Performance of Clay Liners in Near-Surface Repositories in Desert Climate

Sammanfattning: Wars in Iraq (1991 and 2003) generated various types of hazardous waste (HW) in the form of soil contaminated by depleted uranium (DU). Other HW emanated from destroyed army vehicles and remnants of Iraqi nuclear facilities holding various types and amounts of chemical and radioactive material. The negative impact of the various wastes on the health conditions of the population was reported from different parts of Iraq, showing an enhanced frequency of cancer and abnormally born infants. For isolating the wastes, which represent low-level and short-lived intermediate level radioactive wastes, near-surface repositories (NSR) are proposed since they represent the least expensive way of solving future problems with sufficient safety. Internationally, the timeframe of the containment of such wastes is designated to be 300 years. Site selection affects and largely controls the selection of a suitable design the aim being to minimize or eliminate migration of hazardous elements from the waste to the environment. The formulation of siting criteria is the first vital step toward the resolution of the problem. Site selection criteria are proposed taking in account three major factors: environmental, geological and socio-economic factors. Accordingly, Iraqi deserts, which make up 60% of Iraq, represent the number one candidate for locating a safe disposal facility, primarily because of the low population, suitable topography, climatic conditions, seismic stability and availability of raw materials. Long-term performance of NSR is directly related to the function of top and bottom liner systems. They should be designed so that they are mutually compatible and combine to effectively isolate the waste. Liners are considered as the main elements of any disposal facility on the ground surface and a properly designed top liner system is of particular importance since it will minimize or eliminate water percolation into the waste body. Compacted clay liners (CCL) should preferably have with a low hydraulic conductivity, whichis achievable by proper selection of raw materials, compaction density and construction methods. A further criterion is that they must not soften significantly by expansion on wetting, which puts a limit to the smectite content and density. The liners can consist of native material found near the landfill site, and be used after simple processing, primarily drying and crushing, or be mixed with fillers like silty sand. Since the hydraulic conductivity is the key property of a reliable CCL, relevant experimental determination of the hydraulic conductivity is vital. The common practice in geotechnical laboratories is to apply high hydraulic gradients for getting results quickly but this can lead to non-conservative, incorrect results. The present study involved determination of the hydraulic conductivity of a smectite-rich clay sampled at places within reasonable distance from potential NSR sites. Various hydraulic gradients were applied to samples compacted to several different densities, using two permeants and two filter types. It was concluded that the outflow filter can significantly affect the evaluated conductivity especially when applying high hydraulic gradients. This was partly explained by clogging of outflow filters of conventional fine-porous type by torn-off clay particles at such gradients. A major conclusion was that the gradient in laboratory testing should not exceed 100 m/m. In order to assess the suitability of available raw materials within the Iraqi Deserts, two smectitic soils termed as Green and Red clays were investigated for potential use in CCLs. Both clays are fairly rich in smectite, which calls for mixing them with properly graded silt/sand material from the desert for modifying the expandability. The shear strength, swelling pressure, hydraulic conductivity and creep properties were determined and used for defining criteria for selecting suitable clay-sand ratios. The results showed that 30-50% Green clay mixed with sand and 40-60% Red clay mixed with sand were suitable for constructing top liners with a hydraulic conductivity of 1×10-9 - 1×10-10 m/s. For bottom liners, 70% Green clay mixed with sand and 80% Red clay mixed with sand would be suitable; they were found to have a hydraulic conductivity of 1×10-11 m/s. The long-term performance of CCL is controlled by a number of processes like long periods of extreme dryness and short periods of very heavy rain. The percolation of water through the top liner system of a number of design alternatives were simulated using the code HELP 3.95D and subsequently by the FE program VADOSE/W. For the assumed NSR concept the slope stability of the top liner is essential and it was determined by using FE technique considering various slope angles. The engineering properties, primarily the hydraulic conductivity, swelling pressure and shear strength of 30-50% Green clay mixed with sand were introduced in the simulations. Two initial water contents of the compacted materials were considered representing 1) optimum water content (“wet case”), and 2) air‐dry conditions (“dry case”). Application of the HELP code decided the selection of suitable CCL having a thickness of 0.5 m and inclined by 5.7ᵒ. More detailed analyses with VADOSE/W showed that a mixture at the dry case would bring 0.5 mm (0.5 litre of leaking water per square meter) through CCL in an eight year simulation period. Long-term simulations (up to 300 years) showed that CCL would undergo continuous drying without reaching saturation even in the case of periods of very heavy rain (616 mm) for the wet and dry cases. The slope stability factor for the rather steep angle 30ᵒ was found to be 1.5 for the most critical case representing complete water saturation. In conclusion, the proposed materials and design features are believed to be suitable for practical application.