Viscosity and glass transition temperature of hydrous melts in the system CaAl2Si2O8–CaMgSi2O6

The Newtonian shear viscosity and glass transition temperature (Tg) of hydrous melts in the system Anorthite (An)–Diopside (Di)–H2O have been experimentally obtained. Viscosity data on hydrous samples with up to ca. 3 wt.% H2O have been measured using a micropenetration technique in the interval between 108.3 and 1013.1 Pa s and temperatures up to 880 °C at ambient pressure. Measurements of the calorimetric Tg values were performed by using differential scanning calorimetry (DSC). For each sample the cooling rate dependence of Tg was characterised at three different temperatures corresponding to the onset, the peak and the stable liquid regions of the heat capacity curves. These results show strong correlations between these temperatures that can probably be extrapolated to other unequivocally defined metrics of the glass transition interval. Comparison with viscosity data obtained on the same samples shows that glass transition temperatures at each single heating/cooling rate reflect constant viscosity values for these hydrous liquids. Thus observed relationship between calorimetric Tg and viscosity is independent of composition and water content (c.f., [Giordano, D., Nichols, A.R., Dingwell, D.B. (2005) Glass transition temperatures of natural hydrous melts: a relationship with shear viscosity and implications for the welding process. J. Volcanol. Geotherm. Res. 142, 105–118.]). Measured and calculated viscosities and glass transition temperatures for melts in the An–Di–H2O system show substantial differences with those of basaltic composition, suggesting that, despite what commonly assumed, An–Di is not a good rheological proxy for basaltic compositions. The observed differences are reduced at high temperature in the low viscosity range and are significantly more pronounced at low temperature. We infer that such an effect is due to the different contributions to the configurational entropy provided by the simple melts in the An–Di–H2O system compared to the multicomponent basaltic melt investigated. Some implications about the role of water in influencing melt properties are discussed. The results provided here demonstrated that, in some instances, extrapolating the physical properties of simple systems to those of natural multicomponent melts is not appropriate and may result in erroneous evaluation of petrological and volcanological scenarios which require knowledge of those properties.