Development of several different microbeam techniques allows in situ analysis of trace elements in natural and synthetic mineral assemblages, which in turn enables determination of two-mineral partitioning behaviour for trace elements (D-values). With the experimental approach, control of compositional and physical variables is possible, but equilibrium needs to be carefully assessed, preferably by full reversal of the trace-element exchange, or at least by a series of experiments over different times to demonstrate experimental duration needed for "consistency" of results. Also in some instances compositions and optimal experimental conditions are different from the natural situation and extrapolation to the "real" geological state should be justified. With the natural mineral approach, physical conditions must be independently assessed, accepting the uncertainties inevitably involved in geothermobarometric determinations. Also a potentially complex history of formation must be unravelled, with the attendant possibility of non-equilibrium. For example, where the formation of amphibole in initially anhydrous lherzolite is attributed to metasomatism by a trace-element and fluid-rich agent, it may be uncertain whether the trace element content of the pre-existing clinopyroxene in the anhydrous lherzolite equilibrates with the trace element content of the newly formed amphibole in the now hydrated and metasomatized lherzolite. Such a situation could lead to unrealistically high amphibole/clinopyroxene partition coefficients for the trace elements transported in the fluid-rich metasomatizing agent. Also cases exist where isotopic disequilibrium has been demonstrated between coexisting natural mineral pairs in xenoliths, and this points to the unlikelihood of achieving equilibrium trace element distribution between the same minerals.
Comparison of D-values determined for coexisting amphibole and Ca-clinopyroxene indicates overall
good agreement for Sr, Zr, Hf, Y and REE but significant discrepancy for Rb, Ba, Nb and Ta (experimental values for amphibole/clinopyroxene are much lower than natural values). For co-existing Ca-clinopyroxene and garnet the spread of data is greater and fewer elements can be compared (Rb, Ba, Nb and Ta data are not adequate). However Sr, Y, Zr, Hf and Nd-Lu (of the REE) agree reasonably well, but La and Ce experimental values for clinopyroxene/garnet are much lower than the natural values. These differences for both mineral pairs may be attributed in part to critical compositional differences between synthetic and natural minerals, where compositional factors have a key role in controlling accommodation of trace elements in the mineral structure. For example, the TiO2 content of the synthetic pyroxenes is higher than in the natural pyroxenes, possibly favouring acceptance of Nb, Ta into the pyroxene lattice, and hence lower D-values for Nb, Ta for synthetic amphibole-clinopyroxene pairs, compared with the natural pairs analyzed so far. A second factor is the possibility of minute inclusions rich in trace elements trapped in the natural minerals. These may be sub-microscopic and unavoidable, even by microbeam analysis.
Clearly more work on both experimental and natural systems is needed to resolve the discrepancies, and to realise the full potential for using trace elements to assess petrological processes in the mantle, where knowledge of D-value variation as a function of composition, pressure and temperature is essential. Much of the wide scatter of the natural data may be caused by complex compositional variation in the clinopyroxene, garnet and amphibole solid solution series and published studies have started to attempt to correlate variation in D-values with selected compositional parameters. These attempts illustrate the difficulty in separating different compositional factors when relying entirely on natural data. Experimental work has the unique capability of isolating the different compositional controls, ultimately leading to more precise parameterization of partitioning behaviour. This will be a worthwhile goal of such studies in the future. The natural mineral approach, dependent as it is on mineral compositions to infer a P-T history, cannot unequivocally resolve physical and compositional controls on trace element partitioning. Future work on the natural mineral pairs will benefit from concurrent TEM studies, to carefully assess the possible presence of minute fluid inclusions rich in trace elements.