Biased synthesis of L-amino acids and their polymerisation in repeatably
ordered sequences in lightning clouds
Stanley Miller's famous simulation of lightning in a flask containing a
reducing mixture of gases produced several amino acids.
But this compact laboratory setup did not simulate all of the conditions in
a lightning cloud. It it had done so the result of the experiment might have
been even more surprising than it was.
The setup did not create a strong electric field (up to 3 million volts per
metre) over a broad area and depth.
It did not create momentarily a very strong magnetic field.
It did not provide the heating and radiation caused by a great current. (The
main current in a stroke from a typical lightning cloud is about 20,000
amperes, but may be as big as 200,000 amperes. One author implies currents
as big as 300,000 amperes. Much of the research literature describes
mid-latitude lightning; but storms with energies orders of magnitude higher
develop in the tropics.)
It did not provide the freezing conditions ( -20 degrees Celsius) that exist
in the negatively charged 'pancake' of a lightning cloud. It did not provide
supercooled water droplets, small ice crystals, bigger ice crystals (Graupel
particles) and water droplets condensed on nuclei consisting of mineral
particles. In this negative 'pancake' all three phases of water (ice, liquid
and vapour) can coexist. Small droplets can evaporate, or grow into bigger
ones. Small ice crystals can sublimate, or grow into bigger ones.
Yet Miller's achievement in 1953 was so remarkable that everyone who
followed him assumed that the next problem was to work out what happened
when, in the real prebiotic world, his amino acids and adenine fell to the
earth in rain and accumulated in puddles and the primeval sea. The joke
about the 'prebiotic soup' was born: it was about a third as concentrated as
chicken bouillon. Everyone smiled and took their eye off the lightning
The channel containing the current of a lightning stroke is a few
centimetres wide. About 100,000 joules per metre are deposited in it. This
energy dissociates, ionises, excites and heats the air molecules (mainly
nitrogen, oxygen and water in the case of the present atmosphere) in the
channel. They are split into single atoms and, on average, each atom loses
one electron; this happens in a few microseconds. The temperature rises to
about 30,000 degreees C and the pressure to about 10 atmospheres. The
channel then expands in from 10 to 20 microseconds, this taking about 98 per
cent of the total energy. The supersonic expansion of the channel slows to a
sound wave, which is heard as thunder. About 1 per cent of the input energy
is stored in the particles in the channel, and about 1 per cent is radiated
as light and infrared radiation in the band of wavelengths from 4,000 to
A two-tesla magnetic field?
A long unbranched section of the lightning channel can be regarded as a long
straight conductor. To take as an illustrative example a large discharge:
before the channel has expanded, when it has a diameter of 4 centimetres and
carries a current of 200,000 amperes, the magnetic field at a radius of 2
centimetres from the longitudinal axis of the channel would be 20,000 gauss
or 2 tesla.
As it happens, this is the field strength that electrical engineers have
worked with for more than a century; the strongest they have ever been able
to economically achieve. It is very strong. The two-tesla field pulls
seven-kilometre freight trains and twists the screws of the largest ships.
(Stronger fields can be created by extraordinary means in the laboratory,
and cosmologists know of fields much stronger than these.)
For comparison, the magnetic field of the earth is about 0.5 gauss. A small,
inexpensive type of permanent magnet has a field of about 100 gauss.
It seems reasonable to choose the largest known values of the various
aspects of lightning clouds and strokes if they seem to be necessary to
originate life. Life only had to originate once (though it might have
originated often) and it had plenty of time to do it in. It would be helpful
to know the values in the lightning clouds of dust and gas that must have
formed as our planet accreted. Heavier dust particles would have fallen
through the clouds of small dust particles, and the heavy and small
particles must have acquired electric charges of opposite sign. When the
electric field had increased enough to break down the insulating atmosphere
a stroke must have passed between the cloud and the accreting planet, or
between different parts of the cloud.
A similar process (though perhaps not the only one causing electrification)
must happen in volcanic lightning clouds (one above the Sakurajima volcano,
Japan, is shown on the cover of 'Science', volume 275, 28 February 1997).
These clouds contain great quantities of dust at certain phases of an
eruption. During the eruption of the Krakatoa volcano in 1883 a ship 120
kilometres from it saw a black billowing cloud rise to about 25 kilometres.
A ship 64 kilometres away described the cloud as like 'an immense pine tree,
with the stem and branches formed with volcanic lightning'.
Accretion clouds and volcanic clouds might have contained mineral grains
that oriented and selected amino acids once formed, or catalysed their
polymerisation; iron sulphide, for example. (Water clouds might contain
similar mineral grains as the nuclei of droplets.)
Perhaps life originated at various times in all three types of cloud: the
accreting cloud, the volcanic cloud and the water-cloud. Indeed, it might
still be originating today above any volcano emitting a suitable reducing
mixture of gases, if such a volcano exists.
An enantiomeric excess without photolysis
Magnetic circular dichroism
I propose that L-amino acids (or at least an L-enantiomer core from which
L-amino acids could be derived by further chemical reactions) were
preferentially formed from molecules or atoms of the prebiotic atmosphere
that were in or near the channel of a lightning stroke.
In molecules, atoms or ions that have been excited by the energy of a
lightning stroke the orbits of their electrons would be affected by the
strong electric field in the cloud, which persists during a stroke because
only a part of the energy stored in the electric field goes into each
stroke. (A lightning 'flash' commonly consists of three or four strokes; one
flash to the ground had 26 strokes and lasted two seconds. Other observers
have counted 40 strokes in a single flash).
These excited molecules, atoms or ions would also be affected by the strong
magnetic field built up by the electric current of a stroke in the channel,
because the motions of orbiting electrons perpendicular to a magnetic field
are subject to Lorenz forces. The degeneracy of the eigen vibrations of the
electrons is removed by the magnetic field so that their energy levels are
split (this is the Zeeman effect, discovered in 1897). The oscillating
electrons emit radiation in all directions. That emitted in the direction of
the field is left circularly polarised. That emitted in the opposite
direction is right circularly polarised. (These definitions of direction
conform to the chemical convention, as opposed to that used in the physical
literature.) At various wavelengths the left circularly polarised radiation
might be more intense than the right, or the right circularly polarised
radiation might be more intense than the left; these differences in
intensity depending on wavelength are referred to as magnetic circular
Some researchers believe that at certain ultraviolet wavelengths magnetic
circular dichroism would cause ultraviolet radiation to destroy (photolyse)
more of one enantiomer than the other of an organic molecule.
Biased synthesis of L-amino acids
The effect of a strong magnetic field in removing the degeneracy of the
eigen vibrations and splitting the energy levels of the electrons of excited
molecules, atoms or radicals in or near the lightning channel would bias
those molecules, radicals or atoms toward the formation of one enantiomer as
they bonded with each other. I propose that in the formation of amino acids
in or near the channel an L-enantiomer core would be formed in preference to
a D-enantiomer core and, directly or indirectly, L-amino acids would be
formed in preference to D-amino acids.
In this case there would be no need to invoke biased photolysis of the
D-enantiomer to explain an excess of L-amino acids in a meterorite; the bias
would have been in the synthesis of the amino acids, not in their
A universal bias for L-amino acids
In this proposed biased synthesis of enantiomers the bias would be the same
anywhere in the universe for the same synthesis under the same conditions.
Thus this hypothesis differs from one of preferential photolysis, which
depends on the filtering out of certain wavelengths of energetic circularly
polarised radiation and the non-filtering of other wavelengths having the
opposite rotation. Because the particular wavelengths filtered out (by dust,
for example) in different places would depend more or less on chance,
preferential photolysis might result in an excess of one enantiomer in one
part of the universe and the other enantiomer in another part of the
Nature's rules about electric and magnetic fields are asymmetrical
It seems to be assumed by many writers on the origin of life that the
physical conditions in an accretion cloud or on our planet are on average
symmetrical, and that therefore an asymmetry in biological molecules - the
existence of only L-amino acids and D-nucleic acids, for example - must have
evolved by chance. But nature has only one rule relating the direction of an
electric field, the direction of the electric current that it drives and the
direction of the magnetic field that the current creates. And it has only
one rule relating the direction of a magnetic field and its effects on the
oscillations of electrons in orbit about a nucleus or a molecule. There is
asymmetry in these rules.
Other possible effects of a strong magnetic field
Enantiomers facing different ways
If amino acids were diamagnetic or paramagnetic they would become oriented
antiparallel or parallel with a strong magnetic field. Proteins are
sometimes made paramagnetic in laboratory investigations by replacing some
of their hydrogen atoms with fluorine atoms. It seems unlikely that this
would happen to amino acids in nature, except perhaps above a volcano
emitting fluorine gas.
Nonetheless it is illuminating to note that if fluorinated amino acids were
dissolved in water droplets in a volcanic lightning cloud they would firstly
be oriented by the persistent strong electric field of the cloud; then when
a stroke passed close to them they would also be oriented by the strong
magnetic field built up momentarily by the stroke. If both enantiomers were
present in the droplet, and if their electric axis and paramagnetic axis
were not coincident, one enantiomer would have its 'front' facing the
radiation and the shock wave from the stroke; the other enantiomer would
have its 'back' facing the radiation and the shock wave.
The effects of the energy received by the two enantiomers from these sources
might be different because of the different orientation of the molecules
with respect to the stroke. This is another example (though an unlikely one)
of how nature could place enantiomers in an asymmetric situation in a
lightning cloud. It raises the question of whether the strong electric field
in a cloud could slightly polarise the electron orbits of amino acids and
make them paramagnetic enough to be oriented by a strong magnetic field.
A chance of preferential photolysis
Any non-spherical diamagnetic, paramagnetic or ferromagnetic mineral
particles suspended by updrafts in the lightning cloud would be aligned in a
direction antiparallel (in the case of a diamagnetic particle) or parallel
with the direction of the magnetic field surrounding the lightning channel.
These aligned particles might by reflection circularly polarise
electromagnetic radiation travelling radially outward from the plasma in the
channel. (Radiation that began its journey parallel or antiparallel with the
magnetic field would already be circularly polarised.)
The particles might also filter out certain wavelengths.
This radiation would extend to ultraviolet frequencies in an accretion cloud
because of the presence of ions of high atomic number.
However, even if ultraviolet radiation were not present, a beam of
circularly polarised visible electromagnetic radiation close to the channel
would be very energetic and would carry more angular momentum than an
ultraviolet beam of the same energy, because the angular momentum of
circularly polarised radiation varies inversely with the frequency of the
radiation. Thus a very energetic beam of circularly polarised visible light
might have asymmetric effects on enantiomers of organic molecules in its
Polymerisation of L-amino acids in repeatably ordered sequences
Alignment by the electric field
Having produced L-amino acids the lightning cloud might have aligned them
'head to tail', that is, carboxyl group to amino group, and might have
arranged them in order in the line. The strong electric field built up in
the cloud would act on amino acids dissolved in water droplets. Because the
carboxyl group is acidic and the amino group is basic a proton is
transferred from the former to the latter in water, leaving the carboxyl
group with a negative electric charge and giving the amino group a positive
charge. In a strong electric field in a cloud the carboxyl group must be
attracted toward the effective positive part of the cloud or earth and the
amino group must be attracted toward the negative part of the cloud.
(I say 'effective' because typical lightning clouds have a main positive
charge at the top, a negative part in the middle and a small positive part
at the bottom. A lightning stroke to ground connects the negative middle
part of the cloud with the earth, which is relatively positive. Electric
fields within a cloud are complex. However, whatever the orientation of the
field may be at a particular place, it will affect all amino acids dissolved
in water in a consistent way.)
Because water molecules are polar they would be aligned by the electric
field of the cloud; this might strengthen the alignment of the amino acids
dissolved in the water.
Ordering of sequence by the electric field
Some amino acids have extra charged groups in water, so that they have a net
positive charge or a net negative charge (depending on the nature of their R
group). Those with a net positive charge would be pulled bodily through the
water droplet (though remaining inside it) toward the negative part of the
cloud, and those with a net negative charge would be pulled bodily through
the droplet toward the positive part of the cloud or earth. Amino acids with
no net charge would be jostled toward the middle as the charged ones found
their respective places.
Then the water of a small droplet might partly or completely evaporate; in
freezing dry air this could happen quickly. The aligned and ordered amino
acids might be 'dried' and remain as lines of head-to-tail molecules on the
surface of the mineral particle on which the water droplet had condensed.
This mineral particle might be one that catalysed the elimination of water
from the juxtaposed amino group of one molecule and the carboxyl group of
another molecule in a line when a nearby lightning stroke supplied the
Energy for polymerisation from inductive effect
The quickly building magnetic field might induce forces on the positive
amino group and the negative carboxyl group that helped to eliminate a water
molecule and form a chemical bond. As the circular magnetic lines of force
of the building field travelled outward they would induce the electric
charges that they 'cut' to move in the direction perpendicular to the line
of magnetic force and perpendicular to its direction of motion. A positively
charged amino group would be induced to move in one direction and a
juxtaposed negatively charged carboxyl group would be induced to move in the
A short time later, as the magnetic field collapsed, they would be induced
to move in the 'switched' directions. Either during the building-up or the
collapsing of the magnetic field the amino group and the carboxyl group
might be forced together by induction with enough energy to enable them to
eliminate a molecule of water and form a chemical bond.
The same forces would be induced on all of the charged amino and carboxyl
groups in an aligned, ordered sequence of amino acids. Thus a complete line
of amino acids might be polymerised into a short protein.
A variety of ordered sequences
These short proteins, produced by the trillion in the cloud, would not all
be alike. Some might be longer than others. Their arrangements would depend
on which amino acids had been present in their 'mother' water drop, how many
of them there were, how big the drop was before it evaporated and what
mineral 'nucleus' was present. (The drop might alternatively have been
frozen and then sublimated; in this case the aligned and ordered amino acids
would have been 'freeze-dried'.)
Positive, negative and neutral domains
Some proteins (I shall call them proteins to avoid using more cumbersome
names) might have charged amino acid residues at each end (positive at one
end, negative at the other) and uncharged amino acid residues in the middle,
following the order I described above. When redissolved in water these would
fold consistently into specific shapes, depending on the charges of their
parts and other effects, including hydrogen bonding.
Others proteins might have charged amino acid residues at only one end, and
uncharged residues in the middle and at the other end. In some of these
proteins the uncharged residues might be hydrophobic, 'water-hating'. If a
lot of the latter kind of protein gathered together after being redissolved
in a water droplet they might have formed a two-layered membrane: an
essential part of a living cell.
Thus tiny sacs formed by two-layered protein membranes and containing water
and other proteins folded in various configurations might have been formed
in lightning clouds and fallen in raindrops onto the surface of the
A laboratory built to withstand a blast
The conditions I have described would not be difficult to simulate in a
well-funded laboratory. Stanley Miller and his supervisor could hardly have
been expected to simulate them. But 48 years have passed since Miller
synthesised amino acids in a 'prebiotic' atmosphere (and, in a remarkable
coincidence, Watson and Crick described the structure of DNA). No biased
synthesis of L-amino acids, and no polymerisation of L-amino acids into
various short proteins each having a repeatable sequence of amino acid
residues, has been achieved in that time. (By 'repeatable sequence' I mean
each of many specific sequences that will be produced each time the
experiment is done.)
What kind of experimental chamber and atmosphere would be required to
simulate these conditions?
An electric field between plates
A strong electric field could easily be set up between big metal plates at
the top and bottom of the chamber.
Reducing mixtures of 'prebiotic' gases
Various reducing mixtures of gases could readily be sent into the chamber
(there seems to be no consensus on the composition of the prebiotic
atmosphere, so various plausible mixtures should be tried, including
sulphurous mixtures that might have been emitted from primeval volcanoes).
Water as liquid, solid and vapour
Conditions in which water coexisted in its three states as liquid droplets,
ice particles and vapour could be provided by freeze-drying equipment. This
dries the atmosphere in one part of a chamber, but it humidifies the
atmosphere in another. It is not a simple refrigerator. A small fan might be
needed to provide enough updraft to prevent the droplets falling.
A puff of mineral powders
Mineral particles to act as nuclei for water vapour to condense on could be
puffed into the experimental chamber. In some experiments these particles
should include minerals likely to exist in an accretion cloud, such as
nickel and iron.
Lightning stroke and strong magnetic field
A sufficiently strong cylindrical magnetic field (not necessarily as strong
as 2 teslas) could be momentarily built up by the sudden discharge of a
large bank of capacitors through the prebiotic atmosphere in the chamber.
The insulating property of the atmosphere would first have to be broken down
by a high-voltage, small-current electric discharge between pointed
electrodes at the top and bottom of the chamber; this would form a 'channel'
of deposited charge and ionised atoms. This discharge would be immediately
followed by a lower-voltage, great-current discharge through the same
(Experimental work on confinement of plasmas for nuclear fusion has
accumulated much experience on the use of banks of capacitors to provide
currents of several hundred thousand amperes in pulses lasting a few
The passage of so much electric current through the channel would have the
same effects on the gas molecules as a lightning stroke, dissociating,
ionising, exciting and heating them. The temperature in the channel would
rise to about 30,000 degrees C and the pressure to about 10 atmospheres.
Electromagnetic radiation would travel outward from the channel. Then in
from 10 to 20 microseconds the channel would expand supersonically, this
taking about 98 per cent of the energy of the discharge.
The chamber would have to be built strongly enough to withstand the blast.
It should be wide enough to allow the supersonic shock wave to decay to a
The finale: gentle rain
After the experimenters had sent a sufficient number of simulated lightning
strokes through the prebiotic atmosphere in the chamber the fan would be
switched off and the conditions of relative humidity and temperature would
be adjusted so that small droplets of water began to condense, coalesce and
fall like rain, or trickle down the walls of the chamber.
This water would carry dissolved in it whatever organic molecules had been
formed. It would be drawn off from the floor of the chamber.
Then the task of identifying these molecules could begin.
(See a version with more easily read typography at: