 |
| Titel |
Mixing Experiments with Natural Shoshonitic and Trachytic Melts: A comparative Study Under Contrasting Rheological and Fluid Dynamic Conditions |
| VerfasserIn |
Cristina P. De Campos, Diego Perugini, Maurizio Petrelli, Stephan Kolzenburg, Alexander Dorfman, Donald B. Dingwell |
| Konferenz |
EGU General Assembly 2011
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 13 (2011) |
| Datensatznummer |
250055778
|
|
|
|
|
|
| Zusammenfassung |
| Isotopic evidence for open system processes within the Campi Flegrei magma chambers,
(Arienzo et al., 2008, Bull. Volc.) and previous experimental studies on the behavior of major,
minor and trace elements during mixing of alkaline magmas (De Campos et al., 2008,
Chem.Geol.) motivated this study. In order to simulate the evolution of the mixing process,
we performed experiments using natural volcanic samples from this region. The
end-member melts derive from the Agnano-Monte Spina (trachytic) and Minopoli
(shoshonitic) eruptions. Based on previous isotopic data, these are thought to be the most
suitable counterparts for replicating the extreme interacting compositions in this
system.
Variable time series of advection-diffusion experiments using two different techniques
have been carried out: 1) a high-temperature centrifuge and 2) a viscometer. For the
centrifuge experiments the rotating speed was 1850 revolutions per minute and the
acceleration 10 3 g. In this way, dynamic conditions closer to those calculated for magma
chambers (Reynolds Numbers [Re] around 102) could be simulated. For every experiment, a
4 mm thick disk of previously homogenized crystal free shoshonitic glass and an 8 mm thick
disk of homogenized crystal free trachytic glass were loaded in a 5mm diameter Pt
capsule. The capsule was then sealed on both sides, but for a small opening on the
upper end, allowing interstitial degassing during the acceleration. Samples were
arranged in a buoyantly unstable geometry, where the denser material is placed
at the inner side of the rotating circle (basaltic trachyandesite, Ï=2.63 g/cm3 at
1169oC) and the lighter material at the external side (trachyte, Ï=2.45 g/cm3 at
~1000oC). Viscosity ratio was around 35. Temperature has been kept constant at
1,200oC during all experimental runs, with a negligible thermal gradient (< 1ºC).
Forced convection was applied via centrifugal acceleration and density instabilities.
Results from all experimental runs with the centrifuge after 5, 20 and 120 min will be
discussed.
The second set of experiments consisted of two experimental runs (25- and 168-hours
duration) under Taylor-Couette flow, according to De Campos et al. (2008, Chem.
Geol.). Higher amounts of the same end-members, in different proportions, have
been mixed together using a concentric cylinder viscometer. For the second set
of experiments forced convection has been simulated by stirring with a spindle.
Experimental conditions were constrained by: 1) constant angular velocity (0.5
rotations per minute) and 2) constant temperature (1,300Ë Celsius). It this case the
viscosity ratio between melts was around 30 and the Reynolds Number around
10-7. The experiments terminated by stopping all movement, extracting the spindle
from the sample and letting the sample cool to room temperature. Cylinders of the
resultant mixed glasses were recovered by drilling and, prepared for microprobe
analysis.
Microprobe and ICP-MS analyses along longitudinal lines from sections of all the
resulting products reveal a complex non-linear mixing process with different mobility for
different elements. Comparative studies from both experimental sets highlight the importance
of chaotic advection in enhancing the mixing efficiency. . At higher Re (centrifuge) and
very short mixing times, higher viscosity contrast between the mixing melts (lower
temperature - 1,200oC) can be compensated by chaotic dynamics. At very low
Re (concentric cylinder), longer mixing times and chaotic dynamics may equally
enhance mixing, therefore contrasting with calculated predictions from the literature. |
|
|
|
|
|
|