This study investigates the effect of temperature on the interaction mechanisms between reinforcing geotextiles confined in unsaturated, compacted silt. The results and analysis from this study are relevant to the evaluation of the effects of incorporating geothermal heat exchangers into mechanically-stabilized earth (MSE) retaining walls constructed with poorly draining backfill. A thermo-mechanical geosynthetic pullout device was used in this study that incorporates standard components for geosynthetic pullout or creep testing including a rigid soil box with an integrated vertical loading system, a roller grip to apply pullout forces uniformly to the geotextiles, a pulley system for load-controlled creep testing, a servo-motor for displacement-controlled monotonic pullout testing, and instrumentation for monitoring vertical settlement, pullout force, and pullout displacement measurements. Further, the pullout device incorporates heating elements at the top and the bottom of the soil box to apply constant temperature boundary conditions to the soil layer as well as dielectric sensors embedded at different depths in the soil layer to monitor the soil temperature and volumetric water content. Two sets of pullout tests were performed on geotextiles within compacted silt layers having initial degrees of saturation of 0.44. The first involves monotonic pullout of a woven polypropylene (PP) geotextile after reaching steady-state conditions under different boundary temperatures without a seating load, and the second involving monotonic pullout of a woven polyethylene-terephthalate (PET) geotextile after reaching steady-state conditions under different boundary temperatures while under a constant seating pullout load. The second testing series permits evaluation of possible thermally-induced creep displacements. The boundary temperatures investigated in this study are typical of geothermal heat exchange systems and range from 20 to 50 °C. These temperatures are lower than the glass transition temperature of the PET geotextile but greater than that of the PP geotextile.

The results from the two testing series indicate that the ultimate pullout resistance of the geotextiles heated with and without a seating load decreased with increasing temperature. Although heating led to drying of the silt layer throughout most of its height, as expected, water was observed to accumulate at the soil-geotextile interfaces leading to an increase in degree of saturation at this location. An effective stress analysis considering thermal softening mechanisms in soils indicates that the increase in degree of saturation at the soil-geotextile interface was the primary cause of the decrease in pullout resistance. The rate of decrease in ultimate pullout resistance with temperature was similar for both geotextiles tested, indicating that application of temperatures to the polypropylene geotextile greater than its glass transition temperature do not have a major effect on its nonisothermal response.