During the process of aquifer thermal energy storage the in situ temperature of the groundwater- sediment system may fluctuate significantly. As a result the groundwater characteristics can be considerably affected by a variety of chemical, biogeochemical and microbiological reactions. The interplay of these reactions may have a negative influence on the operational performance of ATES-systems. The objective of this thesis was to investigate bacterial clogging processes and the biogeochemical aspects of carbonate and iron precipitation in aquifer material obtained from ATES test facilities.
In the General Introduction presented in Chapter 1 the principle of aquifer thermal energy storage is briefly outlined first. This is followed by more detailed information on the thermodynamics of biogeochemical reactions and the conceivable changes in groundwater characteristics during aquifer thermal energy storage. Chapter 2 reports on some microbiological aspects of well clogging during aquifer thermal energy storage. In column experiments well clogging was simulated using aquifer material from a heat storage site and synthetic groundwater. The well clogging potential of oxic and anoxic column effluents was studied at 10° and 30°C using a hollow fiber membrane from which slime depositions were recovered. Only under oxic conditions a slight increase in slime deposition was observed after a temperature rise from 10° to 30°C. No significant difference in bacterial plate counts was measured in oxic and anoxic column effluents, despite the increase in dissolved organic material concentrations at elevated temperatures. This organic material was mobilized from the soil particular organic carbon fraction. The biologically available organic carbon concentration was less than 1% of the dissolved organic carbon concentration, which was not enough to allow excessive bacterial growth or slime formation. Chapter 3 reports on the thermal mobilization of dissolved organic carbon and the concomitant carbon dioxide production in aquifer material from a heat storage site. These processes have been quantified aerobically and anaerobically within a temperature range of 4° to 95°C in sediment samples containing either quartz-rich coarse sand or peaty clay. At temperatures above 450C dissolved organic carbon compounds, including fulvic acids, were mobilized from both sediments resulting in a substantial increase in the chemical oxygen demand of the water phase. Complexation of calcium and magnesium by fulvic acids resulted in the super -saturation of the water phase with regard to calcite and dolomite and thus prevented the precipitation of these carbonates. The highest rates of carbon dioxide release were measured during the first four days of incubation. Aerobically, the maximum rate Of C0 2 production varied between 35 and 800 (sand) or 15 and 150 (peaty clay) μmol C0 2 per gram volatile solids per day. Anaerobically, the rates were 25 and 500 (sand) or 10 and 110 (peaty clay) μmol C0 2 per gram volatile solids per day. At temperatures above 55°C, C0 2 was produced purely chemically. Chapter 4 deals with ferric iron precipitation in anaerobic Tris-HCl buffered seawater. In these incubations,40 mM lactate was rapidly dissimilated to acetate by sulphate reducing bacteria after a lag period of three days. In presence of added nitrate or ferric iron (both 1 mM) or a combination of both, the initial lactate consumption rate was slowed down and sulphate reduction started after four days at a similar rate as was observed in the absence of nitrate and ferric iron. Nitrate in combination with ferrous iron totally inhibited sulphate reduction. Some lactate was initially oxidized, but its concentration did not change after day six of incubation. In these incubations ferrous iron was oxidized chemically to ferric iron with a concomitant reduction of nitrite to nitric oxide. In this so-called chemodenitrification process, nitrite was formed biologically from nitrate with lactate as a reductant. In Chapter 5 chemodenitrification was studied in details with E.coliE4as a model bacterium. Both, L-lactate-driven nitrate and ferric iron reduction were investigated. Ferric iron reduction in E.coliE4was found to be constitutive. Contrary to nitrate, ferric iron could not be used as an electron acceptor for growth. Ferric iron reductase activity of 9 nmol Fe 2+.mg -1protein.min -1could not be inhibited by well known inhibitors of the E.coli respiratory chain. Active cells and the presence of L-lactate were required for ferric iron reduction. The L-lactate-driven nitrate respiration in E.coliE4 leading to the production of nitrite, was reduced to about 20% of its maximum activity with 5 mM ferric iron, or to about 50% in presence of 5 mM ferrous iron. The inhibition was caused by nitric oxide formed by a spontaneous chemical reduction of nitrite by ferrous iron. Nitric oxide was further chemically reduced by ferrous iron to nitrous oxide. With electron paramagnetic resonance spectroscopy, the presence of a free ferro-nitrosyl complex was shown. In presence of ferrous or ferric iron and L-lactate, nitrate was anaerobically converted to nitric oxide and nitrous oxide by the combined action of E.coli E4 and spontaneous chemical reduction reactions. Chapter 6 reports on aerobic reduction of nitrate to ammonium in E.coli grown in continuous cultures, a novel feature of E.coli Nitrate and nitrite was reduced by E.coli E4 in a L-lactate (5 mM) limited chemostat culture at dissolved oxygen concentrations corresponding to 90 - 100% air saturation. Nitrate reductase and nitrite reductase activity was regulated by the growth rate, oxygen and nitrate concentrations. At a low growth rate (0.11 h -1) the measured nitrate and nitrite reductase activities were 200 and 250 nmol.mg -1protein.min -1, respectively. At a high growth rate (0.55 h -1both enzyme activities were considerably lower (25 and 12 nmol.mg -1protein.min -1). The steady state nitrite concentration in the chemostat was controlled by the combined action of the nitrate and nitrite reductase. Both enzyme activities were inversely proportional to the growth rate. The nitrite reductase activity decreased faster with the growth rate than the nitrate reductase. The chemostat biomass concentration of E.coli E4, with ammonium either solely or combined with nitrate as a source of nitrogen, remained constant throughout all growth rates and was not affected by nitrite concentrations. Contrary to batch, E.coli E4 was able to grow on nitrate as the sole source of nitrogen. When cultivated with nitrate as the sole source of nitrogen the chemostat biomass concentration is determined by the combined activities of nitrate and nitrite reductase and hence, inversely proportional to growth rate.