Predicting Primordial Weather
What was the weather like billions of years ago?
This seamingly straightforward question defies current prediction. An answer requires not a forecast, but a complex backcast. Whether the early Earth had a familiar or exotic climate poses a mystery, one not yet determined by what scientists can reconstruct from combining chemistry and geology. But the question itself underpins a controversy that affects the scenario of how life might have begun here, and potentially elsewhere in the universe.
|"… some warm little pond, with all sorts of ammonia and phosphoric salts, light, heat, electricity etc…", Charles Darwin, on the origins of life in tidal pools|
Charles Darwin first supposed the early Earth was a basic soup of simple organic molecules, which when sparked by sunlight, heat and lightning might have created more complex biological building blocks. He supposed that this soup was basic literally, meaning that it was neither acidic nor neutral in pH–the measure of a solutions ability to react to neutralize hydrogen ions. The modern scientific view is that the earth’s early atmosphere was at most very mildly reducing and more likely neutral. This interpretation poses a big problem for how biochemists at first supposed that life might begin.
The classic view of life’s first molecules date to fifty years ago, when two University of Chicago scientists–Harold Urey and Stanley Miller–first created protein chains from ammonia and other simple organic molecules. In 1953, t hey began their two page paper in Science:
"The idea that the organic compounds that serve as the basis of life were formed when the earth had an atmosphere of methane, ammonia, water and hydrogen instead of carbon dioxide, nitrogen, oxygen and water was suggested by Oparin and has been given emphasis by Urey and Bernal. In order to test this hypothesis…"
In his experiments to simulate the early Earth in a test tube reaction, Miller found that at least 10 percent of the carbon could be converted into a small number of organic compounds and about two percent went into amino acids–the key building blocks of proteins (peptides) and life. Later Carl Sagan would do many experiments varying the chemical percentages, but described the Miller-Urey experiments as "the single most significant step in convincing many scientists that life is likely to be abundant in the cosmos."
But the problem is the organic chemistry doesn’t work out unless the early Earth had what is known as a reducing atmosphere, or ammonia was prevalent. The reactions don’t trigger the right organic chemistry. If the Earth more likely was rich in nitrogen and carbon dioxide– rather than hydrogen, methane and ammonia–, then any amount of sparking delivers a mere drop of organic byproducts. The primordial soup is too dilute. Stanley Miller told Astrobiology Magazine: "I have not found an alternative to disprove the need for a primitive reducing atmosphere."
Even today, only a few definitive things are known about what the Earth might have been like four billion years ago. It is thought that the early sun radiated only 70 percent of its modern power. No free oxygen could be found in Earth’s atmosphere. The rocky wasteland lacked life. Absent were viruses, bacteria, plants and animals. Even the temperature itself is uncertain, since three schools of thought today maintain that the Earth could have been alternatively frozen, temperate or steamy.
Until roughly 3.9 billion years ago, swarms of comets and meteorites whacked the young Earth often enough to occasionally vaporize the surface zones of the oceans and erase any life residing there. The earliest known evidence of microbial life on Earth comes from carbon isotope patterns in 3.85-billion-year-old Greenland sediments.
Astrobiology Magazine had the opportunity to talk about early terrestrial chemistry with Dr. Ian Miller, a Fellow of the New Zealand Institute of Chemistry and a member of the Editorial Board of the journal, Botanica Marina.
Miller holds that the conventional view of the early Earth as ammonia-poor needs to be revisited, and he draws on interpretations for what some of the oldest rocks tell us. His account takes a journey from our planetary neighbors–Mars and Venus–to the deep gold mines of South Africa, where ammonia-rich rocks are being uncovered today.
Astrobiology Magazine (AM): Why do you think there needs to be a change to the standard explanations?
Ian Miller (IM): If we accept that the rocky planets accreted from dust and material that had previously been heated to 2000K, then we are asking dust, stone and small pieces of iron to accrete following collisions. Yet collisions appear to smash asteroids. Mankind has a lot of experience of firing lumps of iron at stone, and never in the entire history of such warfare have walls accreted into something bigger. Nobody has ever conducted a ballistic experiment that results in accretion, and they will not, because stone and iron result in elastic collisions. The result: no planets.
|Rocky planets accrete or aggregate debris from the early solar system |
What is missing is the chemistry of the planetary disc.
If any mix of carbonates (CaO), aluminum oxide (Al2O3) and silicon oxide (SiO2) is heated to 2000 K a hydraulic cement results, which seems to solve this problem remarkably well, and also gets water onto the rocky planets.
In one further stroke, we address a problem nobody seems to have noticed: why else would calcium (Ca) and aluminum (Al) be enhanced over magnesium (Mg) on Earth by about 2 orders of magnitude (100 times) compared with nucleosynthesis probabilities?
But if you accept that, you have to accept that at the same temperatures, nitrides, carbon and carbides must be formed, which are quite solid. This gives the reduced materials which, at one further stroke provide a simple solution to biogenetic precursors, the different compositions of the atmospheres of the rocky planets, and the problem of martian fluvial systems.
Better still, the theory is testable, if you can send the right equipment to Mars.
AM: In your research, why is consideration of a reducing atmosphere important to the early development of organic precursors?
IM: Everybody agrees precursors have to come from reduced species. Either you start with ammonia, or with nitrogen that has to be reduced. Now, nitrogen can be reduced by rocks such as olivine, as can carbon dioxide (CO2). The problem is that at a hot vent these will be rapidly dispersed, and worse, nitrogen gas is not very soluble in water, so it is always very dilute. It is extremely difficult to see how enough of anything can be brought to the same place to form precursors.
|Earth today in proportions of chemical elements that are biologically significant, as measured by the departing Mars Express spacecraft expected to encounter Mars in December. Credit: ESA/Mars Express|
Nitrogen oxides appear to solve the dilution problem, but they do not, because they react with ammonia to make nitrogen. But there is a far worse problem.
My first entry into this field was to test the hypothesis that CO2 could be photoreduced with ferrous ions in solution. It seemed likely that they could, but then it occurred to me that to reduce CO2 and nitrogen (N2) to make one amino acid, dozens of iron Fe(III) ions would be liberated. A quick test of putting Fe(III) in a glucose solution and leaving it in sunlight showed that the carbon-carbon bonds of the sugar were broken in good quantum yield. Precursors and Fe(III) are simply incompatible in sunlight.
AM: How does ammonia play into the prebiotic chemistry?
IM: Ammonia, in my view, has four roles.
Firstly, it provides the nitrogen for amino acids, nucleic acids (which are very nitrogen rich) and also, perhaps more importantly, agents such as FADH and cofactors. It is inconceivable to me that the oxidation/reduction agent should be nitrogen rich in an ammonia-poor environment when quinones provide a perfectly acceptable alternative.
Secondly, it keeps the destructive ferric ions insoluble in water. (Interestingly, insolubilised ferric oxide can photostabilize some organic molecules.)
Thirdly, it provides a weakly alkaline medium. This is important because the real key to life, in my view, is phosphate chemistry, and while this is not properly understood yet, the relevant chemistry seems to operate best at slightly alkaline pH.
Finally, at least for some reactions ammonia should make it easier for crude peptides to function as catalysts, through offering alternative hydrogen-bond options to water at constant energy (terminal versus cross-linking) and may well modulate electric fields during catalysis, e.g. by being a proton acceptor.
|Terrestrial options for early climate. Early earth, snowball, cauldron or temperate?Credit: NASA|
AM: The earliest earth–before microbes and plants–probably lacked much free oxygen. When does oxygen (or ozone) first enter the atmosphere?
IM: My view on oxygen is probably similar to everyone else’s. Water would be photolyzed, and oxygen would form. There would be net oxygen production after the initial reduced components, such as carbon monoxide were oxidized, and much more when photosynthesis got underway.
AM: Does the controversy over the early terrestrial climate bear on this work, such as whether the sun was faint, whether volcanic degassing enterred early or late to add significant carbon dioxide, etc.?
IM: Astronomical evidence seems fairly clear that the early sun was faint. There seems to be reasonable evidence volcanic degassing occurred somewhat after 4Gy (Gy=one billion years) on Mars, and I assume the same for Earth. If we assume the Moon was formed in a massive collision, further degassing can’t occur until magma can get through the outer layer. Magma is slow moving. A recent calculation in Nature indicated that magma takes 1 Gy to move a thousand kilometer. It also takes a while for plate tectonics to get going. The continents have to form, and to float on the underlying magma, they have to be coherently lighter, so I assume there will be a period during which rocks sort themselves as to density, and of course below that the thermal plumes have to get going.
Of course, the Earth may have retained volatiles from the outer shell. If so, there would be CO2 present, and this would be weathered to form calcite. In fact I suggest the runaway greenhouse effect on Venus may not be due directly to its being closer to the sun, but rather indirectly because it is more calcium deficient through, being in a hotter zone initially, setting less cement and having more aggregate during formation. This would require it also to have less water, and that seems to be observed.
As a further comment, I suggest that at primordial times Earth may have had a warmer surface than Venus. The Moon may have been close to the Roche limit, in which case tidal energy would have been very much greater than now, and could have been the predominant form of heating.
|A rock sample from Akilia Island off the coast of Greenland. A zircon crystal determined to be 4.4 billion years old may make it the world’s oldest known terrestrial material. Zircon is a mineral commonly used to determine the geological age of rocks. Chemical analysis of this grain suggests that the Earth was cool enough to have water, a hydrosphere and, possibly, life much earlier than previously thought.|
AM: The oldest terrestrial rocks, (zircon), have been oxygen radio-isotope dated to 4.4 Gy (also using uranium isotope ratios, U/Pb), with some evidence for liquid water early in the Earth’s history . Can you comment on whether those provide any tests for your research?
IM: It strongly suggests that there was early water, presumably liberated from a magma ocean after the moon-forming event. If so, it would have been accompanied by CO2 then, because the high temperatures would have oxidized any reduced gases. However, I believe this would have been weathered before significant volcanism got started. There would also be some nitrogen (N2) in the atmosphere, but that is inert.
When I talk about ‘initial atmosphere’ I really mean, ‘at the period of biogenesis’.
AM: Even in the presence of substantial ammonia, carbon dioxide poisons much of the prebiotic synetheses currently understood. Is there a way to find ratios of ammonia to carbon dioxide (NH3/CO2) in your studies, or if not, an estimated acidity or pH?
IM: The atmosphere/volatiles I am suggesting are a little different from what is generally considered.
The carbon arrives as carbon, and thus to carbon monoxide (CO) and methane (CH4), but also as carbides, thus to acetylene, acetaldehyde, etc, and as cyanamide, which, as an aside, offers the potential for carbodiimides. The importance of these is that they catalyze condensation reactions in water, which, translated for non-chemists, means peptides from amino acids.
Even more importantly, there is a direct route to formamide, thus to hydrogen cyanide, which is essential. Also, within the Earth Fischer Tropsch reactions should proceed, which will produce long chain hydrocarbons terminated with an alcohol group. If these can react with pyrophosphates, phosphate surfactants form. These would form micelles, effectively the first primitive cell walls, which also catalyse condensation reactions. So, in a sense, my proposed set of volatiles get around the problem of the first question by providing an alternative.
The ratio of NH3/CO2 will depend on the availability of calcium silicate, because that will "fix" CO2–however on Mars the CO2 ratio should have been an insurmountable problem, and the proposition predicts large amounts of urea (or whatever it turns itself into over 3.5 GY) beneath the martian surface.
In my wildly optimistic moments, I am hoping the [Mars soft-lander] Beagle 2 will find some buried nitrogenous material.
AM: You mention the Barberton Mountain Land rocks (3.4 Gy, stromatolites) from northern South Africa, have these been compared to Pilbara (Western Australia, 3.5 Gy), Belingwe (Zimbabwe, 2.9 Gy) or Isua (3.8 GY Greenland) for a global perspective?
IM: Out of sheer self-interest, let me put in a plug for more attention to theory. The short answer to the first question is, apparently, no, but I am unsure whether the problem is the nature of the samples, which are not exactly common, or that people were not looking for these components.
The Isua rocks apparently occluded methane, in accord with my proposition, but it is not clear what else.
The study on the Barberton water occlusions were intended to study salt compositions, and because these were measured by ion exchange, and were extracted in acid solutions, ammonia corresponding to several per cent of the nitrogen inventory was found (given certain assumptions as to how much water was degassed).
AM: Does you research affect the figure often quoted that life took 600 million years, and followed less the chemistry than the end of the giant asteroid storms on Earth, called the heavy bombardment?
|The painting titled "K/T Hit" by artist Donald E. Davis. Devasting impacts reshaped the solar system until planets began to rearrange the debris. Image Credit: Don Davis|
IM: Since my assumed timeline for volcanism commences after the Great Bombardment, life could not have started until after that.
I also think that the statement "life started at about 3.8 Gy BP" needs examination. If we assume my proposition, does it get to life? What comprises life? A micelle with nucleic acid inside it? Such a micelle that has made cell wall adjustments with peptides in the micelle wall? Such a micelle that, through osmotic pressure, has learned to split the double helix and reproduce? Such a micelle with porphyrin in the cell walls that photochemically makes ATP? There seems to be an assumption that life started more or less as modern single cells, then did nothing much for 2-3 Gy.
I rather suspect it did not start like it is now. What is the probability of getting 300 amino acids in the precise sequence to form an enzyme to do something that could not be done without the sequence? And how many times does this miracle have to be performed for an organism to do anything useful?
In my view there would be a rather long period of time during which micelles changed from being vaguely reproducing bags of chemicals to becoming life forms as we know them, during which period of time considerable refinement of entities such as ‘enzymes’ would take place. Thus the first enzymes would be quite inefficient compared with what they would evolve to, but they would do a job.
 S.J. Mojzis, T.M. Harrison, R.T. Pidgeon. Oxygen-isotope evidence from ancient zircons for liquid water at the Earth’s surface 4,300 Myr ago. Nature 409. 178 – 181, 2001
Related Web Pages
Primordial Recipe: Spark and Stir
Dinner with Darwin
Life’s Recipe Card
Earth’s Oldest Fossils Reverse Course
Diamond Time Capsules
When Did Life on Earth Begin? Ask a Rock
Earth’s Oldest Mineral Grains Suggest an Early Start for Life
Advances in our Understanding of Life
Life Under Bombardment
Journey to Center of Earth Review (NSF)