European Journal of Mineralogy (Aug 2023)

H<sub>2</sub>O degassing triggered by alkali depletion in bimodal magma injection processes – a new experimental approach

  • P. L. Marks,
  • A. Allabar,
  • M. Nowak

DOI
https://doi.org/10.5194/ejm-35-613-2023
Journal volume & issue
Vol. 35
pp. 613 – 633

Abstract

Read online

The injection of mafic magma into a hydrous felsic magma chamber is a potential trigger mechanism for bimodal explosive volcanism. As H2O is the most abundant volatile component in magmas, the interaction and the degassing behavior of mildly peralkaline hydrous rhyolitic melt in contact with hydrous basaltic melt were investigated by decompression experiments. Preparatory hydration experiments and bimodal magma decompression experiments, as well as reference experiments, were carried out in an internally heated argon pressure vessel. Pre-hydrated rhyolite and basalt cylinders were perfectly contacted together in a precious-metal capsule, heated to 1348 K at 210 MPa, and thermally equilibrated for 10 min. The initial sample properties were determined by a bimodal reference experiment, quenched immediately after equilibration. To simulate the magma ascent, three bimodal samples and a decompression experiment with two contacted rhyolite cylinders for testing the experimental setup were decompressed with 0.17 or 1.7 MPa s−1 to the final pressure of 100 MPa and then quenched. All decompression experiments resulted in vesiculated samples. The H2O vesicles observed in the decompressed sample of the monomodal rhyolite–rhyolite reference experiment are homogeneously distributed throughout the sample. The former interface between the contacted glass cylinders is invisible after decompression and quench. This reference experiment proves that the two-cylinder design does not influence the degassing behavior of the hydrous melt, e.g., an increased formation of vesicles at possible nucleation sites at the contact plane of the cylinders. The undecompressed bimodal rhyolite–basalt sample shows crystal-free rhyolitic glass, whereas 3 µm sized idiomorphic magnetite crystals coexist with glass in the basaltic part of the sample. Within the 10 min run time, a ∼ 300 µm wide hybrid composition zone developed between the hydrous rhyolitic and basaltic endmembers, caused by diffusion-induced mixing processes. Decompression and quenching of the bimodal melts resulted in vesiculated glass samples. A ∼ 100 µm wide zone of alkali-depleted rhyolitic glass as part of the ∼ 300–560 µm wide hybrid zone is covered with an enhanced number of H2O vesicles compared to the pristine rhyolitic and basaltic glass volumes. We suggest that this enhanced vesiculated zone forms by a rapid diffusional loss of alkalis from the mildly peralkaline rhyolitic melt into the basaltic melt of the sample. The reduced alkali concentration significantly reduces the H2O solubility of the rhyolitic melt. This process enhances the H2O supersaturation necessary for vesicle formation during decompression. In summary, the new findings imply that convective magma ascent driven by the injection of hot basaltic magma into a hydrous peralkaline rhyolitic melt reservoir leads to enhanced H2O vesicle formation near the melt interface and thus to efficient degassing. This in turn can accelerate buoyancy-driven magma ascent and mingling and mixing processes that induce further degassing and potentially trigger explosive volcanic eruptions.