The life and death of heterogeneity in magmas

Explosive volcanism is one of the most catastrophic material
failure phenomena. During magma ascent, fragmentation produces
particulate magma, which, if deposited above the glass transition
of the interstitial melt, will sinter viscously. In-conduit
tuffisites, conduit wall breccias and ash deposited from
exceptionally hot pyroclastic flows are scenarios in which
sintering by viscous flow is possible. Therefore, understanding the
kinetics of sintering and the characteristic timescales over which
magma densifies are critical to understanding the degassing
timeframe in conduits and deposits. Viscous sintering is
accompanied by a recovery of material strength towards that of a
pore-free, dense magma. Understanding damage mechanisms and seismic
behaviour prior to failure of sintered volcanic products are also
crucial for the application of micromechanical models and material
failure forecasting laws. Powdered standard glass and industrial
glass beads have been used to explore sintering mechanisms at
ambient pressure conditions and temporal evolution of connected and
isolated pore-structure. I observe that sintering under low axial
stress is essentially particle size, surface tension and melt
viscosity controlled. I found that the timescales over which the
bulk density approaches that of a pore-free melt at a given
temperature is dependent on the particle-contact surface area,
which can be estimated from the particle shape, the packing type
and the initial total porosity. Granulometric constraint on the
starting material indicates that the fraction of finer particles
controls the rate of sintering as they cluster in pore spaces
between larger particles and have a higher driving force for
sintering due to their higher surface energy to volume ratio.
Consequently, the resultant sample suite has a range of
microstructures because the viscous sintering process promotes a
fining of pores and a coarsening of particles. In a volcano, newly
formed sintering material will then further contribute to
magma-plugging of the conduit and its mechanical properties will
affect magma rupture and its associated precursory signals. This
consideration permitted me to explore the effect of sintering on
the stress required for dynamic macroscopic failure of synthesised
samples and assess the ability of precursory microseismic signals
to be used as a failure forecast proxy at conditions relevant to
shallow volcanic conduits. To this end, the samples were subjected
to mechanical tests under a constant rate of deformation and at a
temperature in the region of the material glass transition. A dual
acoustic emission rig was employed to track the occurrence of
brittle fracturing. The monitored acoustic dataset was then
exploited to systematically assess the accuracy of the failure
forecasting method as a function of heterogeneity (cast as
porosity) since it acts as nucleating site for fracture
propagation. The pore-emanating crack model describes well the peak
stress at failure in the elastic regime for these materials. I show
that the failure forecast method predicts failure within 0-15%
error at porosities >0.2. However, when porosities are 100%. I interpret these results as a function of
the low efficiency with which strain energy can be released in the
scenario where there are few or no heterogeneities from which
cracks can propagate. These observations shed light on questions
surrounding the variable efficacy of the failure forecast method
applied to active volcanoes. In particular, they provide a
systematic demonstration of the fact that a good understanding of
material properties is required. Thus I wish to emphasise the need
for a better coupling of empirical failure forecasting models with
mechanical parameters, such as failure criteria for heterogeneous
materials, and point to the implications of this for a broad range
of material-based disciplines.