Another continental pool in the terrestrial silicon cycle

Abstract

Silicon is the second most abundant element on Earth. It is an important nutrient for phytoplankton and is readily absorbed by terrestrial vegetation; it also assists the removal of carbon dioxide from the atmosphere through the weathering of silicates. But the continental cycle of silicon is not well known, and only a few studies have attempted to use silicon stable isotopes (28Si, 29Si and 30Si) to quantify the continental silicon reservoirs. Dissolved silicon in sea and river waters forms a reservoir of mean isotopic value +1.1 per thousand (refs 7, 10). It is enriched in 30Si with respect to the igneous rocks reservoir, which has a mean isotopic value of -0.3 per thousand (refs 4, 9). This enrichment can only be produced by a major fractionation during weathering, and should result in the formation of a continental 30Si-depleted reservoir. Such a reservoir, however, has not been identified to date. Here we analyse silicon isotopes of in situ quartz from a sandstone series in France, using a new-generation secondary ion mass spectrometry apparatus. We show that quartz that precipitates as siliceous cements forms a strongly 30Si-depleted reservoir with isotopic values down to -5.7 per thousand, a more negative value than any previously published for terrestrial samples. Our findings suggest that quartz re-precipitation plays an important role in the biogeochemical cycle of silicon.

Si isotope variability in Proterozoic cherts

“…MC-ICPMS Si isotope data are also accurate to better than 0.1& based on measurements of inter-laboratory standards.MC-SIMS allows sampling in high resolution but is limited to a 2r uncertainty of 0.3& for d 30 Si (Heck et al, 2011) which, although a significant improvement compared to earlier ion-probe studies (e.g. Basile-Doelsch et al, 2005;Robert and Chaussidon, 2006) is still 3 times less precise than MC-ICPMS. Several recent studies have reported high-precision MC-ICPMS Si isotope measurements of cherts, primarily from Archean successions (Andre et al, 2006;van den Boorn et al, 2007van den Boorn et al, , 2010Steinhoefel et al, 2009Steinhoefel et al, , 2010Abraham et al, 2011).…”

“…However, the Si isotope fractionation factor between amorphous silica precipitates and water remains to be determined accurately, and may vary as a function of the precipitation mechanism as well as setting. For example, in contrast to terrestrial pedogenic and groundwater silcretes, where later-formed macro-crystalline quartz is depleted in 30 Si compared to the early-formed micro-crystalline quartz or the original detrital quartz, no isotopic fractionation has been observed between detrital quartz and microcrystalline quartz during marine silicification (Basile-Doelsch et al, 2005). Overall, while it may be expected that silica polymorph transformations, which involve dissolution and reprecipitation, might be accompanied by Si isotopic fractionation, there is no strong evidence for this actually compromising isotopic measurements.…”

“…The Si isotope fractionation factor between silica saturated fluids and silica precipitating from them is not well constrained although it is known that chemical precipitates preferentially incorporate the lighter Si isotopes (e.g. Basile-Doelsch et al, 2005). In this model, the fractionation between the fluid (silica saturated porewater) and the precipitate (cherts), which increases with decreasing temperature, is such that in Case 1 the precipitate is 2.0& lower than the fluid and in Case 2 it is 3.0& lower.…”

Stable isotopes from diatom silica

“…Alexander et al, 2002;Basile-Doelsch et al, 2005;Robert and Chaussidon, 2006), the ability to analyse small grains (> 50 µm) opens the possibility that it may become viable to analyse individual frustules. However, in order to achieve this, a reduction in the analytical error associated with SIMS is required, which are currently higher than those of other techniques (Basile-Doelsch et al, 2005).…”

“…Methods for the analysis of δ 30 Si include fluorine based conversion of purified silica into SiF 4 and IRMS (De Freitas et al, 1991;De La Rocha et al, 1996;Ding et al, 1996;Leng and Sloane, 2008), Multicollector Inductively Coupled Plasma Mass Spectrometry (MC-ICP-MS) (De La Rocha 2002;Cardinal et al, 2003;Reynolds et al, 2006b), acid decomposition and IRMS (Brzezinksi et al, 2006) or ion microprobe analysis (e.g. Basile-Doelsch et al (2005)). Analytical reproducibility for the most commonly used methods, fluorination and MC-ICP-MS are similar at ca.…”