Basic Problems in Thermal-Conductivity Measurements of Soils

Abstract

The thermal conductivity of pozzolanic soil (a fine sandy, unconsolidated, alluvial soil from Lazio, Italy, based on volcanic ash) and blue marlstone rocks (from Alba, Piedmont, north Italy) was measured, using a thermal probe technique, over a wide range of temperatures from ${-}20\,^\circ \mathrm{C}$ to ${+}20\,^\circ \mathrm{C}$. Unfrozen pozzolanic soil thermal-conductivity data display surprisingly low values about 3 to 4 times smaller than water; for frozen soils, the data are just slightly higher than for the unfrozen state but they are still 2 to 3 times lower than for water and seven times lower than for ice. This outcome is probably due to a high internal porosity of individual volcanic ash particles. The influence of the bulk soil porosity on the measured thermal conductivity was found to be rather negligible; the observed slight variation of the thermal conductivity is possibly due to the diverse grain size distribution of soil samples excavated from different depths of the ground. The blue marlstone rock has a considerably higher thermal conductivity than pozzolanic soil, likely due to its very small porosity, consolidated structure, and different implicated minerals. The frozen rock has just about a 30 % higher thermal conductivity than that for the unfrozen state. A temperature-dependent thermal conductivity is observed in the freezing state only. Test results show how heat transfer between the thermal probe and surrounding soil is influenced by storage of heat in the tested material, conduction heat flow, water evaporation due to heating, and finally by vapor diffusion and circulation.