| Abstract |
Characterizing the thermohydraulic properties of reservoirs is essential for the development of geothermal projects. Models to predict the thermal conductivity of a rock formation can be useful when direct measurements on samples cannot be made in the laboratory. The objective of this research was to evaluate the variations of thermal conductivity as a function of porosity in a reference reservoir lithology (sandstone), to determine which analytical models can be used to accurately calculate its thermal conductivity. Crab Orchard Sandstone was chosen for its representative reservoir properties making it a suitable analog for the study of sedimentary basins reservoirs. For example, Crab Orchard is used as an analog to the Cairnside and Covey Hill Formations in the St. Lawrence Lowlands sedimentary basin to study carbon dioxide storage combined with geothermal energy extraction. Measurements were carried out on three core samples. Porosity was determined using a combined permeameter and porosimeter, and bulk thermal conductivity was measured on dry and saturated samples using an optical thermal conductivity scanner (TCS). We then used the arithmetic, geometric and harmonic averages to determine which model best reproduces the sample thermal conductivity that was evaluated with the TCS in dry and saturated conditions. The model that gave the most meaningful results was the geometric average with a relative error in thermal conductivity of 5%. The estimated thermal conductivity of solids was 6.34 W/mK. Under a known porosity, we can now predict the Crab Orchard Sandstone thermal conductivity with acceptable margin of error for samples with similar mineralogy. |