The chemical resistance of glass is an important property for many applications. This property has been extensively studied for many types of glass under static conditions (no liquid is removed during the experiment). There has been little research conducted on the effects of additional stimuli on the dissolution of glass. For this research lithium disilicate was leached in deionized water at multiple temperatures while microwave radiation, ultrasonication or flow conditions were also applied to the system. These results were then compared to static baseline to determine if these stimuli would cause any change to the mechanisms and kinetics of the reaction. It was determined that for the experimental conditions used there was little to no change in dissolution when 2.45 GHz microwave radiation instead of conventional methods was used to heat the reaction. Results from ultrasonication found that samples that experienced erosion showed an increase in dissolution with an increase in dissolution following heavier erosion. This was thought to be due to both an increase in the surface area of the sample to volume of solution (SA/V) ratio (erosion would modify the surface area and release small particulates) and the accelerated removal of the depleted layer due to erosion. Stereoscopic reconstruction was used to semi-quantitatively measure the change in surface area. Regions that experienced minor erosion showed a 3-6% increase in surface area while those that experienced heavy erosion showed a 29-35% increase in surface area. Due to inconsistencies in the size of the eroded area it was not possible to determine the effects of power intensity with this research. Flow dissolution showed similar trends in concentration and different trends for the total normalized mass loss (TNL) to previously published research on more complex glasses. The elemental concentration initially increased before reaching a peak and decreasing to steady state. This peak was thought to be caused by the combination of flow, increasing thickness in the depleted layer, and an initial fluctuation in the forward reaction rate due to changes in pH. For the lithium disilicate glass used in this research both the elemental concentration and the TNL increased with increasing temperature and decreasing flow rate (silica dissolution was an exception as it did not show any change in TNL due to flow). All experimental conditions were shown to achieve steady state (dC/dt~0) by the seventh day of leaching. The contrast in the observed TNL trends between lithium disilicate and more complex glasses was thought to be due to differences in reaction rates and the presence of an additional surface layer in the complex glasses due to precipitation. Microscopy of the leached glass showed that surface features introduced during grinding (scratch lines and microcracks) were preferentially leached and grew in size and number visible during dissolution. A semi-quantitative model was created using stereoscopic reconstruction to describe the preferential leaching of the microcracks as there was little available discussion found in literature outside of associating the growth of these features with localized network dissolution. In this model the microcracks experience preferential dissolution leading to a change in size and shape. The SA/V ratio inside the crack would be much larger than the bulk system (calculated to initially be ~768,000cm-1 compared to the bulk's 0.1cm-1). This would cause massive acceleration in the initial ion exchange, raising the pH of the solution which would in turn cause network dissolution to occur much faster inside the crack. Based on static experiments on lithium disilicate frit (SA/V of 1,010cm-1) the pH inside the crack would jump to above 11 in minutes. As the crack grows, the SA/V ratio inside it would decrease (largest cracks were found to have a ratio ~100,000cm-1). The accelerated leaching caused by these features could have a noticeable effect on the dissolution results. In addition to the accelerated leaching inside a crack, the size of the depleted layer under the crack would be different from the bulk glass. / Doctor of Philosophy / The chemical resistance of glass is an important property for many applications. This property has been extensively studied for many types of glass under static conditions where no liquid was removed and temperature was the major variable. For this research lithium disilicate was leached in deionized water at multiple temperatures while the additional stimuli of microwave radiation, ultrasonication or flow conditions were also applied to the system. The question that this research addressed was how does the aqueous dissolution of glass change when a system is exposed to these additional stimuli? Although glasses are subjected to these stimuli in many everyday applications, their influence on dissolution has not been studied extensively. Lithium disilicate glass was selected because it contains components used in many commercial glasses, has sufficient reactivity in water to allow experiments to be completed in a reasonable time, and because its mechanisms for dissolution under static conditions were well known. Glass is frequently selected to be the container when microwaves are used to heat food or materials. Flow is an important part of many applications involving glass including the storage of nuclear waste glass, glass-lined tanks used in the chemical industries, in the use of glass in the human body (bioglass and dental crowns), and in typical window and laboratory glasses where intermittent aqueous contact and runoff may occur. Examining how cavitation via ultrasonication can be controlled to either minimize or maximize element extraction is important, with the removal of rare earth elements from fly ash being one example.

Multiphase flow and transport in fractured rock is of importance to many practical and engineering applications. In the field of groundwater hydrology an issue of significant environmental concern is the release of dense non-aqueous phase liquids (DNAPLs) which can cause long-term groundwater contamination in fractured aquifers. This study deals with two fundamental processes – fluid displacement and interphase mass transfer – concerning the behavior of the multiphase contaminants in fractured media. The focus of this work has been placed on improving the current understanding of small-scale (single fracture) physics by a combined effort of numerical modeling analysis, laboratory experiments and model development. This thesis contributes to the improved understanding through several aspects. Firstly, the effect of aperture variability, as characterized by geostatistical parameters such as standard deviation and correlation length, on the DNAPL entrapment, dissolution and source-depletion behaviors in single fractures was revealed. Secondly, a novel, generalized approach (adaptive circle fitting approach) to account for the effect of in-plane curvature of fluid-fluid interfaces on immiscible fluid displacement was developed; the new approach has demonstrated good performance when applied to simulate previously published experimental data. Thirdly, the performance of a continuum-based two-phase flow model and an invasion percolation model was compared for modeling fluid displacement in a variable-aperture fracture and the dependence of fracture-scale capillary pressure – saturation relationships on aperture variability was studied. Lastly, through experimental studies and mechanistic numerical modeling of DNAPL dissolution, kinetic mass transfer characteristics of two different entrapment configurations (residual blobs and dead-end pools) were investigated. The obtained understanding from this thesis will be useful for predictive modeling of multiphase contaminant behavior at a larger (fracture network) scale. / Flerfasflöde och ämnestransport i sprickigt berg är av betydelse för många praktiska och tekniska problem. Tunga, svårlösliga organiska vätskor (engelska: dense non-aqueous phase liquids: DNAPLs; t.ex. klorerade lösningsmedel) kan orsaka långvarig förorening av vattenresurser, inklusive akviferer i sprickigt berg, och utgör ett viktigt miljöproblem inom grundvattenhydrologin. Denna studie behandlar två fundamentala processer för spridning av flerfasföroreningar i sprickiga medier – utbredning av den organiska vätskan och massöverföring mellan organisk vätska och vatten. Arbetet har fokuserat på att förbättra nuvarande kunskap om de fysikaliska processerna på liten skala (enskilda sprickor) genom en kombination av numerisk modellering, laboratorieexperiment och modellutveckling. Avhandlingen har bidragit till utökad processförståelse i flera avseenden. För det första har arbetet belyst effekterna av sprickaperturens variabilitet, uttryckt med geostatistiska parametrar som standardavvikelse och rumslig korrelationslängd, på fastläggning och lösning av organiska vätskor i enskilda sprickor, samt utmattningsbeteendet hos dessa källor till grundvattenförorening. För det andra har en ny, generell metod (adaptiva cirkelpassningsmetoden) för att ta hänsyn till effekten av krökningen av gränsytan mellan organisk vätska och vatten i sprickplanet utvecklats; denna metod har visats fungera väl i simuleringar av tidigare publicerade experimentella data. För det tredje, har en jämförelse gjorts mellan en kontinuumbaserad tvåfasflödesmodell och en invasions-perkolationsmodell med avseende på hur väl de kan simulera tvåfasflöde i en spricka med varierande apertur. Här studerades även hur relationen mellan kapillärtryck och mättnadsgrad på sprickplansskala beror av variabiliteten i sprickapertur. Till sist undersöktes lösning av den organiska vätskan i grundvatten för två fastläggningsscenarier (fastläggning i immobila droppar och ansamling i fällor – ”återvändssprickor”) både genom experiment och mekanistisk numerisk modellering. Kunskapen som tagits fram i denna avhandling bedöms vara användbar även för att modellera spridningen av flerfasföroreningar på större (spricknätverks-) skalor.