Abstract
Two Methods (Method A and Method B) used to determine the in-situ horizontal stress of cohesionless soil are presented. Method A is a laboratory test. In test Method A, a high-quality undisturbed sample of cohesionless soil recovered by the in-situ freezing technique was thawed under no-lateral strain condition in a special cell under the in-situ vertical stress. The cell pressure at the time when the frozen sample was completely thawed indicates the in-situ horizontal stress. Method B is a field test. In Test Method B, a special measuring device was inserted into a frozen borehole below subsoil provided by in-situ freezing sampling. The pressure measured in the device when the surrounding subsoil was completely thawed indicates the in-situ horizontal stress. A series of comparative tests to confirm the reliability of these two proposed methods were performed in the laboratory. Sand specimens with given stress histories (overconsolidation ratio = 2, 4 and 6) were frozen one-dimensionally and then used in comparative tests. The horizontal stress in the sand specimen before freezing was directly measured for comparison. Good agreement of the horizontal stress measured with Test Method A and in the sand before freezing was measured. Good agreement of the horizontal stress measured in the Method B and in the sand before freezing was also observed. These results indicate that a freeze-thaw sequence used in this study did not change the state of stress in the sand before freezing. The results also indicated that both Test Method A and Test Method B proposed in this study are reliable for measuring the in-situ state of stress in the cohesionless soil.
