Waste materials often contain increased levels of potentially toxic trace elements compared to natural materials such as soils. In many countries, the recycling of waste materials in the environment, such as in construction works, is regulated by environmental criteria that aim to ensure long-term environmental protection. These criteria are increasingly based on the potential "leaching" of contaminants, i.e. the release of contaminants from the solid phase to the water phase with which the material may be in contact (e.g., percolating rainwater). The extent to which contaminants are susceptible for leaching processes depends on many chemical and physical factors, such as the specific chemical form of the contaminant ("chemical speciation") and transport processes such as convection and diffusion. To better understand the environmental risks associated with the application of waste materials in the environment, it is important to gain a fundamental understanding of the underlying speciation and transport processes that control the leaching of contaminants, as well as the fate of these contaminants in soil and groundwater.The complexity of speciation in combination with transport processes (referred to as "reactive transport") make that the identification of the major controlling processes responsible for observed leaching phenomena is generally not straightforward. However, hypotheses with respect to possible involved processes can often be translated in (computer) models that simulate these processes. The verification of predictions made by such models against experimental data may lead to either confirmation or rejection of the underlying hypotheses. The latter may result in modification and/or expansion of the model, until the system is sufficiently understood and adequate model predictions are obtained. Used in this way, models form valuable instruments in the scientific process of gaining knowledge, and contribute to the identification of the dominant processes that control the behaviour of contaminants in the situation under study. Since processes on a molecular scale have a general validity, models of that are based on these processes ("mechanistic" models) are more suitable for these purposes and of a wider applicability than an empirical models. Once a model, based on gained fundamental insights in leaching processes, sufficiently describes observed leaching phenomena under a wide range of conditions, it may be used for different practical purposes. Among these are the quality improvement of (recycled) waste materials with respect to their leaching properties and the development of realistic regulatory limits for the safe application of waste materials in the environment.The aim of this thesis is to develop a generally applicable, mechanistic geochemical modelling approach with which dynamic leaching and reactive transport processes in "contaminated materials" can be predicted. The term "contaminated materials" ultimately refers to any natural or waste material that may potentially release contaminants by leaching. In this thesis, Municipal Solid Waste Incinerator (MSWI) bottom ash, the major residue that remains from the incineration of municipal solid waste, and contaminated soils are studied as relevant and representative cases of such materials. This thesis focuses on the leaching of inorganic contaminants, although the principles of the approach do also apply to organic contaminants that fall beyond the scope of this work.The approach consists of a number of successive steps. The first step consists of the identification of the (major) processes that control the leaching of contaminants from the material under study. In this step, geochemical model calculations are performed and compared to experimental data in order to verify hypotheses on the underlying leaching processes. As leached concentrations of elements generally vary by orders of magnitude as a function of pH, data generated by batch pH-static experiments (leaching experiments conducted at constant pH values) over a wide range of pH (e.g., pH 2 to pH 12) provide a sensitive verification of such hypotheses. Depending on the outcome of the verification the model may need to be modified and/or further expanded with additional relevant processes, until the model calculations provide an adequate representation of the measurements. In the next step, the resulting geochemical model is coupled with a model for transport processes. Based on the processes identified in the previous step, the time- dependent leaching of contaminants in dynamic (reactive transport) systems is predicted and verified against experimental data from column experiments. Similar to the previous step, the verification of model predictions may lead to modification and/or further expansion of the reactive transport model.The general applicability and (long- term) predictive value of models strongly depends on the way the model is parameterized, i.e. with respect to the used thermodynamic parameters and estimates of material- specific properties/input parameters (e.g., the actual amounts of "reactive surfaces" to which contaminants can bind, such as iron (hydr)oxide minerals and natural organic matter). Therefore, the approach described in this thesis aims for consistency between the hypothesized processes, the selection of (sorption) models to simulate these processes, necessary model input parameters, and experimental methods to determine these parameters. Wherever sorption to a reactive surface is suspected to be an important process, mechanistic sorption models are selected, with a preference for models for which "generic" parameter sets have been derived. The selected models and parameter sets are applied without modification (i.e. without parameter fitting). Whenever sorption models are taken into account, information is needed on the amount of the specific surfaces present in the sample under study. This information is collected using independent, carefully selected experimental procedures that aim to estimate the concentrations of the specific type of reactive surface of interest. Examples of processes that are important for contaminant mobility in soil and waste systems and are treated in the above described way include the adsorption of ions to iron (hydr)oxides and natural organic matter.The successive steps of the modelling approach as outlined above are reflected in the different chapters. The point of departure is described in chapter 2 . For the case of weathered MSWI bottom ash, the at that time available knowledge of processes that control the leaching of contaminants in batch systems is used to predict experimental leaching data obtained from (dynamic) column experiments. This evaluation leads to the identification of potentially important processes on the basis of which the modelling approach can be further improved in the forthcoming chapters. The first step was to use batch pH-static leaching data to verify that heavy metal concentrations as a function of pH could be described adequately with a surface complexation model for sorption to hydrous ferric oxide (HFO) and amorphous Al- (hydr)oxides. The next step was to perform reactive transport simulations, which led to predictions of leached heavy metal concentrations generally within one order of magnitude. Non-equilibrium leaching processes were inferred from the generally more abrupt changes of the reactive transport model predictions compared to the measurements. It was concluded that processes that have to be taken into account for further model development are the influence of non-equilibrium effects and the facilitated transport of heavy metals by dissolved organic matter.In chapter 3 , it is investigated to what extent the batch modelling approach followed in the preceding chapter is also applicable to identify and describe the processes that control the pH-dependent leaching of the metals Ni, Cu, Zn, Cd and Pb from contaminated soils. As soils contain relatively high amounts of natural organic matter compared to MSWI bottom ash, the modelling approach used in the preceding chapter is extended with a model for the adsorption of ions to dissolved and particulate organic matter. The approach is also extended with a model for the non-specific sorption to clay surfaces. The resulting model predictions of heavy metal leaching appeared generally adequate, and sometimes excellent. Results from speciation calculations were consistent with the well-recognized importance of organic matter as the dominant reactive solid phase in soils. Further modelling challenges are to include a model for the pH-dependent leaching of DOC from soils as well as to predict soil pH and buffering processes.In chapter 4 , the leaching of a wide range of major and trace elements from MSWI bottom ash is studied as a function of equilibration time, over a wide range of pH under pH-controlled conditions. Based on recent insights and assumptions on the composition of dissolved organic carbon (DOC) in MSWI bottom ash leachates, a similar "multi-surface" geochemical modelling approach as developed in the preceding chapter for contaminated soils is used to improve the interpretation of MSWI bottom ash leaching test results and to investigate whether "equilibrium" is attained during the time scale of the batch pH-static leaching experiments. Depending on the element of interest and setpoint-pH value, net concentration increases or decreases as a function of equilibration time were observed up to one order of magnitude. In addition, different concentration-time trends (increase or decrease) are observed in different pH ranges. Although the majority of the elements do not reach steady state, leached concentrations over a wide pH range have been shown to closely approach "equilibrium" model curves within an equilibration time of 168 hours. The major result from this chapter is that the different effects that leaching kinetics may have on the pH dependent leaching patterns have been identified for a wide range of elements, and can generally be explained in a mechanistic way.Chapter 5aims to provide a mechanistic insight into the beneficial effects of accelerated aging of MSWI bottom ash using the "multi-surface" geochemical modelling approach developed in the preceding chapters on the leaching of copper and molybdenum. Experimental observations and model calculations in literature and in the previous chapters have shown that the leaching of DOC is likely to be the key process responsible for the generally observed enhanced leaching of copper and possibly other metals. Therefore, a novel experimental method is used to characterize DOC quantitatively in terms of humic, fulvic and hydrophilic acids over a wide pH range in order to identify the processes controlling the solid/liquid partitioning of these reactive ligands and their role in the effects of aging on contaminant leaching. Based on the experimental and model results, a new approach is developed to model the pH-dependent leaching of fulvic acids from MSWI bottom ash. The results of this chapter show that accelerated aging results in enhanced adsorption of FA to (neoformed) iron/aluminium (hydr)oxides, leading to a significant decrease in the leaching of FA and associated Cu. Accelerated aging was also found to reduce the leaching of Mo, which is also attributed to enhanced adsorption to (neoformed) iron/aluminium (hydr)oxides. These findings provide important new insights that may help to improve accelerated aging technology of MSWI bottom ash.In the final chapter 6 , the insights and model developments of the preceding chapters are combined into a novel predictive modelling approach in which the leaching of a broad range of major and trace elements from MSWI bottom ash is predicted simultaneously, based on a single set of model input parameters. The approach is applicable to both batch and dynamic systems, as verified experimentally with data from pH-static and dynamic (column) experiments. To address the possible influence of non-equilibrium processes, the column experiments are operated at different flow velocities and with flow interruption periods. The generally adequate agreement between the model predictions and measurements for MSWI bottom ash shows that the use of equilibrium-based reactive transport models to predict data from dynamic laboratory leaching tests is promising. This finding is supported by the generally low sensitivity of leached concentrations to flow velocity and flow interruptions. The experimental and modelling results indicate physical non-equilibrium processes for non-reactive soluble salts and possible sorption-related non-equilibrium processes for the leaching of molybdenum, FA and associated trace metals.The reactive transport modelling approach, as presented in the final chapter, leads to strongly improved model predictions and understanding compared to previous reactive transport modelling studies performed on MSWI bottom ash in literature so far. Novel aspects of the modelling approach outlined in this final chapter compared to that of the initial chapter 2 include the characterization of DOC in terms of its reactive components HA and FA as a function of L/S (liquid-to-solid ratio) and pH, the inclusion of mechanistic models that predict the binding of metals to these substances, the inclusion of a surface complexation model that predicts FA concentrations, and the combination of these geochemical models with non-equilibrium processes. Further improvement of the modelling approach can be achieved by a more mechanistic description of the (dynamic) leaching behaviour of humic substances. In addition, this chapter makes clear that further research is necessary to develop a generic approach for the estimation of the "availability" of components in different types of contaminated materials.It is shown that the consistent geochemical modelling approach as developed in this thesis allows the identification and prediction of contaminant leaching and reactive transport processes from contaminated materials of very different origin and with different physical and chemical characteristics. This is demonstrated with a number of fresh and weathered MSWI bottom ash samples from different Dutch MSW incineration plants (chapters 2, 4, 5 and 6), and soils from various locations with different contamination histories (chapter 3). Although the prospect for a wide applicability is promising, the major challenge for the future is the verification of the approach against data from (long-term) field applications.