Unmixing of Phosphorus-Bearing Melts on Earth and Mars

• Applicant: Professorin Dr. Sharon Webb
• Subject Area: Mineralogy, Petrology and Geochemistry
• Term: from 2015 to 2019
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 278990208

The structural reasons for the anomalous effect of phosphorus on silicate melt viscosity will be investigated. The addition of phosphorus has been found to increase the viscosity of peralkaline composition melts. Further addition of P2O5 results in a decrease in melt viscosity. Despite indications in previous studies that these phosphorus-bearing melts are unmixed, neither Raman spectroscopy, nor NMR spectroscopy has proven this. Preliminary calorimetric measurements of the glass transition temperature of some of the present P2O5-bearing peralkaline Na2O-Al2O3-SiO2 melts has shown there to be 2 glass transition temperatures, indicating that there are 2 unmixed glasses present in the sample, despite the glass being colourless and transparent. Further investigation of the macroscopic structure via scanning electron microscopy reveals that indeed the glass sample has unmixed. The increase in viscosity observed at both high (literature data) and low (present data) temperature indicates that this melt is unmixed up to 1446 °C. In view of these observations, the possible unmixing of P2O5-bearing melts will be investigated here using calorimetry, viscosity, Raman, and SEM techniques. The array of compositions investigated range from the simple Na2O-Al2O3-SiO2 and CaO-Al2O3-SiO2 to more complex Martian compositions (the Champagne rock in the Gusev Crater). The unmixing of these melts has implications not only for viscosity but also for crystallisation processes in co-existing co-mixed magmas of different composition. Phosphorus is in general classed as a fluxing agent in melts. Other fluxing agents are B, F and H2O. Boron is also known to cause melts to unmix. Further, as the addition of boron to the melt with both CaO and Na2O will result in an increase in peralkalinity with the reaction BO4=BO3+NBO shifting to the right with increasing CaO/Na2O, the presence of B2O3 in the melt is expected to affect the unmixing of the P2O5-bearing melts. Once the question of the unmixing of P2O5-bearing melts has been addressed, the project will extend to the unmixing of P2O5-bearing melts with added B2O3, F and H2O.

• DFG Programme: Research Grants