Welcome to the new version of CaltechAUTHORS. Login is currently restricted to library staff. If you notice any issues, please email coda@library.caltech.edu
Published 1996 | public
Book Section - Chapter

Rare earth elements in carbonate-rich melts from mantle to crust

Abstract

The rare earth elements (REE) are widely distributed as trace elements in mantle rocks and mantle-derived magmas, and in crustal rocks. There is a huge literature on their geochemistry, but much less has been written about how they become concentrated from trace concentrations in the mantle to high percentages in ore deposits. Carbonatite is one host rock for REE deposits, and this review is concerned with the behaviour of REE in carbonate-rich melts in environments ranging from the Earth's mantle to shallow level magmatic carbonatite intrusions and hydro-carbo-thermal veins. A high percentage of the trace REE elements in the upper mantle may be stored in a small percentage of discrete titanate minerals (Haggerty, 1983; Jones, 1989), and most would dissolve in carbonate-rich melts gene rated in the mantle at depths greater than about 70 km; abundances would remain at the trace element levels, however. The immiscible separation of carbonate-rich magma from silicate magma is expected to cause enrichment in REE (Wendlandt and Harrison, 1979), but the experiments of Hamilton, Bedson and Esson (1989) did not reveal the kind of enrichment required to explain the REE in many carbonatites. In the Mountain Pass carbonatite, bastnasite constitutes up to 15 vol.% of the ore body (Olson et al., 1954; Heinrich, 1966, p. 357). Furthermore, experiments on silicate-carbonate liquid immiscibility indicate that although immiscibility is likely to occur within the crust, it appears to be improbable in the lower lithosphere. The high concentrations of REE are probably caused by fractional crystallization of a carbonatite magma already somewhat enriched in REE, and we have conducted experiments to test this proposal. We are approaching the problem in two ways. The first is to build from simple to more complex phase diagrams, in order to establish precisely the behaviour of the rare earth elements in carbonate-rich melts, as a guide for interpretation of the more complex systems and of the rocks themselves. We present here the results for the simplest model of a rare earth-carbonatite magma, the composition join CaCO_3-Ca(OH)_2-La(OH)_3 through the system CaO-La_2O_3-CO_2-H_2O (Jones and Wyllie, 1986), followed by joins to synthetic hydroxylbastnäsite (Deng and Wyllie, in preparation). These experiments outline conditions under which bastnasite and calcite can be coprecipitated from melts at moderate temperatures; this provides a simple analogue for the primary formation of RE-carbonatite from a magma. The second approach is to melt complex mixtures approximating the composition of the ore body at Mountain Pass, to follow paths of crystallization and to locate the parageneses and conditions for the precipitation of bastnasite. The system studied includes baryte and fluorite. Preliminary results have been published (Jones and Wyllie, 1983; Wyllie and Jones, 1985). These experiments show that with fractional crystallization, residual melts will contain high concentrations of REE (18 wt% La(OH)_3). This experimental review outlines likely conditions for the formation of primary RE minerals from carbonatite magmas. Carbonatites are amongst the lowest temperature terrestrial magmas known. If crystallized at shallow depths in the upper crust, primary bastnasite group minerals could form at temperatures as low as ~ 550°C. Subsolidus changes, which may occur frequently in natural systems, are not considered here. However, some examples of possible carbothermal or hydrothermal replacement reactions are given by Gieré in Chapter 5.

Additional Information

© 1996 Chapman & Hall. This research was supported by the Earth Science section of the US National Science Foundation, grant EAR 9218806. Numerous colleagues in the 'carbonatite community' are thanked for their contributions to an enjoyable series of discussions over the past few years. APJ would like to thank, in particular, John Gittins, Mike LeBas and Alan Woolley for their help with earlier versions of this work.

Additional details

Created:
August 22, 2023
Modified:
January 13, 2024