Lithium (Li) and beryllium (Be) are critical metals essential for energy and advanced technologies. However, their geochemical behaviors during water-fluxed melting, a fundamental process during orogenesis, remain inadequately understood. This knowledge gap limits our ability to explain how Li and Be become enriched during orogenic evolution. In this study, we analyzed mineral and whole-rock geochemical compositions of structurally different domains in migmatites derived from water-fluxed melting. In-situ elemental analyses indicate that Be preferentially incorporates into plagioclase (andesine-oligoclase > amphibole > biotite ≈ muscovite > K-feldspar), while Li is dominantly hosted by biotite (biotite > muscovite > amphibole > plagioclase ≈ K-feldspar). Compiled whole-rock datasets suggest that the differential partitioning behaviors of Be and Li lead to the formation of distinct fertile reservoirs during water-fluxed melting, especially at low temperatures (within the muscovite stability field), with Be concentrating in proximal plagioclase-rich leucosomes (up to 6.37 ppm), while Li is retained in mafic residues (up to 254 ppm) alongside fluxing elements P and F. Phase equilibrium modeling confirms that subsequent (ultra)high-temperature, fluid-absent melting effectively remobilizes these elements to form Li- and/or Be-rich melts. Our findings suggest that water-fluxed melting plays a key role in pre-concentrating Li and Be. This points to multi-stage anatectic terranes, those that experienced early water-fluxed melting followed by later (ultra)high-temperature anatexis, as potential targets for Li Be pegmatite exploration.
Xiao et al. (Mon,) studied this question.