In many areas globally, groundwater serves as the sole source of drinking water in rural and urban communities. However, the increased industrial and agricultural activities have resulted in significant contamination of geoenvironment with severe effects on human life and ecosystems. Heavy metals, such as hexavalent chromium [Cr(VI)], and other inorganics, such as phosphate and nitrate, are typically among those contaminants. Approximately 170,000 tonnes of chromium are released annually to the geoenvironment as a result of anthropogenic activities causing contamination of surface water, groundwater and soils. Nitrate and phosphate are contaminants of major concern on a global scale since they are recognized that controls eutrophication in surface water bodies. Thus their transport and fate in geoenvironment must be well understood to better evaluate their environmental impacts. However, the occurrence of chromium in the geoenvironment can also be related to geogenic origin due to the occurrence of specific geological background like ultramafic rocks and ophiolitic complexes. Thus intensive agricultural activities combined with the presence of ophiolitic complexes can lead to groundwater contamination with hexavalent chromium, and phosphates and nitrates. The Greek geological background is highly consisted of ultramafic rocks and ophiolitic complexes and thus hexavalent chromium is often detected in groundwater. The aim of this study was to identify such an area in Greece and investigate the geochemistry of hexavalent chromium and simulate its adsorption efficiency on such type of soils. The selected area was close to Vergina town in northern Greece and exhibited such geological background. Agricultural activities were the only anthropogenic pressure in the area. The selection of the study area was based on an extensive groundwater monitoring data base created by the Greek Institute of Geology & Mineral Exploration. In addition, a new well was constructed in order to be used exclusively for research needs. Groundwater and soils sampling was performed along 100 m depth. Soil mineralogical analysis showed that the collected samples exhibited the typical ultramafic origin containing “ultramafic minerals” such as chrysotile and chromite as well as their weathering products (vermiculite), mixed with minerals typical of a mafic assemblage (chlorite, quartz, albite, hematite). Physicochemical analysis of soil samples showed that pH increased with depth probably as a result of the more intensive presence of organic matter and nitrification process in the upper soil layers and due to the effect of weathering processes in greater depths. Elemental analysis showed that the tested soil was poorer in iron and aluminum, richer in silicon and about average in magnesium, compared with serpentine soils of other areas worldwide. However, the relative abundance of magnesium versus aluminum strongly indicated the relative contribution of ultramafic versus mafic materials in the soil sample. Regarding the presence of chromium the following results were obtained. Total chromium (Crtot) concentration did not exhibit a uniform trend with depth exhibiting firstly a decrease until the first meters, followed by a slight increase for depths down to 10.5 m and a general increasing trend for depths higher than 43 meters. These results are in accordance with other presented in the literature mentioning that weathering processes that occur usually in the shallow unsaturated zone favor leaching of elements such as magnesium and accumulation of others like iron, aluminum and chromium while in higher depths the presence of unweathered serpentinitic phases is more intensive. Contrary to Crtot, Cr(VI) concentration exhibited an almost continuous decrease with increasing depth. Regarding the presence of chromium in groundwater concentrations of Crtot up to 91 μg/L and hexavalent chromium up to 64 μg/L detected. These values are of the highest reported globally in aquifers with similar geological background. Both Crtot and Cr(VI) concentrations in groundwater decreased almost linearly with depth. A high correlation between Cr(VI) and Crtot observed with the Cr(VI)/Crtot ratio being higher than 83%. In addition, the intense agricultural activities in the tested area resulted in high nitrate concentrations. Taking into account that the main processes that regulate the fate of hexavalent chromium produced by oxidation of trivalent in aquifers, are sorption and reduction the occurrence of these processes in the case of ophiolitic soil were investigated. In ophiolitic soils there is a mixture of hexavalent and trivalent chromium, both naturally occurring. Hexavalent, but not trivalent chromium, can be leached out of the soil and enter groundwater. As hexavalent chromium is leached from the soil, the remaining trivalent can slowly oxidize to hexavalent in order to reestablish the equilibrium of the soil. The leached hexavalent chromium can be adsorbed or reduced by the solid phase. As regarding reduction minerals that contain divalent iron like magnetite or (magnesio)chromite can act as reductants for hexavalent chromium. Sorption is strongly affected by the occurrence of iron (oxy-hydro)oxides, which are the most common sorbents for Cr(VI). In addition, iron (oxy-hydro)oxides can efficiently act as adsorbents and for other inorganic contaminants like phosphates and nitrates. In order to determine the processes responsible for hexavalent chromium removal from groundwater batch experiments were performed investigating the effect of several parameters like pH, mineralogy, soil’s particle size, initial concentration of hexavalent chromium, ionic strength of the solution and the presence of other inorganic contaminants. The results showed that both adsorption and reduction processes affected hexavalent chromium removal from the soil solution. For both processes removal decreased with increasing pH values but their contribution to the total removal depends on pH. Reduction was attributed to the presence of a magnetic fraction in the soil sample which includes magnetite and magnesio-chromite as primary minerals. Regarding adsorption is probably attributed to the presence of amorphous iron oxy-hydroxides in the soil. Adsorption was found to be semi-reversible since only a fraction of the adsorbed amount was found to be desorbed in the soil solution. This is probably a result of the formation of inner sphere complexes which are sufficiently stable, including exclusively ionic and/or covalent bonds. However, both processes are surface-driven with reduction being influenced by adsorption since partitioning of hexavalent chromium onto the solid surface is required before reduction occurs. In addition, evaluation of sorption as a function of particle size showed that the finer fraction of the soil, which exhibited and the higher value of specific surface area, dominated the adsorption behavior of the soil. The effect of initial concentration of hexavalent chromium was tested for higher concentrations than these occur due to geogenic origin. Langmuir and Freundlich isotherms fitted very well the experimental data, indicating thus the simultaneous heterogenity of the surface sites on the serpentinitic soils and possibly the formation of a monolayer for hexavalent chromium adsorption. Finally, significant effects observed due to the alteration of ionic strength. Adsorption efficiency was decreased with increasing ionic strength value suggesting thus the formation of outer sphere complexes for the adsorption of hexavalent chromium. In addition, the adsorption capacity of the ophiolitic soil was tested for the inorganic contaminants phosphates and nitrates that are commonly observed in areas with intense agricultural activities and any possible competitive effects between them and hexavalent chromium. The ophiolitic soil exhibited high adsorption capacity for phosphates with the adsorption process being affected by the pH of the solution. More specifically, the increase of pH decreases the adsorption of phosphates, almost linearly. No competitive effects on phosphates adsorption were observed during the presence of hexavalent chromium. On the other hand hexavalent chromium adsorption was strongly affected by the presence of phosphate in the solution. This is probably due to the formation of exclusively inner sphere complexes between phosphate and the ophiolitic surface, creating competition with chromates which also form inner sphere complexes. Regarding nitrates their adsorption on the ophiolitic soil was very low up and almost zeroed at pH values commonly occur in aquifers. The effect of hexavalent chromium on nitrate adsorption was almost negligible and vice versa. This is probably due to different type of surface complexation of nitrates and chromate, since nitrate form exclusively weak outer sphere complexes contrary to chromate which form both inner and outer sphere complexes. The following step of this thesis was to compare the adsorption behavior of the ophiolitic soil with that of a pure ferric oxide. The selected ferric oxide was goethite since it is considered as the most abundant iron oxide in the geoenvironment and the most common oxide in ophiolitic soils. Goethite was tested for its adsorption capacity for hexavalent chromium, phosphate and nitrate. The possible competitive effects between the tested anions were also investigated. The experimental results showed that goethite is an efficient adsorbent for chromate and phosphate ions but not for nitrates. The pH increase caused decrease of the adsorption capacity of goethite for hexavalent chromium, phosphates and nitrate ions. For high pH values the goethite surface becomes negatively charged and thus repulsions of chromates and the inorganic anions with the surface occur. In addition, the effect of ionic strength on chromate adsorption was investigated as a way of revealing the type of adsorption. The increase of ionic strength resulted in noticeable decrease of the adsorption of hexavalent chromium, suggesting the formation of outer sphere complexes, which are based on weak electrostatic forces. The investigation of any competitive effects between chromate and phosphate for adsorption on goethite surface showed that phosphates adsorption is not affected by the presence of chromates. On the contrary, an important effect on the adsorption of chromates, was observed in the simultaneous presence of phosphates. This is probably attributed to the fact that chromates, during the presence of phosphates, are mainly adsorbed via outer sphere complexes since phosphate are adsorbed only via inner sphere complexation. Regarding any competition between chromate and nitrate the results showed that adsorption of chromates was decreased under the presence of nitrates. This is probably due to the significant difference on their concentrations. The significant higher concentration of nitrate probably creates electrostatic repulsions which may affect the complexation of chromates, especially the formation outer sphere complexes. Regarding nitrates the already low adsorption efficiency was not affected by the presence of chromate in the solution. Taking into account the experimental results obtained from the batch experiments the next aim of this thesis was to simulate the adsorption process applying surface complexation models. The behavior, transport, and generally the fate of heavy metals and inorganic contaminants in the geoenvironment depends largely on their sorption reactions with soil particles and so it is of high importance to investigate such reactions. Different empirical approaches have been used for studying the adsorption behavior of natural soils, but several limitations have been observed since these approaches cannot account for changes in groundwater chemistry. Thus, in the last twenty years extensive studies have been performed using surface complexation models for describing the adsorption of heavy metals and inorganic contaminants with quite promising results. Surface complexation models are generally based on providing a thermodynamic description of the reactions between the surface groups and the adsorbed ions, based on charge and mass balances. However, most of these studies have used pure minerals and especially pure hydrous oxide solid surfaces, and only few studies refer to natural materials and particularly to soils like in the present study. Additionally, since each surface complexation model is based on different specific assumptions regarding the solid–solution interface, which are expressed by different surface complexation reactions and thus different adsorption constants the obtained values cannot be used in different models. In this study, three of the most common surface complexation models the triple layer model, the diffuse-layer model and the constant capacitance model were used for simulation. Databases with the corresponding surface complexation reactions were created and inserted in the Visual Minteq software. The application of the aforementioned models was based on the general composite approach for adsorption simulation. This approach considers that the soil composition is too complex in order to distinguish and quantify the several individual phases and thus it is assumed that a general type of active sites exists on the soil surface. This assumption creates the need for using stoichiometry and formation constants which are obtained by fitting experimental data. Thus the created databases with the surface complexation reactions can be used only by altering the adsorption constants in order to fit the experimental data. Among the parameters required in each model a general parameter required for all models in order to perform simulation is the concentration of the solid in the solution. Despite in the case of goethite the solid concentration used in the batch experiments was applied for adsorption simulation this was the scenario in the case of using ophiolitic soil. Taking into account that not all the minerals contained in the ophiolitic soil contribute to contaminants adsorption it was assumed that adsorption is mainly controlled by the presence of iron and maybe of the aluminum oxides. Their concentration was calculated using the mass balance as determined by X-Ray fluorescence and via quantitative X-Ray diffraction analysis. Three different scenarios taking into account the contribution of either a) iron and aluminum oxides, b) only iron oxides and c) only the amorphous iron oxides were used for adsorption simulation using each of the three aforementioned surface complexation models. The application of triple layer model led to satisfactory description of the adsorption process for all the anions tested. The adsorption simulation was effective for any parameters tested like effect of pH, ionic strength and any competitive effects when assuming that adsorption is controlled by either the presence of iron and aluminum oxides or only by the iron oxides. Simulation was not effective when using the concentration of only amorphous iron oxides in the soil sample indicating thus that this solid phase does not correspond to the concentration of the solid phase that contributes to adsorption. Simulating the effect of ionic strength it was revealed that at low ionic strength values adsorption of chromates can be described by the formation of inner sphere, either monodentate or bidentate, and outer sphere complexes. However, the increase of ionic strength showed that adsorption chromate can be efficiently described only by the formation of bidentate complexes. The formation of bidentate complexes of chromates with the soil surface can also describe the competitive effects under the presence of phosphates and nitrates. In addition, satisfactory simulation was achieved for phosphate and nitrate adsorption and their competition with chromates in the soil solution. Thus, Triple Layer Model can in general describe efficiently the adsorption of chromates and the inorganic contaminants on the ophiolitic soil. On the other hand, simulation of experimental data using the diffuse layer model showed that the application of this model is not suitable for describing the adsorption process on the ophiolitic soil. The Diffuse Layer Model was not capable to simulate neither the adsorption of chromates nor of the other inorganic contaminants. A possible explanation is the assumption of monodentate complexes of chromates and phosphates by applying the diffuse layer model and the absence of any outer sphere complexes in the case of chromates and nitrates. Similarly, the constant capacitance model was not capable of simulating the adsorption of neither chromate nor the inorganic contaminants on the ophiolitic soil. Finally, the application of surface complexation models for describing the adsorption behavior of goethite leads to similar observations as in the case of the ophiolitic soil. The triple layer model described satisfactorily the adsorption of hexavalent chromium and inorganic contaminants, as well as their competition on the goethite surface. Simulation showed that chromates are adsorbed via inner and outer sphere complexation. This fact was verified by the decrease of the adsorption efficiency due to ionic strength increase. Chromates complexation is considered to be governed by the formation of bidentate complexes with the solid surface. The adsorption of phosphate and nitrate was also satisfactorily simulated. The competitive effects occurred between chromates and each inorganic contaminant, were also simulated with high accuracy. On the contrary, simulation of the adsorption of chromates and inorganic contaminants using the diffuse layer model was not effective, as in the case of the ophiolitic soil. As mentioned above, a possible explanation is the assumption of formation exclusively inner sphere monodentate complexes by applying the diffuse layer model and the absence of any outer sphere complexes. In closing, the application of constant capacitance model for simulating adsorption of chromate and phosphate on single anion solutions could efficiently describe the adsorption of chromate and phosphate but not adsorption of nitrate. However, it could not describe the competitive effects created between chromate and inorganic contaminants.
