The geochemistry of redox variable elements

The project aims to use synchrotron light to determine the oxidation state (or charge) on metals in magmas at high temperature. The work is motivated by a desire to understand igneous processes through correctly interpreting the geochemical origins of trace element signatures. Igneous petrology is a mature subject which has recently been significantly advanced by the advent of technologies (e.g. laser-ablation ICP-MS) enabling the accurate determination of trace element concentrations. However, the interpretation of the signatures has tended to lag behind the acquisition of the data, such that empirical correlations between elements are attributed to processes (e.g. garnet in the source, crystallisation of plagioclase, island-arcs) with minimal understanding of the fundamental chemical controls responsible for the behaviour. The next quantum leap will be to understand the chemical origin of these signatures. To do this, we need to determine the oxidation state of trace elements over a range of geological variables. The oxidation state defines the charge and size of a trace element and hence its suitability for occupying a lattice site in mineral phases. This suitability controls the crystal-melt partitioning of the element during partial melting and fractional crystallisation, leading to trace element signatures. The oxidation state ratio is influenced by, in particular, the oxygen fugacity, but also the temperature and pressure of the melt, and is thus a powerful potential indicator of magmatic processes. However, the determination of oxidation states is not easy. There has been huge interest in, and effort devoted to, determining the oxidation state ratio of Fe (the most abundant redox variable element). Other elements have been almost impossible to study due to the interfering effect of Fe in most analytical (both wet chemical and spectroscopic) techniques. This is emphasised by statements in the literature such as 'There is no known method to measure the oxidation state of Cr in melts containing Fe'. A relatively new method for determining oxidation states is X-ray Absorption Near Edge Structure (XANES) spectroscopy, undertaken at a synchrotron light source. XANES spectroscopy is an element specific technique, with the capability of sub-micron spatial resolution, and which is suitable to in situ studies of silicate melts. This is important since the redox states that exist in melts at high temperature, and which control the partitioning and geochemical behaviour of an element, are not necessarily retained on cooling due to charge transfer reactions with the large redox variable reservoir of Fe. For example, approximately half the Cr in a mid-ocean ridge basalt (MORB) at 1400 C is Cr(II) even though this oxidation state has never been identified in a terrestrial material (Cr(II) in the presence of Fe(III) is 'unquenchable'). For this research a furnace has been designed that allows XANES spectra to be recorded for almost any element in a melt at temperatures up to 1500 C as a function of the oxygen fugacity. The furnace allows melts to be studied in situ. This work will allow the oxidation state ratio of almost any element in a melt of natural composition compositions to be determined for the first time.