved carbon dioxide gets in contact with basic minerals.

Now, carbon dioxide is the most important variable greenhouse gas in
terrestrial atmospheres. This means that the more carbon dioxide there is in
the planet's atmosphere, the warmer the planet is. The trick is that the warmer
the planet is, the more rapidly the water circuit is turning, and the more
carbon dioxide is washed out of the atmopshere, cooling the planet. (One can
safely assume that geological processes are not significantly influenced by
changes of the planet's temperature by a few degrees, which means that the rate
at which the washed-out carbon dioxide is replenished through volcanoes can be
assumed to be constant.) On the other hand, if the planet cools down for some
reason, the carbon dioxide concentration in the atmosphere rises, warming the
planet. This keeps temperatures on Earth (or a similar planet) stable in the
long run.

However, this has its limits. The upper temperature limit is marked by the
point where no more carbon dioxide is left to be washed out. At that point, a
further increase of temperature can no longer be compensated. With the carbon
dioxide greenhouse effect being reduced to zero, the planet's temperature is
now governed by another greenhouse effect which is normally overshadowed by the
carbon dioxide greenhouse effect: the one of water vapour. Unfortunately, this
one does not stabilize the planet's temperature: the warmer the planet is, the
more water evaporates, enhancing greenhouse effect and warming the planet even
more, until the oceans start boiling away.

This cataclysm pushes the water vapour greenhouse forward, heating the planet
up to perhaps 200 or 250 degrees C. At those temperatures, there is of course
no longer a water circuit. This means that carbon dioxide accumulates again,
adding to the greenhouse effect. Now the planet is so hot that the carbonates
on the surface are slowly broken up, releasing more carbon dioxide and pushing
up the temperature, until there are no carbonates left. The result is a
superdense carbon dioxide atmosphere and temperatures that would melt lead.
Venus is an example for this.

The other extreme is that the planet becomes so cold that the water freezes.
This, however, brings the water circuit to a halt, letting carbon dioxide
accumulate again, which might stop the planet from freezing over. However,
there is a limit to even that, set by a run-away albedo effect. When a planet
cools down, its polar ice caps increase in size. Ice is white, it reflects
light very well. This means that the planet receives less radiation from the
sun and becomes colder. More areas freeze over, more light is reflected, ...

The Life Zone

This marks the boundaries of the life zone, the area in which a terrestrial
planet might be a friendly place with liquid water. For an Earth-sized planet,
the inner limit is supposed to be at about 0.95 AU. This is quite close to
Earth's orbit, but that makes sense, considering that the Earth does not have
very much carbon dioxide in its atmosphere (just 0.035%, and that is not that
much). The outer limit is farther away from us, perhaps at 1.6 AU (a bit
outside Mars's orbit). If Mars had Earth's size, it would have as much liquid
water as Earth, and it might have a breathable atmosphere (if life had created
oxygen there; though it would have much more carbon dioxide, perhaps up to 1%
-- we would feel that Martian air was quite stale). (Note that the average
distance is important. It does not matter much (except that it would of course
affect seasonal temperatures) if a slightly eccentric orbit (like that of Mars)
grazes the limit of the life zone, as long as the planet orbits within it most
of the time.)

For different planet sizes, these limits might differ. The smaller the planet
is, the narrower the life zone becomes, because the outer limit moves inward.
(The planet, being smaller, has a thinner atmosphere and is therefore colder.)

With another sun, the limits are moved according to the star's luminosity. Now,
there is a problem with small stars (those with less than a few percent of
solar luminosity). In these cases, the life zone will be both narrow and close
to the star. If there is a planet close enough to be in the life zone at all,
it is likely to be tidally locked, i.e. it always turns the same side towards
its sun, having a seering hot and an icy cold hemisphere. It seems deliberately
unlikely that a stable atmosphere develops there.

So we can specify the kinds of places where we would want to look for life.
It's terrestrial planets (or moons) of sufficient size, orbiting within the
life zone of a sun-like star (i.e. a main sequence star of spectral type F5 or
`later', with a luminosity between 10% and 300% of that of our sun).

It is, as to now, entirely unknown how likely biological evolution on such a
planet is. The only example we have at hand is Earth, and one of the first
things they teach you in a statistics course is that you can't conclude
anything from a single example, other than that at least one case of the
observed properties exists.

It is also unknown how many special conditions must hold to allow biological
evolution. Our large moon doubtlessly improved the conditions for life on
Earth. Its strong tidal forces stirs up the Earth's interior, which results in
an augmentation of the Earth's magnetic field which serves as a shield against
hazardous radiation from outer space. The same forces also create the ocean
tides which, to some biologists, played a major role in the evolution of life.
But it is unknown whether these things were necessary or not.

Biological evolution

With a suitable planet, the stage is set for the drama of life. However, the
stage is all that is there initially -- a planet with oceans and an atmosphere
which consists primarily of nitrogen, with a few per cent of water vapor and
carbon dioxide, and small amounts of other gases such as methane, ammonia and
sulfur-hydrogen.

The primordial soup

This was the starting point of life on Earth (and probably of other planets as
well). At that time (more than 4000 million years ago), organic material was
scarce. Interstellar matter contains organic molecules such as hydrogen
cyanide, polyacetylene and formaldehyde; traces of these must also have been
present on Earth. The bulk of the organic material from which life formed
(which was, however, only a minute proportion of the organic matter that is
there now!) was created on Earth by abiotic processes.

How this happened, was demonstrated by the chemists Harold Urey and Stanley
Miller in the early 1950s. They irradiated a concoction of methane, ammonia and
water vapour with ultraviolet radiation and had electrical discharges (i.e.,
lightning) set off in it. After a few days, they found a brownish ooze in the
reaction vessel, the so-called tholin. This is a mixture of miscellaneous
organic compounds, among them several amino acids. What had happened was this:
the radiation and lightning bolts cracked up the gas molecules; the resulting
fragments combined into more complex molecules. On Earth 4000 million years
ago, the same thing happened in the atmosphere. Rainfalls washed the tholin
into the oceans; the resulting aquous solution is popularly called primordial
soup.

The first replicators

It is yet unknown how the first living organisms formed out of this `soup'. The
`soup' was a gargantuan natural laboratory where the widest diversity of
chemical reactions took place. At some time a molecule formed which catalyzed
the production of itself -- the first replicator. Chemists actually have
managed to build simple self-replicating molecules which are possibly similar
to those with which the evolution of life started.

Photosynthesis

The first primitive organisms lived in a plentiful environment. The Urey-Miller
processes produced lots of organic molecules which the primordial organisms
could thrive on. But this didn't last forever. On one hand, the organisms
became more and more numerous. On the other hand, ammonia amd other
water-soluble gases were washed out of the atmosphere by rainfalls, and the
remaining methane used up by the Urey-Miller processes.

This meant that life had to find other sources of nutrition to survive.
Doubtlessly, this was the time the first predators hit the scene, organisms
which lived off other living organisms, and the first scavengers, which fed of
those which failed, died for some reason.

However, this was little help against the ongoing crisis. If nothing had
happened, the young life would have used up all the nutrients and died out.
This might have happened on serveral planets. But on Earth, some organisms
began producing their nutrients on their own from simple, inorganic molecules.
This was possible through photosynthesis. These organisms gathered sunlight to
get hold of energy they needed to build carbohydrates from carbon dioxide and
hydrogen; however, as there was no molecular hydrogen available, this in turn
had to be gained from some kind of inorganic hydrogen compounds (which again
involved sunlight).

The first photosynthetic organisms probably used comparatively easy-to-break-up
molecules like ammonia and sulfur-hydrogen (some bacteria use the latter until
today), but then another line arrived which was capable of using water for this
purpose. Though water needs much more energy to break it up, these organisms
had a huge advantage because water was, of course, grossly abundant.

The role of oxygen

This version of photosynthesis, however, put out a waste product: oxygen. This
was a problem. Molecular oxygen is an aggressive gas which does not mix well
with organics. It cleared the atmosphere off what was left over from the
reducing gases with which the evolution of life started, and was deadly to most
of the organisms.

So life faced itself with a crisis again. There might be planets where the
evolution of life came to a grinding halt at that point. Earth was, of course,
none of them. Some organisms developed a protection mechanism against the
oxygen. The principle was to set aside a part of the carbonhydrate supply to
react with the oxygen. But this turned out to be more than just a protection
measure against a dangerous poison! This process provided the organisms with
much more energy than they could ever pull out of their food through
`conventional' means, and these aerobic organisms quickly took over.

Multicellar life

All these organisms were single cells of a primitive, bacteria-like design, so
called procaryonts. At some point, perhaps 1500 million years ago, a new, much
more complex type of cell, the eucaryont, evolved. This cell type, equipped
with a nucleus, mitochondria and other functional structures, became the
building block of the multicellar organisms.

The first multicellar life forms were little more than clusters of identical
cells, but step by step the functions of these cells differenciated. This way,
more and more complex (and larger) organisms evolved. The evolutionary lines of
plants -- mostly non-mobile, photosynthetic organisms -- and animals -- mostly
mobile organisms feeding off plants or other animals -- separated from each
other.

There is speculation if this is a necessary development happening on all
Earth-like planets with multicellar life, or if there could be mobile (or even
intelligent) photosynthetic organisms. The answer is probably that, if there
are self-moving organisms at all, they must be heterotrophic, i.e., live off
other organisms. This is because mobile organisms of course need more energy
than non-moving ones, and macroscopic animals just don't have enough body
surface to gather the energy they need from sunlight. (It is, however,
conceivable -- though probably not very likely -- that a planet is only
inhabited by plants, not by animals.)

The conquest of the dry land

Until perhaps 500 million years ago, the continents of our planet were merely
bare rock, while advanced multicellar life forms were already thriving in the
oceans. Then, the first land plants appeared, to be followed by the first land
animals.

The conquest of dry land was a major step in evolution which demanded special
adaptations. Water-living organisms need not care about loss of liquid. They
did not need a waterproof membrane, for example. Land-living organisms,
however, need such a membrane, and they need to compensate for water losses.
This is a difficult thing, and there might be planets where life, though as old
as Earth or even older, never conquered the continents.

Intelligent life and technology

A sessile plant does not need intelligence. To a self-moving animal, however, a
complex brain is very useful. An intelligent predator might fell prey that is
physically superior by playing tricks on it. On the other hand, an intelligent
animal can evade a physically superior, but dumber, predator.

Intelligence also allows for the usage of tools, which is the very first step
to technology. Real technology, however, probably prerequisites the
domestification of fire. This of course means that only land-living beings can
develop technology. There is no reason to assume that marine animals cannot
develop intelligence that matches, or even exceeds, human intelligence.
Actually, dolphins are at least as intelligent as apes; some scientists assume
that they are as intelligent as us humans. However, such marine intelligent
life forms have no access to technology.

The question how many planets actually spawn a technological civilization is
unresolved. From the point when the first land-living animals appeared, it
lasted more than 400 million years until the first autochthonic technical
civilization -- I am of course talking about us here -- appeared on Earth. On
some planets, this might never take place. And it is the question how long such
a civilization survives. We have the power to blow ourself out of existance, or
wreak so much havoc upon the planet that we (and most other species) cannot
survive any longer. It is possible that all civilizations annihilate themselves
this way after a few thousand years, but it might also be possible that a
civilization overcomes such problems and stays for millions of years.

What will the future bring?

There is no reason to assume that evolution on our planet has come to a halt.
Unless we wipe out ourselves, we will probably continue our development. No-one
can tell us whither the journey goes. We might evolve into a super-intelligent
species, or we might degenerate gradually. We might even evolve into something
we as for now can't imagine at all.

It might also be possible that this post-human evolution will put us on a path
which leads us away from planetary life. It need not even be biological
evolution. The path might lead from humans through cyborgs to intelligent,
self-replicating machinery living in outer space. It is therefore possible that
when searching for intelligent life in the vicinity of sun-like stars, we are
searching in the wrong place: they once lived on a planet of such a star, but
now don't need Earth-like planets any more and are now to find in the
vicinities of other kinds of objects which are more useful to them -- and where
we would never expect life, such as blue supergiants or pulsars.

A civilization that destroys itself might foul up its planet to such a degree
that no life survives. This is especially likely if it comes to a nuclear war.
Other extinct civilizations may leave behind a somehat changed and disturbed,
but still habitable planet (or many such planets, if they are an interstellar
civilization) on which an other sentient species might evolve later. Through
the eons, a planet might therefore spawn several different civilizations
subsequently.

Life in non-Earthlike environments?

So far we have considered life on planets similar to Earth: terrestrial planets
of sufficient size, orbiting within a zone of moderate temperatures, with
liquid water on their surface. But can there be life in other environments as
well? The Apollo astronauts had to go through quarantine measures upon their
return from the Moon because no-one was willing to take the risk of importing
something alien, infectious from there. (However, not a single trace of life
was found on the Moon -- not very surprising.) Mars can be considered a close
call towards an Earthlike world; it is considered possible that primitive life
exists there. The Viking probes couldn't find anything; but it is yet not
certain that there is no life on Mars.

Other places in our solar system have been considered to be possibly
life-bearing as well. No-one really expects life on Mercury or Venus (though
even this need not be entirely impossible!), or, on the other hand, on the
moons of Saturn (except Titan -- see below), or in the Uranus, Neptune or Pluto
systems. But it might be possible that there is life on Jupiter: while the
outer layers of its vast atmosphere are colder than -100 degrees C, its inner
parts are hot -- there must be an area of moderate temperatures in between.
Jupiter's atmosphere contains methane, ammonia, water, hydrogen sulfide and
other molecular compounds, and huge thunderstorms and lightnings have been
observed. Thus, Urey-Miller processes might be possible, and some scientists
actually assume that Jupiter's clouds are coloured by complex organic
compounds. This, however, is the first step towards biological evolution, and
perhaps, actually life exists on Jupiter, floating in its atmosphere. These
life forms, however, would be incredibly alien, different from anything ever to
be found on an Earth-like planet.

The next place sometimes mentioned as a possibly life-bearing world is the
second of Jupiter's Galilean moons: Europa. At first glance, Europa doesn't
seem very convincing -- it is entirely encrusted with ice and has no
atmosphere. However, its ice crust is extremely smooth -- just as if it had
been molten several million years ago. It is assumed that this is actually the
case: Europa, on its orbit between Io and Ganymede, suffers intense tidal
friction; this could have molten its crust and might still keep the lower layer
of it in liquid state, i.e. there might be an underground ocean beneath its
solid surface! However, if there is life there, it is life in utter darkness,
which rules out sunlight as energy source. It therefore would have to live off
something else, perhaps volcanic gases. It is unknown whether this is feasible,
which means that Europa might be just as dead as the other moons of Jupiter.
(Nobody expects life on Io with its sulfur volcanoes, especially in light of
the deadly radiation this moon is exposed to, or on the frozen worlds of
Ganymede and Callisto, which show no hints of an underground ocean.)

Another place often suspected to be capable of bearing life is Saturn's largest
moon, Titan. It has a dense, opaque atmosphere containing organic material.
However, one thing is certainly missing there: liquid water. Titan is simply
too cold to allow for this. It is, however, assumed that there is something
liquid on Titan -- a mixture of light hydrocarbons (like methane and ethane)
and several more complex substances. The possibility of life on Titan depends
on whether such a medium can play the role of water or not. This issue is
discussed below.

Alternatives to oxygen?

Is it possible that the kind of photosynthesis we know on Earth doesn't evolve,
but something else instead? There are many other molecules which could provide
for the hydrogen needed in photosynthesis, instead of water, and most of them
are easier to break up. These have doubtlessly played a role in early
bio-evolution on Earth, and at least one of them -- sulfur-hydrogen -- is used
by certain bacteria until today. But none of these molecules is in such a
plentiful supply as water. On Earth, those making use of water won over. On
some other planet, this variety of photosynthesis might never evolve. But on
those planets, life will get stuck on a primitive level -- anaerobic digestion
just does not put out enough energy to drive complex, mobile, intelligent
multicellars.

The only thing which has the chemical potential to rival oxygen is chlorine.
This could be a by-product of photosynthesis drawing hydrogen from hydrogen
chloride. It could also work with chloride salts; in this case, the hydrogen is
actually drawn from water (leaving behind hydroxide), but what is oxodized is
chlorine `Chlorine breathers', popular in bad SF, are chemically possible. But
are they likely? Chlorides are just as hard to oxidize as water, but they are
less abundant, giving advantage to those drawing their hydrogen from water.

Another possibility is that the photosynthesis process draws hydrogen from
water, but does not release the oxygen as O2. There are other things that could
come out of this. Water could be oxidized to hydrogen peroxide, nitrogen to
nitrate, or ferrous oxide to ferric oxide, to give some examples. Hydrogen
peroxide would be probably accompanied by molecular oxygen which forms from the
decomposition of the peroxide, making oxygen breathing possible. (The planet
would, however, still be uninhabitable, with the surface water and atmosphere
contaminated by hydrogen peroxide.) Nitrates would be accompanied by nitric
oxides. These could be breathed by alien life forms, and support fire, forming
a feasible alternative to oxygen.

Thus, we have a few alternatives to our oxygen: chlorine, nitric oxides, and
oxygen split off from hydrogen peroxide. However, Earth's oxygen-producing
photosynthesis is probably a more likely path than those.

Alternatives to water?

All life processes on Earth take place in an aquous environment. The origin of
life lies in the oceans, and even those organisms living on land or in the air
are internally mainly composed of water. On Earth, there is no life without
water. But is life imaginable where something else plays the role of water?

The most likely candidate for such an alternative solvent is ammonia. There is
no planet or moon where liquid ammonia exists in our solar system. However,
such worlds might exist in other solar systems. If a large enough planet
orbited the Sun at 3 A.U., it could have lakes or even an ocean of liquid
ammonia. What makes ammonia a likely candidate is its similarity to water. To
us, it is a poisonous gas with an evil stench, nothing like the liquid which,
to us, means life. However, from a more objective, abstract viewpoint, it shows
several commonalities. Ammonia, like water, is a polar molecule, with a
positive charge on one end and a negative charge on the other. Liquid ammonia
has similar chemical properties to water; salts (and water ice) dissolve in
ammonia, hydrophobic substances do not. There are even equivalents to acid/base
reactions. In light of this, it might be possible that life exists in liquid
ammonia or mixed ammonia/water environments. However, these environments are
very unlikely to spawn technological civilizations, because oxygen or similar
`breathing' and fire-feeding gases are not stable in a liquid ammonia
environment.

Methane and other hydrocarbons, as they are possibly present on Titan, however,
are entirely different. These solvents are non-polar, i.e. they behave entirely
different than water. Salts and other hydrophilic substances (e.g.
carbohydrates and water ice itself) don't dissolve in them, but on the other
hand, hydrophobic substances do. Biochemical reactions like those found on
Earth, however, heavily rely on the presence of a polar solvent and could not
work in a non-polar environment. If life exists in non-polar environments, its
biochemistry is probably more alien than we can imagine today.

Might there be any other liquids? Science fiction books and movies (such as
Alien) sometimes feature beings with acid-based biochemistry. Acids are polar
solvents; thus, such beings might be possible -- if something provides
significant amounts of a certain acid on a planet's surface. However,
concentrated inorganic acids tend to destroy all organic matter (except some
highly hydrophobic substances), but life based on water-diluted acids might
still be possible -- or inorganic life forms (see below). Concentrated sulfuric
acid has a boiling point of 3380C at Earth-normal pressure; on a planet like
Venus with its much denser atmosphere, it is much higher.

Alcohols are polar solvents, but to a lesser extent than water; they don't
dissolve salt, but dissolve some hydrophobic substances. However, being organic
in themselves, they are not very likely to occur in large enough quantities
before biological evolution sets in on a planet. Molten salts or even lava are
also sometimes mentioned in bad `science' fiction stories; however, these are
too hot to co-exist with organic matter. If there are life forms with molten
salt or lava for `blood', they must be of inorganic composition -- and this is
a matter of speculation.

Inorganic life?

Are there any other elements which could form as broad a variety of compounds
as carbon? Pulp-quality space fantasies often envision organisms based on
silicon instead of carbon. The authors of these works usually have silicon
equivalents to organic compounds in mind (every carbon atom replaced by a
silicon atom). This, however, is impossible. Silicon equivalents of some simple
compounds are known to exist, but they are highly unstable substances which can
only be sustained under laboratory conditions. Most organic compounds, however,
including the more complex ones, don't have silicon equivalents. But there
might be another kind of silicon-based life. Silicates show a wide variety of
forms (though not as wide as organic compounds). Molecules consisting of
chained silicate (SiO4), phosphate (PO4), sulphate (SO4) and aluminate (AlO4)
units might exist under suitable conditions. It is, however, entirely unknown
whether such conditions do occur in nature, and whether life could evolve from
such molecules. But if it exists, it might thrive in places where no organic
life could exist, such as Venus.

But even weirder things might exist. Physicists assume that atomic nuclei have
a shell structure similar to the electron hull. It is speculated that under the
immense pressure to be found at the surface of neutron stars, these shells
might interlock just as electron shells do and form `nuclear molecules'. No-one
knows what properties such nuclear molecules have and whether these are able to
form complex, self-organizing and self-replicating structures. There might even
be life forms which are not composed of matter as we know it at all, but of
supersymmetric or bosonic matter, or just standing wave patterns of creative
energy.

And finally, there might be life forms around that have evolved from more
`normal' life forms into something utterly bizarre. Intelligent computers,
networks of superconducting nanotechnical nodules or anything else. The
advanced descendants of star-faring races might be perfectly adapted to living
in open space such that they consider planets to be dangerous gravity traps and
avoid getting near them whenever they can; an Earth-like world might be as
hostile to them as the oceans or the reducing atmosphere of a juvenile
terrestrial planet are to us. They probably won't care about sun-like stars at
all (the only interesting thing about sun-like stars is the possibility of
Earth-like planets, anyway, so if you don't care about Earth-like planets, why
care about sun-like stars?), but cluster around those stars which put out
really big amounts of energy -- supergiants, recurrent novae and their like,
which are all places where our SETI researchers who focus on Earth-like life
never look.

-------------------------------------------------------------------------------
Jvrg Rhiemeier
e-mail: rhiemeir@ibr.cs.tu-bs.de

