The present study was directed to the modification of the fine grained fraction of a typical clayey soil from the Southeast of Brazil, specifically from the Zona da Mata de Minas Gerais, for geoenvironmental applications. The study encompassed the following steps: (i) construction of an equipment for performing column percolation test using high hydraulic gradients; (ii) development of soil ionic complex change processes regarding field applications; (iii) column percolation tests in the soil and modified soils in order to evaluate the influence of the hydraulic gradient in theirs hydraulic conductivity and retention capacity of the heavy metals Cd, Mn, Ni, Pb and Zn; (iv) comparison of geotechnical parameters of the soil and the modified soils; and (v) evaluation of the distribution of heavy metals over the tested soils specimens and of the potential adsorption intensity of metals by the soils via sequential extraction. For the purposes of terminology, the natural soil and the chemically modified soils were named: natural soil (ST); soil ST treated with calcium (TCa); soil ST treated with sodium (TNa); and soil ST treated with hydrated lime (TCAL). A multispecies solution with the heavy metals lead, cadmium, manganese, nickel and zinc and the hydraulic gradients of 15, 45, 66 and 87 were used in the column percolation tests. Based on the testing data program, and for practical engineering purposes it was concluded that: (i) the equipment developed to perform the column percolation tests met the desired objectives; (ii) in comparison with the soil ST, the soil TCa and TNa presented significant changes in the Atterberg limits named LL and PL, depicting more plastic behavior, and increase in the maximum dry specific weight and drop in the optimum moisture content, as well as non-significant change in grain size distribution; (iii) in the TCAL soil, it was noted reduction of the soil fine fraction, and non-significant change in the parameters optimum compaction parameters; (iv) changes in hydraulic gradient, regardless of the type of soil, did not produce significant variations in coefficient of hydraulic conductivity of soils; (v) regarding soils ST, TCa and TNa, hydraulic gradient variations reflected in different behaviors of soils heavy metals retention processes, revealing a trend of reduction of the retardation factor with reduction of the applied hydraulic gradient; however regarding the soil TCAL, it was not possible to calculate its retardation factor, since all metals were retained in the soil mass; (vi) from an engineering point of view, the tested soils showed no significant changes in their mechanical strength but the soil TCAL who presented major improvement based on data from unconfined compression, direct shear and quasi-static cone penetrometer tests, also presenting this soil more ductile mechanical behavior than the others; (vii) with regard to the results of the shrinkage tests, it was not detected significant change in comparison with soil ST; and (viii) taking up the soil ST as a reference, the soil TCAL was one that presented greater ability of heavy metal adsorption, since practically all the amount of multispecies solution was adsorbed on its mass; with respect to other soils, they showed similar behaviors, and sometimes even lower adsorption capacity than the soil ST.
