This project involved four distinct research activities, (1) the influence of temperature on lime-stabilized soils, (2) the influence of temperature on cement-stabilized soils (3) temperature modeling of stabilized subgrade and (4) use of calcium chloride to accelerate strength gain of cement-stabilized soils.
Significant conclusions from the lime-focused research include that the minimum lime content of a soil increases as temperature decreases. Increased curing duration leads to decreases in pore fluid pH. However, this reduction in pH is less at lower temperatures, indicating that little reactivity occurs below 50°F. Exposure to freeze-thaw cycles or initial curing at 35°F resulted in significant reductions in strength gain for a given curing duration. However, once the freeze-thaw cycles or temperature reduction was removed, strength gain resumed at approximately the same rate. Overall, these results suggest that current specifications may be modified to allow lime stabilization to proceed in cooler temperatures, provided a corresponding increase in curing time and/or thermal protection is provided prior to loading.
The soil cement data indicate that curing soil-cement at lower temperatures will result in lower strengths. For example, the 7 day strength for samples cured at 25oF was less than the strength of samples cured at 50oF or 70oF by a factor ranging from 2-6. Likewise, the 7 day strength for samples cured at 35oF was less than the strength of samples cured at 50oF or 70oF by approximately 20-25%. Additionally, on the basis of 15 repeat tests for 3 and 7 day curing periods, for three different soils, results indicate that the mean strength at 3 days is 84-93% of that for 7 days, in support of a potential change in current subgrade evaluation practice predicated on the longer duration.
Specifications for stabilization work have often been based on air temperature measurements, however the performance of lime or cement treated soil is expected to be more closely related to the in situ temperature. This research has found that the thermal diffusivity of both lime and cement-stabilized subgrades varies from 3.8 x 10-7 m2/s (2.14 in2/hr) to 9.8 x 10-7 m2/s (5.46 in2/hr). These data were incorporated into a model that relates air and soil temperatures. A computer application was developed to use the model to make predictions of subgrade temperatures and cured strength.
A window of efficacy was observed for Buncombe, Guilford and Johnston county soil cement mixes, with ideal ranges at 50oF curing conditions of 0.25% ¿ 1.0%, 0.25%-0.75% and 1.25%-1.75%, respectively (percentages reflect mass of CaCl2per mass of cement). However these optimum ranges vanished or changed to levels untested at 35oF curing conditions. A field trial was conducted with CaCl2 doses of 2.3% and 8.3%, and both of these dosage levels resulted in weakening of the material, as evaluated by in situ dynamic cone penetration tests and unconfined compression testing of field-mixed samples. Laboratory mixed samples of the same material resulted in strength increases. CaCl2 modification increases the electrical conductivity and dielectric value of soil-cement mixtures which might imply increased susceptibility to longer term moisture-induced weakening at high dosage levels. The cost of CaCl2 modification at effective doses is likely to be less than 10% of cement costs. The overall body of research presented in this report suggests that CaCl2 modification of soil-cement is not a mature enough approach to serve as a method for mitigating the effects of low temperatures on strength gain. Additional data are required to probe the sensitivity of temperature, mixing method and soil type.
