Large explosive phonolitic eruptions are commonly characterised by aphyric juvenile eruptive products. Taking into account the low density contrast among phonolitic composition and settling phases (i.e., feldspar and leucite), the almost complete lack of crystals in these differentiated compositions rises the question of which process could produce such an efficient crystal-melt separation. Seeking for an answer, we have investigated crystallization in presence of a thermal gradient as a possible mechanism for crystal-melt separation, considering both chemical and physical effects acting on a variably crystallized system. Using a natural tephri- phonolitic composition as starting material (M.te Aguzzo scoria cone, Sabatini Volcanic District, Central Italy), we have reproduced thermal gradient-driven crystallization in order to simulate the crystallization process in a thermally zoned magma chamber. Crystallization degree (paragenesis made of clinopyroxene±feldspars±leucite) as well as melt composition varies along the thermal gradient. In particular, melt composition ranges from the tephri-phonolitic starting composition at the bottom of the charge (hottest and aphyric zone) to phonolitic at the top (cooler and heterogeneously-crystallised zone). Backscattered images of experimental products clearly evidence: i) the aphyric tephri-phonolitic melt region at the bottom of the charge; ii) a drop-shaped crystal clustering in the middle zone; and iii) large aphyric belt and pockets (up to 100 μm wide) of phonolitic melt, with large deformed-shaped sanidine occurring at their margin, at the charge top region. The latter two features, resulting from solid-melt displacements, suggest that the segregation of phonolitic melt can be related to crystal sinking and compaction. On the other hand, the compositional variability of the melt along the thermal gradient is directly related to the crystallization degree, indicating that chemical diffusion and thermal migration have negligible effect at the experimental scale. Experimental results suggest that, in presence of a thermal gradient, a filter-press differentiation mechanism (i.e. sinking+compaction) is able to produce heterogeneous magma differentiation characterised by a wide range of melt compositions (in our case from tephri-phonolitic to phonolitic). Although the limitation due to the vertical shape of the charge and consequent shear effects occurring at the lateral walls (thus limiting the mobility of the crystal clusters), experimental duration of 24 h is enough to allow crystal-melt separation by means of settling and compaction, indicating that timescale for such a process is extremely rapid and effective at experimental conditions. Actually, experimental textures and phase relations are in good agreement with those observed in natural lithic enclaves (from Sabatini Volcanic District) representative of the crystallizing boundary layer of a phonolitic magma chamber. Thus, we speculate that gravitative collapses of a mushy zone from the magma chamber roof of a thermally zoned magma chamber may produce top accumulation of highly differentiated and aphyric melts.
Origin of Aphyric Phonolitic Magmas: Natural Evidences and Experimental Constraints
MASOTTA, MATTEO;
2010-01-01
Abstract
Large explosive phonolitic eruptions are commonly characterised by aphyric juvenile eruptive products. Taking into account the low density contrast among phonolitic composition and settling phases (i.e., feldspar and leucite), the almost complete lack of crystals in these differentiated compositions rises the question of which process could produce such an efficient crystal-melt separation. Seeking for an answer, we have investigated crystallization in presence of a thermal gradient as a possible mechanism for crystal-melt separation, considering both chemical and physical effects acting on a variably crystallized system. Using a natural tephri- phonolitic composition as starting material (M.te Aguzzo scoria cone, Sabatini Volcanic District, Central Italy), we have reproduced thermal gradient-driven crystallization in order to simulate the crystallization process in a thermally zoned magma chamber. Crystallization degree (paragenesis made of clinopyroxene±feldspars±leucite) as well as melt composition varies along the thermal gradient. In particular, melt composition ranges from the tephri-phonolitic starting composition at the bottom of the charge (hottest and aphyric zone) to phonolitic at the top (cooler and heterogeneously-crystallised zone). Backscattered images of experimental products clearly evidence: i) the aphyric tephri-phonolitic melt region at the bottom of the charge; ii) a drop-shaped crystal clustering in the middle zone; and iii) large aphyric belt and pockets (up to 100 μm wide) of phonolitic melt, with large deformed-shaped sanidine occurring at their margin, at the charge top region. The latter two features, resulting from solid-melt displacements, suggest that the segregation of phonolitic melt can be related to crystal sinking and compaction. On the other hand, the compositional variability of the melt along the thermal gradient is directly related to the crystallization degree, indicating that chemical diffusion and thermal migration have negligible effect at the experimental scale. Experimental results suggest that, in presence of a thermal gradient, a filter-press differentiation mechanism (i.e. sinking+compaction) is able to produce heterogeneous magma differentiation characterised by a wide range of melt compositions (in our case from tephri-phonolitic to phonolitic). Although the limitation due to the vertical shape of the charge and consequent shear effects occurring at the lateral walls (thus limiting the mobility of the crystal clusters), experimental duration of 24 h is enough to allow crystal-melt separation by means of settling and compaction, indicating that timescale for such a process is extremely rapid and effective at experimental conditions. Actually, experimental textures and phase relations are in good agreement with those observed in natural lithic enclaves (from Sabatini Volcanic District) representative of the crystallizing boundary layer of a phonolitic magma chamber. Thus, we speculate that gravitative collapses of a mushy zone from the magma chamber roof of a thermally zoned magma chamber may produce top accumulation of highly differentiated and aphyric melts.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
