Estimates of Crystallinity Utilizing Differential Scanning Calorimetry: Application to the Kīlauea 2018 Lower East Rift Zone Eruption

Abstract

Abstract Rocks produced by diverse processes, from condensation in space to impacts on planetary surfaces to volcanism, contain both crystals and amorphous material. Crystallinity provides information on the thermal history of the sample and is especially important in characterizing volcanic rocks and pyroclasts because lava rheology is profoundly influenced by the crystal content. Crystallinity is typically quantified via microscopy, using transmitted light or backscattered electrons. However, many samples present visibly ambiguous textures such as intimate intergrowth of crystal phases, and/or crystal sizes extending down to the nanometer scale. Here, we apply calorimetric methods involving heat capacity and enthalpy to assess the crystallinity of a series of volcanic samples. We tested three different approaches, using differential scanning calorimetry, on 30–40 mg aliquots of powdered basalts from the 2018 Kīlauea lower East Rift Zone. The first approach involves determining the magnitude of the increase in heat capacity at the glass transition, which can determine crystallinity to a 1σ precision of ±3%. The second approach is based on the enthalpy of fusion, which requires a longer more complex procedure with results that are typically more uncertain than for the heat capacity method, with a 1σ of ±6%. A final method utilizing differences in enthalpies calculated from the heat capacities required the most complex procedure and has the greatest uncertainty of ±18%. Preliminary results for lavas with microscopically determined crystallinities ranging from 11 to 98% indicate that crystallinity based on calorimetric data can be tens of percent higher than the average value identified using microscopy and petrographic analysis. Image-based methodologies applied to sections of samples reveal spatial heterogeneity and details in texture and crystallinity, whereas calorimetry-based methodologies capture the overall ‘bulk sample’ properties, unbiased by section effects or imaging resolution limits. These techniques are a powerful combination that can present complementary views of crystallinity.