LABORATORY TEST METHODS Soil Mixture Preparation

treatment temperature increased, especially at temperatures in excess of 450~ However, the strength decreases were small (less than 15%), and the concrete strength was still within typical values for structural concrete. The leachability tests showed that as treatment temperature was increased, levels of PAHs in the soil significantly decreased. From the leachability tests performed on samples of crushed concrete that contained contaminated soil, it appears that only low molecular weight PAHs (i.e., less than or equal to166 g/moO were stabilized by the concrete. The poor stabilization of higher molecular weight PAHs may be an artifact of the sample preparation method which required that the concrete be crushed prior to leaching. Graduate Research Assistant, Civil & Environmental Engineering, Tufts University, Medford, MA 02155 2 Assistant Professor, Civil & Environmental Engineering, Tufts University, Medford, MA 02155 3 Undergraduate Research Assistant, Civil & Environmental Engineering, Tufts University, Medford, MA 02155 INTRODUCTION Between the mid-1800s and the 1960s, manufactured gas plants (MGPs) were widely used in the U.S. to produce gaseous fuel from available coal, coke, oil and other fossil fuels (Hatheway, 1997). A by-product of coal gasification was coal tar, a dense, non-aqueous phase liquid that was often disposed of on-site in wells, pits and lagoons. Coal tar was also introduced into the subsurface environment due to leaks and spills from tanks and piping networks, as well as the dismantling of plants taken out of service (Luthy et al., 1994). The total number of sites in the U.S. contaminated with coal tar is estimated to be more than 32,000 (Hatheway, 1997). Coal tar contamination is of particular concern because it contains a wide range of hazardous chemicals, including polycyclic aromatic hydrocarbons (PAHs), many of which are known or suspected carcinogens (IARC, 1983). Coal tar-contaminated soil is commonly placed in landfills or treated using thermal processes such as thermal desorption, incineration, and coburning. However, treated soils ot2en contain residual contamination and consequently must be disposed of in solid waste landfills. In a review of 200 coal tar-contaminated sites at which remediation activities have been performed, Owen and Unites (1999) found that 55% of these sites landfilled contaminated soils and 49% employed thermal treatment (desorption or coburning). Reusing contaminated soil or treated soil containing residual contamination may be an alternative to disposal in landfills. One possible use for this soil is as an aggregate in concrete. Incorporating coal tarcontaminated soil into concrete may help to stabilize residual contaminants, preventing them from leaching out of the soil. Also, the possibility of marketing soil with residual contamination for use as aggregate in concrete provides a possible financial incentive for remediation. To date, little work has been done to determine whether thermally treated coal tar-contaminated soil is a suitable concrete aggregate. Several studies have reported relatively high (25% to 98%) stabilization of pure organic compounds (not compounds associated with contaminated soils) in cement (Owens et al., 1996; Faschan et al., 1996; Conner, 1995; Hebatpuria et al., 1999). However, only one study was found in which soils contaminated with organics were incorporated into concrete. In this study it was found that the effective diffusivity of benzene was reduced by as much as five orders of magnitude when contaminated sand was incorporated into concrete (Ezeldin and Vaccari, 1995). Because the investigators did not report on the physical properties of the concrete (i.e., tensile and compressive strength, elastic modulus, etc.), applications for concrete containing contaminated aggregate are not known. This paper presents an evaluation of some of the physical and chemical properties of thermally treated coal tar-contaminated soil and of concrete incorporating this soil as an aggregate. The specific objectives of this study were to determine the effects of thermal treatment on the physical characteristics of natural and coal tar-contaminated soil, evaluate the strength characteristics of concrete containing thermally treated natural and coal tar-contaminated soil as aggregate, and assess the leachability of PAHs from concrete containing coal tar-contaminated soil as aggregate. LABORATORY TEST METHODS Soil Mixture Preparation and Geotechnical Testing Procedure The soil used in this study was a gap-graded mixture of tan to brown subangular to sub-rounded sand and gravel. This type of soil is typical of many former MGP sites (Luthy et al., 1994). The contaminated soil was prepared by mixing coal tar sludge obtained from a former MGP site in New Bedford, Massachusetts with the air-dried sand and gravel mixture. The final contaminated soil was mixed to contain approximately 3,000 ppm PAHs by weight. A furnace was used to thermally treat 10 kg batches of the sand and gravel mixtures. The furnace was preheated to a specified temperature between 250~ and 650~ before each treatment. The soil was heated for one hour, then promptly removed and spread in a large metal pan to cool. The residence time of one hour was meant to approximate actual practice and temperatures of 250~ to 650~ represent the range commonly employed during thermal desorption and incineration. However, the laboratory method did not employ mixing by rotary kiln during thermal treatment or cooling by quenching with water, as is commonly performed in thermal treatment processes. Mechanical sieve tests (ASTM D422) and direct shear tests (ASTM D3080) were performed to characterize the physical properties of the natural and contaminated soil before and after thermal treatment. Two sieve tests were used in order to characterize the full range of particle sizes. A coarse sieve analysis, with sieve sizes between 26.67 mm and 2 mm was initially performed. The portion of soil retained on the 2 mm sieve and in the pan of the coarse sieve analysis was set aside for a fine sieve analysis using sieve sizes between 2 mm and 0.075 mm. The direct shear tests were performed using normal loads of 36 kg and 66 kg during shear. Based on ASTM specifications and the shear mold dimensions, 6.5 cm diameter and 5 cm height, the maximum particle size appropriate for the mold should be 6.5 mm. The soil particle size in this investigation was reduced using a 9.42 mm sieve. Soil was compacted into the shear mold by tamping three equal lifts each with ten blows using a Harvard miniature compactor (290 g hammer dropped 38 cm). The thermally treated soil was also mechanically sieved after direct shear to evaluate particle breakage using the fine sieve analysis. Concrete Mix Design and Strength Testing Procedure The concrete mix design used in this study combined cement, water, coarse aggregate, and sand in the proportions 1 : 0.52 : 3 : 2, respectively, by weight. Table 1 shows the actual mix proportions for the 13 batches of concrete prepared for this study. From each batch, three 10.2 cm x 20.3 cm cylinders were prepared for compression tests, and three 10.2 cm x 15.2 cm cylinders were prepared for splitting tension tests. Elastic modulus, Poisson's ratio, and compressive strength were determined following the procedures of ASTM C369 and ASTM C39. The indirect tensile strengths of the concrete were determined by performing Brazilian splitting tensile tests as per ASTM C496. All specimens were cured in a humid room for 21 days prior to testing.