Giant lithium pegmatites may begin with “sanukitoid-infused” greenstone roots

01

Why this matters

Lithium (Li) pegmatites supply a large proportion of the world’s hard-rock lithium, and the Archaean cratons of Western Australia contain some of the most productive examples. Yet the source of the lithium remains debated. The dominant model for many LCT pegmatites invokes sediment-rich crust: weathering concentrates lithium and other rare metals in sediments, those sediments are buried and melted, then a pegmatitic melt is formed through low-degree anatexis, extensive fractional crystallisation, or a combination of both. That model is plausible in terranes with abundant fertile metasedimentary rocks, but it is difficult to apply to the Yilgarn and Pilbara cratons, where such sedimentary source rocks are not regionally voluminous and strongly peraluminous S-type granite suites are not a major feature of the Archean record.

Smithies et al. address this mismatch by asking whether Li-rich pegmatite provinces can instead be generated from predominantly metaigneous crust that was chemically modified before melting. Their proposal is important because it moves the exploration question back in time: rather than searching only for evolved granites in orogenic settings, explorers should also identify the crustal architecture, mantle-derived magmatism and alteration processes that prepared may prepare a fertile source.

02

What the study found

The authors combine a large regional dataset of 4,531 whole-rock samples with zircon U-Pb ages, whole-rock Sm-Nd isotope data and geological mapping. The Yilgarn dataset is the focus, with the Pilbara used as an independent comparison. They define anomalous Yilgarn leucogranites as those containing more than 70 ppm Li, approximately the 95th percentile of the dataset, or anomalous Cs where Li is unavailable.

Several observations form the basis of the model. First, the Yilgarn and Pilbara leucogranites are generally metaluminous to weakly peraluminous, with ASI values commonly between 1.0 and 1.1 and trends that resemble TTG-derived rather than sediment-derived melts. Second, Li- and Cs-anomalous leucogranites are concentrated near greenstone–granite contacts. In the Yilgarn, approximately 60% of anomalous samples occur within 5 km of a known contact, increasing to about 65% in the eastern craton. Third, these leucogranites have relatively radiogenic εNd values, broadly between -2 and +2 at 2.65 Ga, overlapping the juvenile end of the wider felsic-rock array. Similar spatial and isotopic relationships occur in the Pilbara, especially in the juvenile western corridor containing Wodgina and Pilgangoora.

The proposed precursor is sanukitoid magmatism. Sanukitoids are hydrous, mantle-derived high-Mg diorite to monzodiorite magmas that are strongly associated with greenstone belts and major shear zones. In the Yilgarn, approximately 90% of exposed sanukitoid samples lie along shear zones within or at the margins of greenstone belts, and 97% occur within 5 km of greenstone–granite contacts. The authors suggest that these magmas and their fluids introduced water, Li and other incompatible elements into the mid-crust beneath greenstone belts. Subsequent biotite-forming alteration stored part of this budget in the crust. Later biotite-dehydration melting then remelted the modified source to produce Li-rich leucogranites, which could ultimately evolve into LCT pegmatites.

03

Our interpretation

The strongest contribution of this paper is not the claim that sanukitoids themselves were necessarily highly Li-rich. The regional sanukitoid data show only modest and poorly defined Li enrichment, with primitive samples reaching approximately 21 ppm Li and some least-altered samples up to about 50 ppm. Instead, the key process is repeated alteration and storage. Hydrous sanukitoid intrusions could have generated biotite-rich reaction zones in the surrounding crust, and biotite is an efficient carrier of Li, Cs, Rb and volatiles. The relevant source is therefore a refertilised crustal volume, not simply a primitive magma with an unusually high initial Li concentration.

This distinction matters for pegmatite genesis. The paper does not replace fractionation with source enrichment; it relocates the first stage of enrichment. Once a leucogranitic melt inherits a larger lithium budget, crystal fractionation, unmixing, liquid extraction and local structural focusing can still be required to produce a mineralised pegmatite. Source fertility establishes the available inventory, but it does not by itself determine whether a particular intrusion becomes economic.

The evidence is persuasive at the regional scale but remains indirect at the source-to-pegmatite scale. The spatial association between greenstone contacts, sanukitoids, radiogenic Nd and anomalous Li in leucogranites is systematic. However, the proposed source is not directly sampled, and the paper does not present a full mass balance demonstrating that sanukitoid alteration can generate the Li, Cs, Rb and volatile budgets required by the associated giant pegmatites. The link from anomalous leucogranite to pegmatite is supported by temporal and isotopic association, but the intermediate evolution from leucogranite to pegmatitic melt is deliberately not investigated in detail. This makes the model an excellent regional fertility framework.