The science associated with the incorporation of intermediate level toxic and radioactive waste into cement systems is an important field to understand for the safe and cost effective storage and disposal of these wastes. A fundamental understanding of the effects of different wastes and, possibly, a prediction of their effects, is necessary to fully utilise the material. Composite cement systems are an attractive alternative to Portland cement for a number of reasons. The use of mineral additions consumes a waste product which would otherwise need a disposal route and, due to the decreased amount of Portland cement used, reduces cost and energy consumption. Strength of the cemented waste is not of primary importance so the lower strength of the composite cements should not pose a problem. However, one of the main advantages of composite cements over Portland cements is that they evolve less heat during hydration, which is a useful property especially for the storage of cements in an underground repository where the ambient temperature could be significantly higher than usual. A Portland cement and three composite cements (3:1 BFS:OPC, 9:1 BFS:OPC and 3:1 PFA:OPC) were studied with the Portland cement acting as a reference material. To these cements five inorganic metal salts were added. These were ZnCl2, SnCl2.2H20, SnCl4.5H20, A1C13.6H20 and As203 and they were added at levels of 0.1%, 1.0% and, in some cases, 5% by weight of metal. To allow the study of the effects of zinc chloride and the two tin chlorides to be carried out in depth, the number of samples to be studied was selectively reduced during the project. The kinetics of early hydration, the formation of non-standard hydration products and the location of the metals within the microstructure were investigated using such techniques as isothermal conduction calorimetry, X-ray diffraction, and scanning electron microscopy with energy dispersive spectroscopy analysis. It was not possible to generalise the effects of the metal salts on the hydration of the cement as the variation between the different cement systems was considerable. Zinc chloride caused a retardation of set and hydration in all cement systems, the severity of which increased with increased loading of metal salt. Tin(II)chloride also caused a retardation of set but this was not as great as in the zinc system. Tin(IV) chloride had mixed effects which were dependent on the amount of salt present in the system. All the additions affected the various cement systems to different extents, with the 9:1 BFS:OPC being the most retarded. The formation of calcium hydroxo stannate, CaSn(OH)6, and calcium zincate, CaZn2(0H)6.2H20, was identified in the appropriate cement systems with the formation of FriedeTs salt, Ca3Al2C>6.CaCl2.10H2O, also apparent alongside the appearance of these phases. The identification of the metals (zinc or tin) within the microstructure revealed that the metal did not locate in grains of the mineral addition but, in all cases, was incorporated in the Portland cement phases