Abstract

The objective of this research was to investigate the effects of thermal gradient and fines content and the interaction between these two factors on the frost heave characteristics of a typical Alaska base material. The laboratory frost heave testing involved one type of aggregate base material, three thermal gradients, and three fines contents in a full-factorial experimental design with two replicates. The aggregate was classified in the American Association of State Highway and Transportation Officials soil classification system as A-1-a; the thermal gradients were 0.15, 0.30, and 0.45 ºC/cm; and the fines contents were 6, 8, and 10 percent. After frost heave testing, a stepwise regression analysis was performed to identify significant independent variables for each of nine separate dependent variables, including frost heave, heave-uptake ratio, steady-state frost heave rate, gravimetric water ingress, and gravimetric water content in each of the five individual lifts tested following frost heave testing. Soil suction, specific gravity, salinity, and hydraulic conductivity testing were also performed on samples prepared at each of the three fines contents to support numerical modeling of the frost heave test results using the computer program ICE-1. The results of the stepwise regression analysis indicate that thermal gradient is a significant predictor of six of the nine dependent variables and that the square of thermal gradient is a significant predictor of five of these six dependent variables. As the thermal gradient increased, the samples experienced decreasing amounts of water ingress and frost heave. However, the data show that neither fines content nor the square of fines content is a significant predictor of any of the dependent variables. Thus, although previous research has shown that higher fines contents are generally associated with greater susceptibility to frost heave, this effect is not manifest in the comparatively small increases in fines contents evaluated in this research. The interaction between thermal gradient and fines content is a significant predictor of only one independent variable. Differences between the modeled and measured frost heave values ranged from 0.01 to 0.92 cm, with the larger differences typically associated with the lowest thermal gradient and the lowest fines content.

Degree

College and Department

Ira A. Fulton College of Engineering and Technology; Civil and Environmental Engineering

Rights

http://lib.byu.edu/about/copyright/