The first observation of the solubility of PGE in a silicated melt has been provided by the discovery of Ir crystals in yttrium-gadolinium garnet (YGG) monocrystals obtained from a bath melted in an iridium crucible, and the presence of these crystals was related to the oxygen fugacity in equilibrium. The proof of dissolution of a PGE in a "dry" silicate melt was given by this experiment Application of such observation regarding the behaviour of PGE in a basaltic melt appeared as a mean to understand the physicochemical mechanism of precipitation of PGE in the mafic-ultramafic suites.
The first attempts that we have performed with Ir (IPGE) and Pt (PPGE) (Amossé and Allibert, 1986, 1987, 1990, 1993) has confirmed the importance of oxygen fugacity whose variations influence the solubility of the PGE dissolved in basic silicate melts. The bath choosen was of basaltic composition at a temperature of 1430°C. The sulfur fugacity which looks more effective at the end of differentiation process, has also been studied in order to determine the conditions of precipitation of PGE in the base metal sulfides. In the present study we have extended the experimental study for the determination of behaviour of Rh and Ru. We have also used different values of oxygen fugacities which are not relevant to geochemical conditions (fO2 > 10-5 bar) for new experiments dealing with Pt and Ir.
The negative slope of the variation of solubility of Pt and Ir v.s. oxygen fugacity (e.g. Amossé and Allibert, 1986, 1987, 1990) has been confirmed for the fO2 values which are relevant for the geochemical conditions prevailing during the differentiation of the mafic-ultramafic suites. For higher fO2 values, a positive slope appears with the equilibrium between PtO, PtO2 on one hand and Pt2+,Pt4+ on the other hand. The same behaviour was also observed for Ir. For Rh, an equilibrium between RhO and Rh3+ with a caracteristic slope of 0.25 is observed at the fO2 which corresponds approximately to the precipitation of chromite. The corresponding oxydation reaction being: RhO + 1/4 O2 Þ Rh2O3. This interesting feature agrees with the observation of Capobianco et al. (1990) who observe that Rh (and Ru) can enter in the network of spinel. Rhodium spinel have been described a long time ago by Bertaut and Dulac (1961). Our study also confirms the presence of such a behaviour of Rh in a spinel in a natural system. A chromite from an ophiolite of the district of Tropoje (Albania) provides distribution of Rh, Ru and Os which seems related to the existence of a solid solution of the PGE in the chromite and exolution in subsolidus conditions at high temperature.
The behaviour of PGE in equilibrium with a sulfur fugacity confirms the previous data. Ir and Ru are not easily complexed by S, but Pt and Rh form complex compounds which are easily soluble in the basic silicate melt. This behaviour explains easily the precipitation of these metals in the base metal sulfides.
Such silicate melts in equilibrium with PGE metals vs. fO2 have been studied by some researchers but in conditions which cannot be compared with ours. As an example, Borisov and Palme (1995), O'Neill et al. (1995) have used baths with high CaO and having no iron content. From recent results obtained in our Laboratory (Dable et al., 1996, to be published) we have pointed out that a high CaO content can favors the domain of stability of ionic species (Ptn+) even at low fO2, and we have drawn a phase diagramm of Pt species on a large fO2 scale with equilibrium between Pt,PtO,PtO2 on one hand and Pt°, Pt2+, Pt4+, Pt6+ on the other hand. For high CaO content and Fe-free bath, ionic species can be seen even at low fO2. (fO2= 10-8 bar). Even though the iron is a major element of the ultrabasic suites, the diffusion of iron in PGE metals in equilibrium with the bath cannot be clearly pointed out previously by us. A surface layer containing iron can be expected. which seems to favor the passivation of PGE metals.for which determination of passivation conditions cannot be obtained from data concerning pure PtO or IrO2, but must include interaction of Pt and Fe in the surface layer. On the other hand numerous experiments published recently were obtained at fO2 which are far from these observed during differenciation of magmas, for instance in air. Capobianco et al. worked at fO2 between 10-6 and 10-4 but at a temperature of 1261°C. To compare these experiments with ours , obtained between 10-7 and 10-5 but at a temperature of 1430°C, it would be necessary to work at fO2 values two order of magnitude below ours (10-9 and 10-7).
The present study confirms the previous data obtained in our Laboratory and concerning the passivated state of PGE in equilibrium with iron-rich silicate melt. These data were extended to Rh and Ru. But for these elements the formation of ions Rh3+ and Ru4+ has been pointed out by a positive slope of the line giving solubility of the elements vs. fO2. at fO2 values relevant of geochemical conditions of chromite crystallisation. A solid solution of PGE in spinel was expected as reported by some authors. Such a feature has been found in an ophiolitic chromite from Albania. Ionic species have also been pointed out for Pt and Ir in our melt of basaltic composition with positive slopes as also reported by many authors, but for fO2 which are not relevant of geochemical conditions.These last results agree well with those published by many researchers.
Amossé, J. & Allibert, M., Colloque PIRSEM: Facteurs de concentration des matières premières minérales. Montpellier, 29-30 Septembre (1986).
Amossé, J., Allibert, M., Fischer, W. & Piboule, M., C.R.Acad.Sci. Paris 304, 1183-1185 (1987).
Amossé, J. & Allibert, M.,Chem. Geol. 81, 45-53 (1990).
Amossé, J. & Allibert, M., Geochim. Cosmochim. Acta 57, 2395-2398 (1993).
Bertaut, E.F. & Dulac, J., Phys. Chem. Solids 21, 118-119 (1961).
Botrisov, A. & Palme, H., Geochim. Cosmochim. Acta 59, 481-485 (1995).
Capobianco, C.J. & Drake, M.J., Geochim.. Cosmochim. Acta 54, 869-874 (1990).
Capobianco, C.J., Hervig, R.L. & Drake, M.J., Chem. Geol. 113, 23-43 (1994).
O'Neill, H. St C., Dingwell, D.B., Borisov, A., Spettel, B. & Palme, H., Chem. Geol. 120, 255-273 (1995).