Vol 56, No 13 (2018)

The generation of crustal material and the formation of continental crust with a thickness of ≈40 km involve different physical mechanisms operating over different time-scales and length-scales. This review focusses on the building of a thick crustal assemblage and on the vertical dimension where the consequences of gravity-driven processes are expressed most clearly. Continental crustal material is produced by a sequence of crust and mantle mlelting, fractionation of basaltic melts and sinking of dense mafic cumulates. The repeated operation of these mechanisms over tens of million years leads to a thick stably stratified crust. We evaluate the main mechanisms involved from a physics perspective and identify the key controls and constraints, with special attention to thermal requirements. To form magma reservoirs able to process significant magma volumes and to allow the foundering of mafic cumulates, melt must be fed locally at rates that are larger than that of average crustal growth. This requires the temporary focussing of magmatic activity in a few centers. In some cases, foundering of dense cumulates does not go to completion, leaving a deformed residual body bearing tell-tale traces of the process. Crust must be thicker than a threshold value in a 30–45 km range for mafic cumulates to sink into the mantle below the crust. Once that threshold thickness has been reached, further additions lead to increase the proportion of felsic material in the crust at the expense of mafic lithologies which disappear from the crust. This acts to enhance radiogenic heat production in the crust. One consequence is that crustal temperatures can be kept at high values in times of diminished melt input and also when magmatic activity stops altogether, which may lead to post-orogenic intracrustal melting and differentiation. Another consequence is that the crust becomes too weak mechanically to withstand the elevation difference with neighbouring terranes, which sets a limit on crustal thickening. The thermal structure of the evolving crust is a key constraint on the overall process and depends strongly on radiogenic heat production, which is surely one of the properties that make continental crust very distinctive. In the Archean Superior Province, Canada, the formation of juvenile continental crust and its thermal maturation 2.7 Gy ago can be tracked quite accurately and reproduced by calculations relying on the wealth of heat flow and heat production data available there. Physical models of magma ascent and storage favour the formation of magma reservoirs at shallow levels. This suggests that crustal growth proceeds mostly from the top down, with material that gets buried to increasingly large depths. Vertical growth is accompanied by lateral spreading in two different places. Within the crust, magma intrusions are bound to extend in the horizontal direction. Deeper down, lateral variations of Moho depth that develop due to the focussing of magmatic activity get relaxed by lower crustal flow. This review has not dealt with processes at the interface between the growing crust and the mantle, which may well be where dikes get initiated by mechanisms that have so far defied theoretical analyses. Research in this particular area is required to further our understanding of continental crust formation.

The paper presents first nitrogen isotope analyses in shungites sampled from the core material of boreholes and from adits in the Onega region, Russia. The samples represent all varieties of the rocks, both stratified and migrated, with concentrations of reduced carbon ranging from a few percent to almost 100%. Analyses were carried out on a mass spectrometric complex (designed at the Open University), with the simultaneous measurements of the isotopic composition and concentrations of nitrogen and carbon using the step oxidation technique. It was determined that nitrogen is released mostly when the carbon is oxidized. The individual samples differ from one another in nitrogen isotopic composition and the nitrogen/carbon ratio, which indicates that these samples contain three distinct organic components of biological origin. Variations in δ¹⁵N and C/N that correspond to mixing of organic matter of the three types have been also observed during stepped oxidation. One of them has a relatively low C/N of approximately 200 and contains isotopically light nitrogen (δ¹⁵N up to –10‰). This component is thought to be primary, produced when the shungite rocks have been formed in a biological cycle involving reduced nitrogen species and formation of chemoautolithotrophic organisms, similar to what nowadays takes places in hydrothermal systems related to volcanic activity in spreading zones in the seafloor. Another component of the organic matter, with a relatively high δ¹⁵N from 0 to +3‰ and C/N of approximately 1000, was likely derived from the primary one in the course of diagenesis and metamorphism, due to nitrogen losses, resulted in heavier isotopic composition of the residual nitrogen. Finally, during post metamorphism time, the shungites have been contaminated with organic matter brought by meteoric waters. This organic matter had a relatively high δ¹⁵N of approximately +10‰ and the lowest C/N < 50. The different oxidation temperatures of the three organic components explain the reason for the variations in δ¹⁵N and C/N during step combustion of the material.

The paper presents pioneering data on the isotopic composition and elemental ratios of nitrogen, carbon (carbon dioxide), helium, and argon in the fluid phase of quenched tholeiitic glasses from different segments of the Bouvet Triple Junction area (BTJ). The data reflect a complicated geodynamic and tectonic history of the area evolution and indicate that the variations in the elemental ratios of the volatile components of the fluid–gas phase were controlled by a number of various factors: elemental fractionation during melt degassing, mixing of gases from different sources, postmagmatic diffusion-controlled helium loss. The nitrogen–argon and noble gas isotope systematics suggest a significant contribution of the atmospheric component to the mantle source of fluids for the samples from the Spiess Ridge and the segment of the Southwest Indian Ridge (SWIR) and a smaller contribution for the Mid-Atlantic Ridge (MAR) samples. For the Spiess Ridge and SWIR, the most probable contaminating agent was water fluid with dissolved gases of atmospheric composition. This fluid may have been brought to the mantle with ancient crustal rocks involved in magma generation. These crustal rocks may represent small fragments of the Gondwana continent with which sedimentary organic matter could be brought into the magma source.

We report the carbon isotope compositions of a set of diamond crystals recovered from an investigation of the experimental interaction of metal iron with Mg–Ca carbonate at high temperature and high pressure. Despite using single carbon source with δ¹³C equal to +0.2‰ VPDB, the diamond crystals show a range of δ¹³C values from –0.5 to –17.1‰ VPDB. Diamonds grown in the metal-rich part of the system are relatively constant in their carbon isotope compositions (from –0.5 to –6.2‰), whereas those diamonds recovered from the carbonate dominated part of the capsule show a much wider range of δ¹³C (from –0.5 to –17.1‰). The experimentally observed distribution of diamond’ δ¹³C using a single carbon source with carbon isotope ratio of marine carbonate is similar to that found in certain classes of natural diamonds. Our data indicate that the δ¹³C distribution in diamonds that resulted from a redox reaction of marine carbonate with reduced mantle material is hardly distinguishable from the δ¹³C distribution of mantle diamonds.