LITHIUM PEGMATITES IN THE HIMALAYAN CREST 2: A NEW FRACTIONATION MODEL FOR WATER- AND METAL-RICH MELTS

Lithium-rich pegmatites have recently been discovered in the South Tibetan Detachment System (STDS) at the crest of the Himalaya (see companion presentation of Liu et al.). However, past models to explain how these pegmatites form are either qualitative or require large numbers of cyclic fractionation-extraction events that have little direct field evidence. Here, we propose that Li-enriched melts may form and separate via an unusual combination of rapid crystal growth, limited diffusion of Li and water, and deformation in the context of the STDS.

Local melt compositions adjacent to rapidly-growing crystals (boundary layer compositions) depend on what is crystallizing. Crystallization of anhydrous quartz, feldspar, monazite and zircon can theoretically form Li- and H₂O-rich, inviscid boundary layers, whereas crystallization of micas can form Li- and H₂O-poor, viscous boundary layers. However, rapid crystallization is required for boundary layers to form at scales less than typical grain sizes of a few mm: about 1×10^-7 to 1×10^-8 m/s (c. 1 mm/day) at likely magma temperatures of 700 °C. Even at 700 °C, water-saturated boundary layer melts are still too viscous to separate via density. Consequently, formation of transient pressure gradients during deformation must preferentially remobilize H₂O- and Li-rich melts to form Li-rich pegmatites. Our model for Li-pegmatite formation therefore involves three steps: 1. Intrusion of parent magmas into relatively cold country rock (rapid crystallization), 2. Magma lockup and formation of Li- and H₂O-rich boundary layers adjacent to anhydrous minerals, and 3. Deformation to mobilize the least viscous layers. This model explains field observations and distinct compositional gaps between Li-rich pegmatites and potential coeval source magmas from the STDS in southern China.

If our model is correct, the c. 2200 km-long STDS may represent an important zone of Li enrichment and site for future Li reserves. More generally, the rock record of fractional crystallization may not reflect whole system fractionation. Preferential extraction of hydrous boundary layer liquids that develop adjacent to anhydrous minerals implies some evolved felsic magmas, such as Li-pegmatites, may not record crystallization of hydrous minerals.