Carbonatites and associated rocks are the main source of rare earth elements (REEs), metals essential to modern technologies. REE mineralization occurs in hydrothermal assemblages within or near carbonatites, suggesting aqueous transport of REE. We conducted experiments from 1200°C and 1.5 GPa to 200°C and 0.2 GPa using light (La) and heavy (Dy) REE, crystallizing fluorapatite intergrown with calcite through dolomite to ankerite. All experiments contained solutions with anions previously thought to mobilize REE (chloride, fluoride, and carbonate), but REEs were extensively soluble only when alkalis were present. Dysprosium was more soluble than lanthanum when alkali complexed. Addition of silica either traps REE in early crystallizing apatite or negates solubility increases by immobilizing alkalis in silicates. Anionic species such as halogens and carbonates are not sufficient for REE mobility. Additional complexing with alkalis is required for substantial REE transport in and around carbonatites as a precursor for economic grade-mineralization.

Nolans-type ore deposits contain REE and Th mineralization hosted in fluorapatite veins. These veins intrude granulite facies rocks and are surrounded by a diopside selvage. Nolans-type deposits are thought to form by REE, F and P-rich hydrothermal fluids derived from alkali or carbonatitic intrusions. However, REE are not effectively transported in F and P-rich systems. REE ore deposits are commonly hydrothermally overprinted, possibly obscuring the igneous nature of the primary mineralization. We conducted a series of piston cylinder "sandwich" experiments, testing the hydrothermal fluid hypothesis, and a newly suggested process of carbonatite metasomatism. Our results confirm theoretical predictions that REE are hydrothermally immobile in these systems and the experimental phase assemblage is not compatible with the natural rocks. Our results show that fluorapatite can only host several weight percent levels of REE at temperatures higher than ϳ600°C. Below that temperature, a miscibility gap exists between REE-poor fluorapatite and REE-rich silicates such as britholite or cerite. In contrast, experiments reacting P and REE-rich carbonatite with silicate rock above 700°C closely resemble natural rocks from Nolans-type deposits. Selvage mineralogy is sensitive to the MgO content of the carbonatite. A diopside selvage formed at carbonatite MgO/(CaO؉MgO) Ϸ 0.2 while wollastonite and forsterite formed at lower and higher ratios, respectively. Phosphate solubility in carbonatites decreases with decreasing MgO contents. As diopside formed, REE-rich fluorapatite preferentially crystallized from the selvage inwards. Thus, carbonatites are effective at simultaneously mobilizing REE, F and P to the site of deposition. Nolans-type deposits are the cumulate residue of this reaction, with the carbonatite liquid migrating elsewhere. At temperatures below 700°C the carbonatitesilicate reaction additionally formed monticellite, cuspidine and magnesioferrite, resembling a skarn assemblage. Whereas skarns form by infiltration of silicate magmas or related fluids to carbonate rocks, our experiments are the opposite: intrusion of carbonatite into silicate rock. These mid-crustal skarn-like rocks may host elevated ore elements of carbonatitic affinity, such as F, P, Y, REE, Th, Ba, Sr, and Nb. We propose the term "antiskarn" to describe such systems, and suggest they trace the migration of carbonatite liquids through the crust. Hydrothermal reworking, retrogression, or metamorphism of antiskarns may obscure the carbonatitic genesis of the rocks. These metasomatic zones are the crustal equivalent of wehrlites that form by peridotitecarbonatite reaction at mantle depths.

causing crystallisation of allanite-(Y) from hydrothermal fluids in the miarolites. The REE pattern of the hydrothermal allanite-(Y) is characterised by LREE and Eu depletion, similar to the parent plagiogranitic magma. As allanite had sequestered most of the REE in the fluid, epidote took over as the principle hydrothermal mineral. Epidote in Troodos plagiogranites records a fluid evolutionary trend beginning with REE-rich-Eu-depleted similar to allanite-(Y) and gradually transforming into the REE-depletedEu-enriched pattern prevalent throughout 'conventional' sub-seafloor fluids. A comparison of allanite-bearing and allanite-absent plagiogranites from the same locality suggests that REE-bearing fluids migrated from the plagiogranites. Similar fluid evolution trends observed in diabase-hosted epidote, located adjacent to a large plagiogranite body, suggest influx of plagiogranite-derived REEbearing fluids. Epidotisation in oceanic settings is usually considered to be the result of alteration by high fluxes of seawater-derived hydrothermal fluids. Although epidotisation by magmatic fluids has been suggested to occur in plagiogranites, our study shows that this autometasomatic process is the dominant mechanism by which epidosites form in plagiogranites. Furthermore, epidotisation of diabase has been attributed solely to seawater-derived fluids, but we show that it is possible for diabase-hosted epidosites to form by migration of plagiogranite-derived fluids.