HR: 08:45h
AN: V51B-04    [PDF]
TI: An Experimental View on the Vesiculation-Driven Fragmentation of Viscoelastic Media and its Implications for Plinian-Style Eruptions
AU: * Taddeucci, J
EM: taddeucci@ingv.it
AF: Dept. of Earth and Environmental Sciences, Univ. of Munich., Theresienstrasse, 41 III, Munich, 80333 Germany
AU: * Taddeucci, J
EM: taddeucci@ingv.it
AF: INGV-RM1, Via di Vigna Murata, 605, Rome, 00143 Italy
AU: Spieler, O
EM:
AF: Dept. of Earth and Environmental Sciences, Univ. of Munich., Theresienstrasse, 41 III, Munich, 80333 Germany
AU: Ichihara, M
EM:
AF: CNEAS, Tohoku Univ., Aobaku-Sendai, Sendai, 980-8576 Japan
AU: Dingwell, D B
EM:
AF: Dept. of Earth and Environmental Sciences, Univ. of Munich., Theresienstrasse, 41 III, Munich, 80333 Germany
AU: Scarlato, P
EM:
AF: INGV-RM1, Via di Vigna Murata, 605, Rome, 00143 Italy
AB: The exsolution and expansion of volatiles from magma are well acknowledged as the driving force for Plinian-style volcanic eruptions. However, it is still unclear whether during such activity magma fragments as a result of gas expansion within individual bubbles or the fast flow that expansion imposes upon it. Here we present the results of shock-tube experiments on the exsolution, expansion, and fragmentation of a gas-saturated viscoelastic medium, and their implications for Plinian-style eruptions.\\ Our sample material is a silicon polymer commercially named ``Silly Putty\copyright''. We measure its viscoelasticity with a forced oscillation rheometer: the polymer is shear-thinning in the strain timescale 10$^{3}$-10$^{-2}$ s, with its viscosity decreasing almost two orders of magnitude, and its relaxation time is of 0.2 s at 25$\deg$C. To evaluate the effect of argon exsolution on the viscosity of the sample we measure the rate of sinking of a steel rod into the sample under gas and atmospheric pressure. The results shows only a minor increase in viscosity as a consequence of sample compaction under gas pressure.\\ The experiments take place in a high-pressure Plexiglas chamber (volume ca. 10 cm$^{3}$, pressure up to 20 MPa) containing a sample of argon-saturated polymer. Sudden decompression of the chamber causes argon exsolution, bubble growth, expansion and flow of the polymer, and its fragmentation. A digital camcorder and pressure transducers record sample expansion and fragmentation and pressure profiles at the bottom and above the sample, respectively. Experimental variables include duration of the sample saturation phase, saturation pressure, and lubrication between sample and chamber. On decompression, the sample: 1) starts to exsolve gas and, after an interval of the order of 0.05 s, to expand; 2) accelerates for another 0.1 s; 3) reaches its maximum expansion after 1 s; 4) the now-foamed sample collapses in 10 s. During the second phase brittle cracks form in the top, fast-accelerating part of the sample, with geometries ranging from sub-horizontal in experiments with no lubricant to complex patterns in those with glycerin. The cracks rarely cut the whole sample into fragments, but mostly remain confined and, as the sample keeps on expanding, tend to heal. Using the video images we calculate the following parameters as a function of time: velocity of the sample front, porosity of the sample, and elongational strain rate: peak values are 1 ms$^{-1}$, 0.9, and 13 s$^{-1}$, respectively. From the length of sample that vesiculates and the saturation duration we estimate the diffusion coefficient of gas in the sample to be in the order of 5x10$^{-9}$ m$^{2}$s$^{-1}$.\\ Brittle fragmentation of the polymer implies a solid-like behavior. The measured peak elongational strain rate is of a time scale significantly shorter than the relaxation time of the polymer, which, during the acceleration phase, is thus forced to cross a rheological boundary. At this point the unrelaxed sample cannot flow any longer and, on overcoming the material tensile strength, velocity differentials are accommodated by fracturing. Note that after the acceleration phase the strain rate drops below 5 s$^{-1}$ and the fractures in the now relaxed sample starts to heal. A process similar to the one described above has been suggested for the fragmentation of magma during Plinian-style eruptions and has received confirmation by numerical simulations. Our experiments support this mechanism and have the potential to further parameterize the process.
DE: 8414 Eruption mechanisms
SC: Volcanology, Geochemistry, Petrology [V]
MN: 2003 Fall Meeting