Quartz as Monitor for Sediment Provenance

Quartz belongs to the most abundant minerals in the Earths crust and is an important constituent of sediments and sedimentary rocks. In recent years it has become evident that the incorporation of trace amounts of protons (so called OH defects) reflect initial crystallization conditions and/or later thermal overprint. OH defects and trace metals in quartz allow discrimination between grains from different source rock systems, such as those from young igneous bodies versus those from bodies with metamorphic overprint. Mineral and melt inclusions reflect the crystallization history of quartz, too, but are more inert towards thermal treatment than OH defects. Knowing both characteristics (OH defects and melt/mineral inclusions) allows for the reconstruction of initial crystallization condition and later metamorphic overprint of individual quartz grains. In this way, individual quartz grains bear a wealth of information concerning their geological history and have a high potential as a tracer that may be used to determine sediment provenance and infer geological processes within the source area, even if the respective parent rocks have long been eroded away. The quartz-monitor research project investigates the intrinsic chemical variations of detrital monocrystalline quartz and inclusions (mineral and melt) as an innovative sediment provenance tool. The aim of this project is to generate the first comprehensive database of detailed OH defect characteristics of detrital quartz from selected sedimentary systems with properties that make them model systems for other localities worldwide, and to test applicability and limits of hydrous defects and trace metal geochemistry in quartz as a monitor for sedimentary processes and provenance analysis in large-scale source to sink settings. The sample material will be derived from existing offshore wells drilled by international scientific ocean drilling, ongoing IODP expeditions and on- shore reference material. As main analytical methods Fourier-Transform Infrared (FTIR) Spectroscopy, Secondary Ion Mass Spectrometry (SIMS), and Electron Probe Micro Analyzer (EPMA) will be used. Results of this project will help to further understand the evolution of different contributors in large sedimentary systems in terms of provenance and mixing ratios of quartz from different lithologies (metamorphic, igneous, and reworked sedimentary material), as quartz populations better reflect mixing ratios of contributing sources than accessories such as heavy minerals, or zircon age spectra.

OH-defect contents in quartz grains were determined from 6 sediment samples, one from the lower Amazon, 2 Andean tributaries (Solimões, Madeira), and 3 Shield tributaries (Tapajs, Xingu, Rio Negro). IR spectra and thus the total OH-defect inventory is dominated by AlOH and similar in all samples. The average defect water content in quartz grains is higher in samples with Andean origin than the average defect water content with Shield provenance. Low average defect water contents are mostly caused by a larger fraction of defect-water poor grains. Taken together, the lower Amazon sample is more like the grains with Andean origin, but may be a mixture with Shield component. However, Andean signal and Shield signal show a strong overlap making an assignment to one source impossible. This result is rather unexpected, because in a similar case study in Europe (old Scandinavian basement versus relatively "young" Variscan granites from Middle Europe) the defect water content differed by an order of magnitude. Surprisingly, the most remarkable difference between the 6 sediments was the occurrence of fluid inclusion (hitherto not considered as good marker): in quartz grains from Solimões this signal was most much stronger than in the Shield samples. Grains from the lower Amazon grains were in between, suggesting (again) that the quartz grains from the lower Amazon represent a mixture of Andean and Shield components. The molecular water signal showed subtle (though significant) variations, similar to synthetically produced solutions with different salt contents, suggesting that a discrimination of grains from different sources by this parameter would be in general possible. However, application for provenance analysis did not give conclusive results. Trace metal contents determined by electron microprobe showed - if detectable at all - an unspecific spread for all measurable elements in all subsets of grains. Trace element contents analysed by secondary ion mass spectrometry (SIMS) show a fair correlation between Li and Al in all subgroups; weak correlations were observed for the pairs Al - Ti and Li - Ti, and in some subgroups also for Li - B and B - Al, but the differences between the different subgroups are too small to be useful for provenance analysis. Absolute trace elements contents and trace element ratios show a strong overlap for all subsets of grains, making it impossible unequivocally distinguish grains from one source from another. With a critical view to all results, it can be summarized that (1) the different sources from the lower Amazon are quite similar in terms of the investigated characteristics of their quartz grains, (2) a mixture with substantial contribution of both Andean and Shield components is very probable, and (3) the molecular water signal could be a valuable tool for discrimination.