DEVELOPMENT OF PREDICTIVE MODEL FOR INVESTIGATING EFFECT OF SODIUM CHLORIDE ON SOIL GEOTECHNICAL PROPERTIES Gbenga M. Ayininuola, Oluwole A. Agbede University of Ibadan Department of Civil Engineering University of Ibadan Ibadan, Nigeria gm.ayininuola@mail.ui.edu.ng and oluwoleagbede@yahoo.com ABSTRACT The study centred on the development of predictive equations to determine soil contaminated with sodium chloride geotechnical properties at given level of cation exchange capacity (CEC), exchangeable anion and cation. Soil samples for the work were taking from two borrow pits in Ibadan, Nigeria. Aside the control, each sample was divided into sixteen and grouped into. Each group received a single dose of 10, 30, 50 and 70 g/dm3 and their geotechnical and chemical monitored for 365 days. The geotechnical properties obtained were used as dependent variables while chemical properties as independent variables in multiple regression model to generate fourteen equations for estimating soil geotechnical properties at specified levels of chemical properties. The results revealed that the two soils geotechnical properties except coefficient of permeability decrease with increase in salt concentrations and time. Also, the differences between the measured and predicted or simulated geotechnical properties were in the range of - 10.08 and +6.51 %. KEY WORDS Model development, Geotechnical properties, Salt concentration, Soil contamination, Mathematical modelling 1. Introduction Nearly all structures are sited on soils: The foundations of structures such as buildings, roads, dams, bridges, etc are designed based on the prevailing circumstances of the soil. After thorough geotechnical study, the type of foundation for a particular structure is agreed upon and construction of interfacial element would commence. Site investigation results show the immediate condition of the soil even though engineers employ factors of safety to predict the likely future condition of the soil with response to loading. This approach often time may not be economical or totally misleading. In many site investigations, little emphasis is placed on the estimation of chemicals present in the soils. When such is considered, their effect on concrete structure, underground water and health is always examined. Nowadays, it has been accepted that soil contains numerous chemicals originating from its formation and other contaminants introduced due to man-made activities. These chemicals undergo a lot of interactions or reactions that cannot be easily ascertained. The products of such reactions would have direct consequence on the resulting soil properties. In addition, structures are built on contaminated soils such as old refuse dumpsite without proper geophysical studies. Little efforts have been exercised to know the impacts of contaminants in soils. The study centered on developing predictive equations using multiple regression statistical model for investigating the influence of sodium chloride on soil geotechnical properties. Yanful (1982) reported increase in hydraulic conductivity of approximately two orders of magnitude for samples of glacial clays from Ontario permeated with brine. Ho (1985) presented data on the effects of a NaCl brine on the grain-size curve for Indian Head till and Regina clay. He found no significant difference in grain size between water samples and samples prepared with NaCl brine. Barbour and Fredlund (1989) and Maio (1996) have shown that on exposure to concentrated salt solutions, volume changes of saturated clay soils also occurred from fluid flow out of the clay in response to chemical concentration gradients. Barbour and Yang (1993) experimented on the effect of brine on clay. They concluded that an increase in salt pore fluid concentration results in a decrease in the liquid limit, little change or a slight increase in the plastic limit, and consequently a decrease in the plastic index or the plasticity of the soil. Ikejir and Tanimoto (1993) showed that the effect of cation was very little on tan φ. Clays show depending on their mineralogy, considerable changes in properties when exposed to salt solutions. Torrance and Ohtsubo (1995) reported the decrease of the liquid limit of the Na- saturated Ariake clay as the salinity of the pore water decreased. The liquid limit of a Na-saturated sample for Bangkok-clay increased with increasing NaCl concentration and when substituting Ca for Na in a similar manner to Ariake clay, but the extent of the liquid limit change by such treatments was much smaller for Bangkok clay than for Ariake clay (Ohtsubo et al, 2003). Using four different clays and three chloride salts (NaCl, KCl and CaCl2), Schmitz and van Paasen (2003) reported reduction in the liquid limit as salt molarity Proceedings of the IASTED International Conference September 3 - 5, 2012 Botswana, Africa Modelling and Simulation (AfricaMS 2012) DOI: 10.2316/P.2012.761-006 126