Laboratory and field evaluation of soil stabilization using cement kiln dust

A field and laboratory study was undertaken to evaluate Cement Kiln Dust (CKD) as a soil stabilizer. The perfonnance of CKD from three different sources, Holnam of Ada, Blue Circle Cement of Tulsa, and Lone Star Industries, Inc. of Pryor was compared to the performance of Quick Lime. The field component involved the construction of four test sections along a rural highway near Lula, Oklahoma. During construction, observations were made so that the construction requirements for different additives could be compared. In addition, soil samples were collected before stabilization for the purpose of soil classification. Treated soil samples were collected prior to compaction to prepare laboratory specimens for unconfined compression testing. Field testing included Dynamic Cone Penetration (DCP) testing in the stabilized subbase, and Falling Weight Deflectometer (FWD) testing after completion of the pavement. An in-depth laboratory study was conducted on a clay and sand soil obtained from a cut section next to the test site. These two soils represented a broad range of possible subgrade soil types along the test sections. Unconfined compression tests were conducted on samples prepared with each of the four additives (3 CK.Ds, lime) after moist-curing for 3, 7, 14, 28, and 90 days. Tests on raw shale were made for comparison. In addition, Atterberg limit tests were conducted on both moist- and dry-cured mixtures of shale, as well as two additional soil types taken from different sites. One of the CKDs and lime were used to investigate the durability of treated soils under freeze-thaw and wetdry cycles. This was accomplished by evaluating the unconfined compression strength of samples after different number of durability cycles. Other laboratory testing included: swell testing, California Bearing Ratio testing, pH testing, and Scanning Electron Microscopy. Results from the field study showed that the CKD obtained from the Holnam plant in Ada gave the best performance overall. Quick Lime, Lone Star CKD and Blue Circle CKD perfonned similarly. Results of the laboratory study were more conclusive because many of the variables that can not be controlled in the field, such as degree of mixing, weather, and subgrade variation, are not a significant factor under laboratory conditions. The laboratory test data showed that overall, CKD was at least as effective if not more effective than Quick Lime for stabilizing the shale. For sand, CKD is clearly a more effective stabilizer than Quick Lime, as expected. The influence of CKD and lime on the PI of the three soils that were tested was similar. Durability tests showed that CKD-treated shale was more durable that lime-treated shale, and CKD treated sand was much more durable than CKD treated shale. In addition, durability of the treated soils was better than the untreated soils. Overall, the results of the study suggest that CKD can be an effective soil stabilizer; however, because of the variation in effectiveness between the three CKD sources, the use of CKD should be evaluated on a job specific basis.Final Report (October 1997-April 1999)N