Thermal properties, density, and porosity of Rakaia Terrane, Torlesse Composite Terrane, New Zealand
New Zealand geological literature is, until now, yet to include a study of the thermal properties of its most common rock type. Rakaia Terrane is the largest constituent of the Torlesse Composite Terrane and represents the basement rock for much of South Island and lower North Island. It comprises an assortment of argillite, greywacke and metamorphosed equivalents. In this study, 89 samples were cut into 4 cm cubes and thermal properties, porosity, density and chemistry were measured. A portable electronic divided-bar apparatus (PEDB) was used to measure thermal properties and portable X-ray fluorescence spectrometer (pXRF) was used to measure chemistry. Elemental ratios show that 82 of the 89 samples cluster, i.e. they have a chemistry typical for a Torlesse rock, and an argillite-greywacke index (AGI) is formulated that quantifies how sandy each typical sample is: mean argillite has AGI=0; and greywacke has AGI=1.
Porosity of typical Rakaia Terrane rocks is low (generally < 3 %) and density (2600 ± 200 kg m-3), volumetric heat capacity (2.2 ± 0.3 MJ K-1 m3) and specific heat capacity (835 ± 15 J K-1 kg-1) are similar across all rock types and metamorphic grades. Thermal conductivity is positively correlated with AGI. Argillite has less variation in chemistry than greywacke but greater variability in thermal conductivity, because phyllosilicates play an important role in thermal properties. Thermal conductivities are: 2.59 ± 0.74 W m 1 K 1 for argillite; 3.07 ± 0.50 W m 1 K 1 for greywacke; 2.51 ± 0.40 W m 1 K 1 for semischist measured perpendicular to foliation; 3.89 ± 0.63 W m 1 K 1 for semischist parallel to foliation; 2.62 ± 0.72 W m 1 K 1 for schist perpendicular to foliation; and 4.10 ± 0.84 W m 1 K 1 for schist parallel to foliation. Thermal diffusivity is correlated with thermal conductivity since volumetric heat capacity has low variability: 1.43 ± 0.26 mm2 s-1 for greywacke; 1.14 ± 0.30 mm2 s-1 for argillite; 1.20 ± 0.29 mm2 s-1 for semischist measured perpendicular to foliation; 1.75 ± 0.37 mm2 s-1 for semischist measured parallel to foliation; 1.27 ± 0.41 mm2 s-1 for schist measured perpendicular to foliation; and 1.90 ± 0.35 mm2 s-1 for schist measured parallel to foliation. Mean thermal conductivity slightly increases with textural zone and metamorphic grade, reflecting higher thermal conductivity of constituent minerals. Metamorphism introduces anisotropy via fabric formation and compositional segregation. The anisotropy factor, i.e. the highest thermal conductivity of a sample divided by its lowest value, of each lithotype is: 1.15 ± 0.14 for greywacke; 1.39 ± 0.34 for argillite; 1.68 ± 0.35 for semischist; and 1.82 ± 0.50 for schist. Theoretical models imply that compositional layering is less important than alignment of phyllosilicates: bedded greywacke-argillite units have anisotropy indistinguishable from uniform argillite; and the quantity and composition of mica-rich layers (rather than the number of quartz segregations) is the primary cause of thermal anisotropy in schistose rocks.
Atypical rocks (7 samples analysed) provide anecdotal insights into the effects of volcanic, biogenic, and hydrothermally-altered components. Biogenic calcium carbonate slightly increases thermal conductivity of argillite to 2.5-3.1 W m 1 K 1. Anomalously-coloured argillite with variable volcanic components has conductivity 2.1-2.9 W m 1 K 1, which is indistinguishable from typical argillite. Hydrothermal alteration can leach potassium and dissolve mica, resulting in higher thermal conductivities of 3.1-3.7 W m 1 K 1; however, one measurement of 2.3 W m 1 K 1 shows that hydrothermal alteration can also lower thermal conductivity.