Prior to construction of building foundations, dams, roads, and other infrastructure components, soil improvement is often required to improve the strength and/or liquefaction resistance of sand deposits. After improvement, initial tests can show low strength, seemingly indicating inadequate improvement. However, after manifestation of time-dependent strength gain in recently improved, and therefore disturbed granular materials, commonly called sand aging, low initial test results are shown to be not indicative of long term behaviors. The motivation for this research is to predict sand aging effects on penetration resistance in order to prevent construction delays due to initial failure to meet quality assurance metrics.

Three field experiments and a laboratory experimentation program were conducted as part of this study. Loose sand layers in Griffin, Indiana and New Madrid, Missouri were disturbed using explosive densification, vibroseis shaking, and impact pier installation. Different methods of disturbance were used in order to study the effects of varying energy inputs and aeration of the pore fluid. In situ geotechnical tests, including the cone penetration test (CPT), dilatometer test (DMT), vision CPT (VisCPT), and shear wave velocity (V_s) measurements, were conducted at each experiment site to quantify sand aging. Time-dependent strength gain was recorded following explosive compaction and impact pier installation. Additionally, cyclic triaxial testing on reconstituted samples of Griffin sand showed an increase in liquefaction resistance with time following sample preparation.

The effects of factors proposed to influence sand aging behavior were investigated using a larger database of sand aging case histories, including the three experiments performed as part of this research project. A method of predicting sand aging based on disturbance method, effective vertical stress, and time is presented. This prediction method represents an improvement over previously proposed prediction methods due to its ease of application at new sites requiring soil improvement. The long term impact of this research is to improve current methods of predicting sand aging effects, reducing the risk of construction delays and unnecessary additional soil improvement.