At the Goldschmidt conference in Davos 2009, chemist Anthony Chappaz told me about the potential of using synchrotron radiation to determine the speciation of amorphous (non-crystalline) materials. This approach could potentially provide fundamental insights to how molybdenum (Mo) is cycled in the environment. The technique was first applied in the geochemical literature by the famous Mo-geochemist, Prof. emeritus George Helz. Helz had shown that ancient black shales contain Mo coordinated to oxygen (O) and sulfur (S), similar to what he produced in the laboratory when Mo is precipiated with Fe in anoxic and sulfidic solution. However, nobody had ever tested if this occur in nature today.
Anthony had just seen my presentation on the cycling and isotope geochemistry of Mo in anoxic and sulfidic Lake Cadagno – a modern analogue to Proterozoic ocean chemistry. The Mo removal process is very efficient in anoxic and sulfidic environments, and largely control the Mo removal in the ocean today (below the seafloor) and was more important in the distant past, when atmospheric O2 levels were low. We decided to study the speciation of Mo in the sulfidic muds to test if Mo precipitation in this modern setting compared to what Helz had found. Also, we suspected that its chemical form would tell us more about the removal process from oxic waters thru sulfidic deep waters into the sulfidic sediments.
Remarkably, we were able to determine the Mo oxidation state (its atomic charge) and coordination environment (number of ligands and type of neighboring atoms) in the muds from Lake Cadagno, in samples with only >88 ppm. Further, we performed oxidation experiments to confirm the presence of reduced compounds in the samples, and provide improved means to infer the oxidation state from the XANES part of the x-ray absorption spectrum. Our results show that the muds contain predominantly a Mo compound coordinated exclusively to (four) sulfur atoms – distinct from molybdenite and other isolated reference materials. The Mo speciation is most similar to the famous Mo-Fe-S cubane in the N2-splitting enzyme, nitrogenase. However, there was no evidence that our samples actually contained nitrogenase, since nitrogen-fixing organisms have too little Mo relative to organic carbon to explain the high Mo content in our samples. Instead, we concluded that the chemical compounds are similar.
Upon oxidation, the Mo-S bonds break and form Mo=O double bonds similar to what Helz and colleagues observed in black shales. This suggests that Mo in shales experienced oxidation either in nature or during processing in the laboratory. The emerging view is that the Mo compounds can be synthesized in anoxic and sulfidic solution in the presence of iron. The Mo reduction step is found to be associated with zero-valent sulfur, S(0), donating electrons for reaction. This leads us to propose that Mo removal is fast at the chemocline and slower at depth for two reasons: 1) S(0) is available in higher concentrations at the chemocline, producing more reactive Mo-Fe-S species, 2) Mo depletion with depth below the chemocline cause Mo removal to decline (given that some of the reactive Mo-Fe-S from the chemocline is transported horizontally and not all precipitating vertically). Together this features illuminate the Mo removal process in marine anoxic environments and improve our ability to use Mo enrichments and Mo isotope compositions to infer water column redox conditions in the past.
Dahl et al. GCA (2013): Molybdenum reduction in euxinic lake