Viscosity is perhaps the most important physical property for modeling the behavior of an ascending magma. Although the effect of temperature and composition on the viscosity of silicate melts has been proven to be strong, the effect of pressure remains poorly constrained and is often assumed to be small. Experimental data on the pressure dependence of viscosity is particularly lacking in the high viscosity range. To address this problem, preliminary viscosity measurements have been carried out using a newly developed high-pressure viscosimeter.
The system consists of a dilatometer, which is easily inserted into
an internally heated pressure vessel (see Annual Report 1997). The device
works in the high viscosity range between 108-1011
Pa s, with a precision of ± 0.06 Pa s
log units. We conducted a series of viscosity measurements using the standard
glass DGG-1 (Tg = 543.3°C) at 500 bar (Fig. 3.8-6).
Fig. 3.8-6: Comparison of viscosity data obtained at 1 bar and 500 bar of DGG-1 standard glass of the German Glass Technical Society. Also plotted are the data from Dorfman et al. (1997).
We observed no obvious difference in measured viscosities between data obtained at elevated pressures and at 1 bar. Moreover, the results are in good agreement with the data obtained by Dorfman et al. (Eur. J. Mineral., 9, 345-350, 1997). This leads us to the preliminary conclusion that the pressure effect is weak in the high viscosity range for this composition. Further investigations at 1000 bar and 2000 bar will be conducted to better constrain better how pressure influences the viscosity of silicate melts at very high viscosities.