Silty soils are a common occurrence and can exhibit low strengths, and minimal bearing capacity, causing widespread construction and performance problems. These soils are highly moisture-sensitive and their stability is greatly influenced by the degree of densification achieved during compaction. The strength and stiffness of silty-subgrade soils is also greatly reduced when moisture infiltrates the compacted soil during post construction period. This investigation concentrated on further developing the description of the pumping problem for soils with high silt content and the development of guidelines for stabilization of problematic silts. Using documented field experiences of the DOTD districts, the study presents a testing program to investigate the nature of the problem, the character of the silt materials, and their performance with modifying/stabilizing agents. Seven representative silt soil samples were used in the laboratory program. The unique characteristics that identify a soil with a high potential to pump were determined with standard laboratory tests. The response and stability of the silts under compaction and loading with various moisture levels and compaction efforts was also tested. The susceptibility to pumping of the different samples was reviewed in terms of their physical characteristics. In addition to the silt content percentage, the plasticity character was noted as being significant. The potential for the modification/stabilization of the problem silt soils was also studied. The laboratory tests were selected with respect to construction needs and possible post construction conditions. A limited number of specific additives were proposed with consideration for their ability to dry the subgrade silts sufficiently to be compacted and with the strength to provide a "working table" for the construction of the base and pavement. The additives selected included hydrated lime, portland cement, and class c fly ash. A unitary set of guidelines for stabilizing problematic silts is proposed.

Frost heave damage to roadways costs millions of dollars every year. The need for an improved understanding of the fundamental mechanisms associated with frost heave and methods for efficiently improving frost-susceptible materials prompted the Department of Civil and Environmental Engineering at Brigham Young University (BYU) to undertake a project to design, construct, and verify the functionality of a new frost heave testing apparatus. Frost heave research has been carried out for more than 75 years. The equipment used to conduct this testing has advanced in accuracy and utility over the years. To establish a background in past and current frost heave research, a survey of 12 frost heave devices, including their construction and capabilities, was performed in this research. Several design objectives were then delineated, and a nine-specimen frost heave device was designed and constructed to meet the specifications. The apparatus uses one collective heat source and one collective heat sink for all nine specimens. Heave data and temperature data are collected electronically, while the weights of the specimens before and after frost heave testing are measured manually. Preparatory tests were conducted to confirm the functionality of the data acquisition systems, the uniformity of conditions experienced by all specimens, and the replication of natural roadway freezing conditions. Once preparatory testing was complete, a full-scale frost heave test was performed using the apparatus to investigate the efficacy of cement stabilization in reducing the frost susceptibility of a Montana silt and to validate the functionality of the finished device. Results from the testing indicate that adding 2.0 percent cement actually induces frost heave in excess of that exhibited by the untreated soil. However, additions of 3.5 percent and 5.0 percent were found to be effective in preventing frost heave. Although minor, recommendations for further improvements to the frost heave apparatus include provisions to further decrease the thermal gradient across the specimen and installation of an automatic temperature control device for the water source.

The purpose of this research was to investigate the effects of reclaimed asphalt pavement (RAP) content and cement content on the strength and durability of recycled aggregate base materials. Specifically, the unconfined compressive strength (UCS) and final dielectric value in the Tube Suction Test (TST) were measured in a full-factorial experimental design including five RAP contents, five cement contents, and three replicate specimens of each possible treatment. Specimen mixtures consisted of 0, 25, 50, 75, or 100 percent RAP and 0.0, 0.5, 1.0, 1.5, or 2.0 percent Type I/II Portland cement. Both the RAP and base materials were sampled from the I-84 pavement reconstruction project performed in Weber Canyon near Morgan, Utah, during the summers of 2004 and 2005. The laboratory testing procedures consisted of material characterizations, specimen preparation, and subjection of the specimens to strength and durability testing, and the data were evaluated using analysis of variance (ANOVA) testing. Both the RAP and base materials included in this research were determined to be non-plastic, and the AASHTO and Unified soil classifications for the RAP material were determined to be A-1-a and SM (well-graded sand with gravel), respectively, and for the base material they were A-1-a and SW-SM (well-graded sand with silt and gravel), respectively. The optimum moisture contents (OMCs) for the blended materials were between 5.6 and 6.6 percent, and maximum dry density (MDD) values were between 129.7 and 135.5 lb/ft3. In both cases, decreasing values were associated with increasing RAP contents. The results of the ANOVA performed on the UCS data indicate that UCS decreases from 425 to 208 psi as RAP content increases from 0 to 100 percent and increases from 63 to 564 psi as cement content increases from 0.0 to 2.0 percent. Similarly, the final dielectric value decreases from 14.9 to 6.1 as RAP content increases from 0 to 100 percent and decreases from 14.0 to 5.8 as cement content increases from 0.0 to 2.0 percent. With design criteria requiring 7-day UCS values between 300 and 400 psi and final dielectric values less than 10 in the TST, the results of this research suggest that milling plans should be utilized to achieve RAP contents in the range of 50 to 75 percent, and a cement content of 1.0 percent should be specified for this material. Cement contents less than 1.0 percent are not sufficient to stabilize the material, and greater cement contents may cause cracking. Because control of the actual cement content in the field depends on the contractor's equipment and skill, inspection protocols should be implemented during construction to ensure high-quality work. Additional recommendations are associated with the construction process. The specimens prepared in this research were compacted to relative densities of 100 percent using modified Proctor energy. Therefore, field compaction levels must approach these density values if the same material properties are to be achieved. In addition, all specimens tested in this study were cured at 100 percent relative humidity. Following compaction in the field, cement-treated layers should be moistened frequently during the first few days after construction or promptly sealed with a prime coat or wearing surface to ensure that the cement continues to hydrate. Variability in RAP and cement contents should also be minimized to achieve consistent material properties.

Frost heave is a primary mechanism of pavement distress in cold regions. The distress exhibited is dependent on the frost susceptibility of the soil within the depth of frost penetration, the availability of subsurface water, and the duration of freezing surface temperatures. Cement stabilization is one technique used to mitigate the effects of frost heave. The tube suction test (TST) is one possible method for determining the frost susceptibility of soils in the laboratory. The purpose of this research was to assess the utility of the TST for identifying non-frost-susceptible (NFS) materials stabilized with cement. This research investigated two aggregate base materials from Alaska that have exhibited negligible frost susceptibility in the field. The unconfined compressive strength (UCS), final dielectric value in the TST, and frost heave at three levels of cement treatment and in the untreated condition were evaluated for both materials. The data collected in this research indicate that, for the two known NFS materials included in this study, the TST is a good indicator of frost heave behavior. The total heave of the untreated materials was approximately 0.15 in. at the conclusion of the 10-day freezing period, which classifies these materials as NFS according to the U.S Army Corp of Engineers. Both materials had final dielectric values of less than 10 in the TST, indicating a superior moisture susceptibility rating. The results of this research suggest that the TST should be considered for identifying NFS materials, including those stabilized with cement. Additional testing should be performed on known NFS materials stabilized with cement and other additives to further assess the validity of using the TST to differentiate between frost-susceptible and NFS materials. Consistent with previous studies, this research indicates that, once a sufficient amount of cement has been added to significantly reduce frost heave, additional cement has only a marginal effect on further reduction. Therefore, to avoid unnecessary expense in construction, the minimum cement content required for preventing frost heave should be identified through laboratory testing and specified by the engineer. In this work, UCS values ranging between 200 psi and 400 psi after a 7-day cure were typically associated with this minimum cement content. Because the scope of this research is limited to two aggregate base materials, further testing is also necessary to validate this finding.

Reuse of reclaimed asphalt pavement (RAP) in the full-depth recycling (FDR) process is a cost-effective and environmentally responsible method of asphalt pavement reconstruction. Although FDR has been used for several years in some locations, the effect of RAP on the mechanical properties of recycled base materials has not been well documented. The purpose of this research was to investigate the influence of RAP on the mechanical properties of recycled base materials typical of northern Utah. Two sources of RAP, two sources of base, and RAP contents of 0, 25, 50, 75, and 100 percent were utilized in a full-factorial experimental design with three replicates of each unique combination. Testing procedures consisted of material classifications, compaction tests, and evaluations of strength, stiffness, and moisture susceptibility of each material blend. The California bearing ratio (CBR) test was used to measure strength, the free-free resonant column test was used to measure stiffness, and the tube suction test (TST) was used to measure moisture susceptibility. Once all the testing was completed, a fixed effects analysis of variance (ANOVA) was performed on each of the test results, or dependent variables. The independent variables were RAP content, RAP type, and base type, together with all their interactions. Results of the ANOVA were used to quantify the effects of RAP on the mechanical properties of the base materials. The data indicate that CBR values decrease as RAP content increases, with the greatest percentage reduction occurring with the addition of 25 percent RAP. For stiffness testing at the optimum moisture content determined for each blend, the general trend was a decrease in stiffness from 0 percent RAP to 25 percent RAP, followed by a steady increase in stiffness as the RAP content was increased from 25 to 100 percent. Following a 72-hr drying period at 140ºF, however, the general trend reversed; an increase in stiffness occurred as the RAP content was increased from 0 to 25 percent, and a steady decrease in stiffness was observed for RAP contents above 25 percent. The TST data suggest that additions of 25 and 50 percent RAP actually increase the moisture susceptibility of the recycled material compared to the neat base, although the blended material tested in this study was classified as non-moisture-susceptible when the RAP content was 75 percent or higher. Because of the marked impact of RAP content on the mechanical properties of recycled base materials, engineers should accurately determine asphalt layer thicknesses prior to pavement reconstruction and carefully determine the optimum blending depth for each project. While asphalt milling or base overlays may be required in some locations to avoid excessively high RAP contents, reduced blending depths may be warranted in other areas to prevent the use of low RAP contents. In summary, while the use of RAP in the FDR process is environmentally responsible and offers potentially significant cost savings, thicker pavement base layers, base stabilization, or both may be required in many instances to ensure adequate long-term pavement performance.

The Portland Cement Association commissioned a research project at Brigham Young University to compare selected laboratory durability tests available for assessing stabilized aggregate base materials. The laboratory research associated with this project involved two granular base materials, three stabilizers at three concentration levels each, and three durability tests in a full-factorial experimental design. The granular base materials consisted of an aggregate-reclaimed asphalt pavement blend obtained from Interstate 84 (I-84) and a crushed limestone obtained from U.S. Highway 91 (US-91), while the three stabilizer types included Class C fly ash, lime-fly ash, and Type I/II Portland cement. Specimens were tested for durability using the freeze-thaw test, the vacuum saturation test, and the tube suction test. Analyses of the test results indicated that the unconfined compressive strength (UCS) and retained UCS were higher for specimens tested in freeze-thaw cycling than the corresponding values associated with vacuum saturation testing. This observation suggests that the vacuum saturation test is more severe than the freeze-thaw test for materials similar to those evaluated in this research. The analyses also indicated that the I-84 material retained more strength during freeze-thaw cycling and vacuum saturation and exhibited lower final dielectric values during tube suction testing than the US-91 material. Although the I-84 material performed better than the US-91 material, the I-84 material required higher stabilizer concentrations to reach the target 7-day UCS values specified in this research. After freeze-thaw testing, the Class C fly-treated specimens were significantly stronger than both lime-fly ash- and cement-treated specimens. In the vacuum saturation test, none of the three stabilizer types were significantly different from each other with respect to either UCS or retained UCS. Dielectric values measured during tube suction testing were lowest for cement-treated specimens, indicating that cement performed better than other stabilizers in reducing the moisture/frost susceptibility of the treated materials. The results also show that, as the stabilizer concentration level increased from low to high, specimens performed better in nearly all cases. A strong correlation was identified between UCS after the freeze-thaw test and UCS after the vacuum saturation test, while very weak correlations were observed between the final dielectric value after tube suction testing and all other response variables. Differences in variability between test results were determined to be statistically insignificant. Engineers interested in specifying a comparatively severe laboratory durability test should consider vacuum saturation testing for specimens treated with stabilizers similar to those evaluated in this research. The vacuum saturation test is superior to both the freeze-thaw and tube suction tests because of the shorter duration and lack of a need for daily specimen monitoring. Although the Class C fly ash used in this research performed well, further investigation of various sources of Class C fly ash is recommended because of the variability inherent in that material. Similar research should be performed on subgrade soils, which are also routinely stabilized in pavement construction. Research related to long-term field performance of stabilized materials should be conducted to develop appropriate thresholds for laboratory UCS values in conjunction with vacuum saturation testing.