How do we know the gases that were in the primitive atmosphere at the time the earth was forming?

Answered by John Mercer, Assistant Professor of the Practice, Duke University Biology Department.

Short answer:

The Earth’s atmosphere has had a complex and evolving character.  It is unlikely we will ever know the exact composition of paleo-atmospheres, but the types of gases we believe were present at various times are known from the chemical makeup of the Sun, from materials left over from the formation of the Solar System such as comets and primitive meteorites, and from the typical composition of molecular clouds out of which stars form.  To origin of life researchers the precise composition is less important than knowing whether the atmosphere is strongly reducing since such atmospheres are one of several potential sources of organic compounds.  Because of its small atomic mass, its current distribution in the Solar System, and early impacts of extra-planetary objects, it is unlikely that the most reducing of gases, hydrogen, was present in abundance. Reduced gases can still be released to the atmosphere by volcanic release of gases dissolved in the Earth’s interior if significant amounts of iron are present in the mantle, but isotope studies suggest that much of the Earth’s iron was sequestered by the core early in the planet’s history so it now appears more likely that volcanic gases were similar to the present and more nearly neutral.  However, some reducing gases can come from cometary and primitive meteorite impacts.  We know that such impacts would have been more frequent on the early Earth from looking at the cratered surfaces of the Moon of a similar age.

Long answer:

The answer to your question is not well known, but there is no shortage of inferences on the composition from indirect evidence.   One factor affecting how well we can infer the state of the atmosphere is the absence of any geological record with proxies for the state of the atmosphere before about 4 Ga.  Absence of a record is due in part to continued bombardment of the Earth throughout the Hadean (4.5 - 3.8 Ga) together with subduction of the rock record throughout Earth history.

The early stages of Earth formation would initially have been a reasonably gentle accretion of dust and small objects into planetesimals, followed by much more violent accretion of planetesimals into the Earth.  Impacts in the last stage of accretion would have made the Earth very hot and molten with an additional contribution of thermal energy from the decay of radioactive elements in the Earth (They were more abundant then.).   Recent isotope studies suggest that core formation would have occurred within 30-50 million years post accretion.  Throughout this period temperatures would have been high enough for the Earth to be molten.  With such high temperatures, and because of the high abundance of silicates in the Earth, the atmosphere during Earth formation was most likely a hot silicate vapor (sort of like vaporized glass).

As mentioned above, the Earth would have been subjected to continued bombardment even after the Earth was formed.  Many of those early impactors would be energetic enough to partially or completely melt the Earth's surface even as it had begun its cooling phase. We also have strong evidence, principally from material brought back from the Apollo missions to the Moon, that tell us that the Moon was very likely formed from the impact of the Earth with a Mars-sized body some time between about 4.5 and 4.4 Ga.  With respect to atmosphere formation, the point to take away from this information is that the first atmospheres would likely have had an episodic existence since impactors would have repeatedly blown away substantial parts of the atmosphere, and vaporized part or all of any oceans that formed.   Nevertheless, isotopic analysis of small crystals of zircon found in rocks from Australia has provided evidence for liquid water as far back as 4.4 Ga.  A corollary would be that there was at least some water vapor in the atmosphere.

It is well known that most of the material that makes up the molecular clouds out of which stars and planetary systems form are the gases hydrogen and helium.  Information from astrophysical studies, indicating that Jupiter and Saturn have large amounts of these two gases, once were used for early arguments that the early Earth would have had a reducing atmosphere with significant amounts of hydrogen.  However, information from astrophysical studies tells us that in the early stages of the formation of a planetary system, the inner part of the accretion disk would have been rapidly cleared of gas by the solar wind.  In light of that understanding we are able to explain why Jupiter and Saturn have atmospheres with so much hydrogen and helium and why the inner planets (Mercury through Mars) have so little – but that's another story. From the point of view of Earth's atmosphere, if it ever had a hydrogen rich atmosphere, it would have likely been gone before the Earth was cool enough for processes leading to life got under way.

Molten rock on and in the early Earth would have contained reducing (e. g. hydrogen), neutral (e. g. nitrogen), and oxidized gases (e. g. carbon dioxide) that would have been released into an early atmosphere. Such gases would be expressed by early volcanism, and, while the proportion that is reducing is unknown, the potential for outgassing significant amounts of reducing gases could be high if the Earth’s mantle were reducing.  However, that now seems unlikely due to the early formation of the Earth's core. The Earth's core is predominantly iron that settled out of the mantle during core formation.  Because of the ability of iron to produce reduced gases from relatively oxidized gases, a high abundance of iron in the mantle would have favored reduced gases in volcanism, but the rapid sequestering of iron to the core made the mantle more nearly neutral rather than reducing, so it is likely that the gases of volcanism were actually more like what they are now - neutral to slightly oxidized.  The current view of the state of the atmosphere during the period in which prebiotic evolution occurred is that it was likely neutral to weakly oxidized with significant amounts of carbon dioxide, nitrogen, and water vapor.

There remain several other potentially important sources for an early atmosphere: comets and primitive meteorites (chondrites).   Especially from recent interplanetary encounters (e. g. Tempel 1), we have confirmed that comets have substantial amounts of water ice, and ices of relatively reducing gases like methane and carbon monoxide, and of oxidized gases like carbon dioxide.   Significant amounts of organics are present as well.  Cometary impacts continue up until the present (recall the recent one with Jupiter).  They would have been much more frequent in the early Solar System, and are likely an important source of water, reducing gases and organics. Some chondrites also contain volatiles that could have contributed to the Earth's early atmosphere.  Isotopic studies on noble gases point to a greater contribution from comets than chondrites.

In any case, contributions from outgassing of the mantle and/or extra-planetary impacts of comets and chondrites must have become more important as the Earth cooled and violent impacts of big rocky objects declined in order to develop a permanent atmosphere.  No free oxygen would have been present since it is a highly reactive gas and would have formed compounds with other elements rather than remain in the atmosphere.  Free oxygen occurs much later in the atmosphere's history, and is another story.

I hope that gives you some idea of the complexities and current limits on inference about the state of the early atmosphere, and, I hope, gives the atmosphere a much more dynamic and evolving character. 

Links:

Haze over Saturn moon mirrored on early Earth Experiment shows ample food here for first living organisms.

http://sfgate.com/cgi-bin/article.cgi?f=/c/a/2006/11/07/MNG8MM7BJN1.DTL

Back to top

 

Symposium Home         Speaker Biographies         Directions to Sigma Xi