As mentioned several times in Earth Pages News, geologists have been slow to accept that the Earth’s evolution has been substantially affected by impacts of extraterrestrial bodies.  In hindsight, this stubborn scepticism seems perverse.  The discovery of impact-induced melt spherules in the Late Triassic sediments of SW England (see Britain’s own impact in EPN, December 2002) went almost unnoticed.  However, there is still an entrenched view that nothing really big has happened.  When similar spherule beds were reported from the Early Archaean greenstone belts in Australia and South Africa in 1986, and deduced to have formed by an impact, the authors were pounced on by those who thought they could plausibly explain the very odd rocks by unremarkable, Earthly processes. How satisfied Donald Lowe and Gary Byerly, of Stanford and Louisiana State Universities must be to find their view now proven beyond doubt, and to share in publishing the evidence.  The proof comes from isotopic studies of three spherule beds in the 3200 Ma-old Barberton greenstone belt in South Africa (Kyte, F.T. et al. 2003.  Early Archean spherule beds: Chromium isotopes confirm origin through multiple impacts of projectiles of carbonaceous chondrite type.  Geology, v. 31, p. 283-286).  Chromium isotopes in the rocks are so unearthly, that explaining them requires that they contain up to 60% of extraterrestrial material, probably from carbonaceous chondrite impactors.  Compared with the global spherule-bearing and iridium-rich K/T boundary layer (3 mm thick on average), that is the ejecta from the Chicxulub impact, the Barberton beds are much thicker (10-20 cm).  The authors estimate that, if the Barberton layers are globally representative, the impactor responsible for their formation could have been 50 to 300 times more massive than that which terminated the Mesozoic Era.  Besides that, three such layers formed within 20 Ma, and that suggests bombardment flux more than ten times that late in Earth evolution.

Triggering core formation at the microscopic level

Since Birch’s discovery in the 1950’s that the Earth’s excessive density compared with exposed rocks could be explained by a metallic, iron rich core, whose presence was detected by studies of seismic waves, there have been many explanations for core formation.  Some regarded the process as a slow accumulation of iron-rich melt as it sank from the mantle, others that it formed during Earth’s initial accretion from the iron-rich parents of metallic meteorites.  Lead and tungsten isotope studies indicate clearly that the core formed very early in Earth’s evolution, taking as little as 30 Ma.  However, for such a vast mass to have quickly segregated from the rest of the Earth poses awesome mechanical problems.  Alloys of iron, nickel and sulphur do have much lower melting temperatures than silicate minerals, and planetary accretion releases gravitational potential energy.  That serves to heat up a growing planet, but core-forming materials would melt long before the dominant silicates that envelop them, if indeed mantle materials did melt substantially.  So, at the centimetre scale of rocks, a melt fraction, however dense, would have to migrate and accumulate in globules with sufficient gravitational potential to sink through the viscous early mantle.  The boundaries of pores in which melts form are critical.  If the angles between silicate facets and melt-filled pores are large, tiny amounts of molten metal cannot become interconnected and migrate, unless the silicates begin to melt too or are actively deformed.  Since coexisting silicate and metal melts are not supported by geochemical evidence and deep planetary interiors are probably static, the fact that the interfacial angles of crystalline minerals are high poses quite a problem.  Geochemists at the University of Yokohama in Japan have performed complex experiments at high pressure and temperatures to simulate likely conditions during planetary accretion (Yoshino, T. et al. 2003.  Core formation in planetesimals triggered by permeable flow.  Nature, v. 422, p. 154-157).  They discovered that if metallic melts account for more than 5% by volume of the accreting body, then this melt can percolate through the solid rock, because the angles separating melt and solid fall below the critical value of 60º.

The implication is that even quite small planetesimals (>30 km radius) can quickly develop metallic cores, using energy released by the decay of short-lived isotopes that were plentiful early in Solar System history.  This is borne out by studies of metallic meteorites  Of course, the immense gravitational energy released by accretion of larger planetary bodies would result in the same differentiation, but if they formed by accumulation of smaller differentiated bodies there is no need to postulate within-planet processes on the microscopic scale.  The core would be “pre-manufactured”, only requiring blending of many smaller cores of accreting planetesimals

See also: Minarik, B. 2003.  The core of planet formation.  Nature, v.  422, p. 126-127.