Kwak and Abeysinghe reported in 1987 that at least six separate rare earth and uranium-bearing daughter crystals occur in fluid inclusions hosted by andraditic garnet from the Mary Kathleen REE-U ore skarn, Queensland, Australia. The daughter minerals are particularly high in La, Nd and Ce which reflects the relatively high concentration of these in the bulk ore (La2O3 = 33.5%, Nd2O3 = 9.1% and Ce2O2 = 51.5% of the 2.6 wt. % REE common in the ore). The host garnets themselves contain up to 7600 ppm REE and 5 to 2700 ppm U. The energy-dispersive spectra (EDS) are consistent with the following minerals: a (Y, Ce, U, Ca, Fe, Nb, Ta) mineral; a (Ca, Fe, Ce) carbonate(?) mineral; a (Fe, Ca, Y, Ce, Nb, Ta) mineral; a possible carbonate of La, Mn and Nd; a chlorite of Mn and La as well as a possible chloride or oxychloride of K, Mg, Mn and La.
Their occurrence infers that relatively high concentrations of REE and U prevailed in the original, oxidized, concentrated (30-70 wt. % total dissolved salts), high-temperature (550-670 ~ ore solutions. Their presence as daughter crystals may be due to the fact that CaC12 is a dominant salt in the solutions and that the latter's solubility was sufficiently high to 'salt out' the less soluble REE-U components.
These conclusions are compatible with the view that these solutions represent largely reacted or 'exhaust' ones, produced as a result of ore solution interaction with carbonate to produce skarn. This is apparent from the extremely high CaC12 contents present which can be shown to be produced during skarn genesis (see Kwak and Tan, 1981). The REE and U-bearing mineral species allanite, uraninite and stillwellite which occur in the skarn obviously postdate the crystallization of the garnet which hosts the fluid inclusions. The REE and U thus tend to remain in solution during early skarn genesis. It is notable that, like the REE:U ratio in the skarn (i.e. 2.6:0.1), REE bearing daughter crystals are much more abundant. It is likely that REE and to a much lesser extent U must be very soluble in saline, high-temperature solutions at the relatively high fO2 values at which andradite garnet (+pyroxene) form.
The probable reason REE daughter crystals have not previously been reported is that no solutions have been studied in which either REE concentrations were sufficiently high so as to be supersaturated on falling temperatures or there was no extremely soluble salt present such as CaC12 to 'salt out' such components. As no mineral having appreciable REE and U crystallized at this period of the skarn's evolution, the elements stayed in solution rather than precipitating as REE-allanite, uraninite and stillwellite as they did at a later time.The occurrence of the unusual daughter minerals in fluid inclusions in the Mary Kathleen skarn may thus reflect the unique environment of this type of mineralization.
Also see: http://www.i2mconsultants.com/downloads/corp/JGG-2-023.pdf