Chloroperoxidase (CPO) from Caldariomyces fumago is a versatile heme enzyme with great potential for environmental and pharmaceutical applications. It catalyzes a plethora of reactions including halogenation, dismutation, epoxidation, and oxidation. The diverse catalytic capabilities of CPO have long been attributed to the protein’s distinct active site that combines structural features of peroxidases and cytochromes P450. Particularly, the role of the axial thiolate ligand in CPO catalysis has been much debated. Furthermore, no data are available on the role of hydrogen bonding between Arg 26-Asn 37 and Ala 27-Asn 33 of the proximal helix in defining the structural and catalytic properties of CPO. In order to investigate the influence of the proximal thiolate and the proximal hydrogen bond network on the structural and biochemical properties of CPO, several mutant CPOs were constructed and characterized using various spectroscopic techniques and enzymatic assays. Cysteine 29, which coordinates to the heme, was replaced with a His (C29H) to mimic the proximal ligation of classical peroxidases. The UV-Vis spectrum of the carbon monoxide complex of ferrous C29H mutant remained essentially identical to that of wild type (WT) CPO and P450 although the ferric state of the variant enzyme showed a spectral pattern reminiscent of a classical histidine ligated heme peroxidase. Histidine ligation was further confirmed by paramagnetic NMR spectroscopy. Contrary to a previous report, the specific chlorination activity of C29H was essentially abolished (less than 1% of that of WT CPO) but the epoxidation and peroxidation activities were enhanced 10-fold and 55-fold, respectively. These findings demonstrate for the first time that the heme ligand, Cys 29 in CPO, is not a prerequisite for CPO’s unique P450-like spectroscopic signatures but is constitutive for the protein’s versatile catalytic activities. Arginine 26 and Asparagine 33 in the proximal heme pocket were replaced with Ala (R26A, N33A, and R26A/N33A) to disrupt hydrogen bonding. Tertiary structures and heme environments of R26A, N33A, and R26A/N33A differed from those of WT CPO as determined by CD spectroscopy. The specific chlorination and dismutation activities of all mutants were almost abolished but the peroxidation and epoxidation rates were increased. These results show that the proximal hydrogen bond network plays an important role in maintaining the structure and catalytic diversity of CPO.

Chloroperoxidase (CPO) is one of the most versatile heme enzyme isolated from the marine fungus, Caldariomyces fumago. Functionally, CPO can catalyze four types of reactions: peroxidation (peroxidase-like), dismutation (catalase-like), halogenation (halogenase-like), and peroxygenation (P450-like). Structurally, CPO has a distal and proximal pockets that can be best described as a hybrid of classical peroxidase and P450s. As a heme-thiolate protein, CPO contains the conserved proximal Pro28-Cys29-Pro30 stretch found in other members of the family. However, the structural and functional roles of these proline residues remain poorly understood. Site-directed mutagenesis was undertaken to generates three CPO mutants, P28A-, P30A-, P28A/P30A-CPO. The replacement of the rigid proline with a more flexible alanine residue, freed up the back bone amide for the formation of additional amide-sulfur hydrogen bond, allowing the investigation of the importance of these residues in CPO catalysis. The three CPO mutants displayed dramatic difference in ligand binding affinity and catalytic activities relative to WT-CPO. Any mutations on the proline resides within the proximal loop eliminated the halogenation and dismutation activities but enhanced the vii epoxidation and peroxidation activities by 4-14 fold. As the binding affinity for cyanide, the CPO mutants displayed significantly higher dissociation constant relative to WT-CPO. Our results revealed that Pro28 and Pro30 play important roles in maintaining the versatility of CPO. As a versatile enzyme, CPO has great application potential in pharmaceutical and chemical industry due to its ability to catalyze the formation of chiral epoxides. Phe103 and Phe186 located on the distal pocket have been proposed to guard the access of substrates to the ferryl oxygen of the heme center. The interactions of these two phenylalanine residues restricted the size of substrates and regulates CPO’s enantioselectivity. F186A- and F103A/F186A-CPO were generated and characterized where the rate of peroxidation and epoxidation were significantly enhanced at the expense of halogenation and dismutation activities. Our results demonstrated that Phe186 played a subtler role relative to Phe103 in terms of substrate specificity and product enantioselectivity of CPO.

As propriedades físicas incomuns exibidas por fluidos confinados em meios porosos desempenham um papel importante em diversos processos químicos, geoquímicos e ambientais. Atualmente, muitos aspectos da estrutura e dinâmica dos fluidos espacialmente confinados ainda são pouco compreendidos. Nesse contexto, fenômenos interfaciais influenciam consideravelmente os processos que ocorrem em meios nanoporosos, podendo resultar em efeitos relevantes para o desenvolvimento dos dispositivos nanofluidicos. Esses sistemas multifásicos e com fenômenos multifísicos podem apresentar propriedades eletrônicos e dinâmicos envolvendo diferentes escalas de tamanho e tempo na interface sólido/fluido. Atualmente, uma única metodologia não é capaz de resolver toda a complexidade encontrada em tais sistemas pelo fato de cada qual estar restrita a uma escala ou demanda computacional específica. Além disso, as metodologias habitualmente aplicadas para investigar as fases bulk através da modelagem computacional, em geral, não são adequadas para acessar sistematicamente os efeitos de superfície que ocorrem na interface sólido/fluido. Os desafios impostos à modelagem molecular pelos sistemas nanofluídicos requerem iniciativas inovadoras (dentre as metodologias disponíveis) para acessar as propriedades de interface. Nessa tese, desenvolvemos e aplicamos novas abordagens computacionais em nível atômico a fim de modelar e caracterizar sistemas nanofluidicos. Nesse contexto, introduzimos um método multinível hierárquico top-down, que combina simulações de dinâmica molecular com cálculos ab initio de transporte eletrônico, para abordar fenômenos de multiescala. O potencial dessa implementação foi demonstrado em um estudo de caso envolvendo o fluxo de água e o transporte de íons através de um nanotubo de carbono tipo (6,6). Mostramos que o traço iônica pode representar uma mudança na condutância elétrica do nanocanal, e levar a uma medida indireta da corrente iônica. Também implementamos uma versão modificada da análise de rede de ligações de hidrogênio baseada em teoria de grafos, a fim de fazer o estudo das propriedades estruturais e dinâmicas em diferentes regiões do poro. Com essa abordagem, nós fomos capazes de explorar sistematicamente os efeitos de interface em fluidos espacialmente confinados. Combinando-se simulações de dinâmica molecular com a análise da rede de ligações de hidrogênio em camadas, nós pudemos avaliar a extensão dos efeitos de superfície nas propriedades dinâmicas e os detalhes da interface calcita/salmoura. Com a abordagem desenvolvida, conseguimos isolar os efeitos específicas dos íons da solução aquosa na rede de ligações de hidrogênio. Mostramos que a camada superficial exibe uma topologia de rede semelhante à observada em água pura, uma vez que a barreira eletrostática e física exibida por essa região, inibe a adsorção de íons na superfície da calcita. Fora dessa faixa, os íons influenciam consideravelmente a rede de ligações de hidrogênio: observamos a formação caminhos geodésicos mais extensos em relação àqueles observados em água pura. Esses ramos, que são formados por ligações de hidrogênio contíguas, podem conectar moléculas de baixa a alta dinâmica. Tal estrutura, pode explicar as propriedades mecânicas adesivas observadas em fluidos altamente confinados. Nossas principais contribuições decorrem na descrição da estrutura do solvente, dos íons da solução aquosa na interface calcita/fluido; e suas indicações físicas, e seu potencial significado nos processos de crescimento e dissolução de cristais. Nossas implementações fornecem contribuições interessantes para a compreensão atual dos processos que ocorrem em meios porosos. Especialmente, podendo contribuir para um desenvolvimento racional de novos dispositivos nanofluidicos. / The unusual physical properties exhibit by fluids within nanoscopic porous media play an important role in the plethora of chemical, geochemical and environmental processes. Currently, many aspects of the structure and dynamics of the spatially constrained fluids are still poorly understood. Additionally, the interfacial phenomena considerably influences the processes occurring in nanoporous media, which can have a major effect on nanofluidics devices. These multiphase systems and multi-physics phenomena occurs at solid/solution interfaces, with electronic and dynamic effects taking place across size and time scales. Currently, a single methodology is not capable to disentangle all the complexity find in such systems because it is restricted to a specific scale or computationally demand. In addition, the usual computational modeling methodologies applied to investigate bulk phases, they are, in general, not suitable to systematically access the surface effects occurring at solid/fluid interfaces. The challenges imposed by the nanofludics-based systems within the molecular modeling framework require innovative initiatives (among the available methodologies) to correctly access the interface properties. In this thesis, we develop and apply novel computational approaches to properly design and characterize nanofluidics-based systems at atomic level. In this context, we introduced an hierarchical top-down multilevel method by combining molecular dynamics simulations with first principles electronic transport calculations to address the multiscale phenomena problem. The potential of this implementation was demonstrated in a case study involving the water and ionic (Na, Li, and CL) flow through a (6,6) carbon nanotube. We showed that the ionic trace, observed on the electronic transmittance, it may handle an indirect measurement of the ionic current that is recorded as a sensing output. We implemented also a layered version of hydrogen bond network analysis based on graph theory. With this approach, we were able to properly explore interface effects arising on spatially confined fluids. By combining molecular dynamics simulations with the layered hydrogen bond network analysis, we evaluated the extension of surface effects on the fluids dynamics properties and the interaction details at calcite/brine interface. With the developed approach, we have been able to isolate the specific features of the aqueous solutions ions on the hydrogen bond network. We showed that the surface layer near the calcite/brine interface displays similar network topology as observed in pure water, since the electrostatic and physical barrier displayed by this layer inhibit the adsorption of ions on the calcite surface. Outside that region, these ions affect the hydrogen bond network. We observed a more extended geodesic paths with respect to that observed in pure water. Such hydrogen bond branches may connect low to high dynamics molecules across the pore and hence, it may explain the glue-like mechanical properties observed in confinement environment. Our main contributions in this work relies on describing the structure of solvent and electrolyte aqueous solution at calcite/fluid interface and their physical indications and potential significance on the crystal growth and dissolution processes. Our implementations provide interesting contributions to the current understanding of processes occurring in porous media. Specially, it may contribute on the rational design of novel nanofluidics devices.

As propriedades físicas incomuns exibidas por fluidos confinados em meios porosos desempenham um papel importante em diversos processos químicos, geoquímicos e ambientais. Atualmente, muitos aspectos da estrutura e dinâmica dos fluidos espacialmente confinados ainda são pouco compreendidos. Nesse contexto, fenômenos interfaciais influenciam consideravelmente os processos que ocorrem em meios nanoporosos, podendo resultar em efeitos relevantes para o desenvolvimento dos dispositivos nanofluidicos. Esses sistemas multifásicos e com fenômenos multifísicos podem apresentar propriedades eletrônicos e dinâmicos envolvendo diferentes escalas de tamanho e tempo na interface sólido/fluido. Atualmente, uma única metodologia não é capaz de resolver toda a complexidade encontrada em tais sistemas pelo fato de cada qual estar restrita a uma escala ou demanda computacional específica. Além disso, as metodologias habitualmente aplicadas para investigar as fases bulk através da modelagem computacional, em geral, não são adequadas para acessar sistematicamente os efeitos de superfície que ocorrem na interface sólido/fluido. Os desafios impostos à modelagem molecular pelos sistemas nanofluídicos requerem iniciativas inovadoras (dentre as metodologias disponíveis) para acessar as propriedades de interface. Nessa tese, desenvolvemos e aplicamos novas abordagens computacionais em nível atômico a fim de modelar e caracterizar sistemas nanofluidicos. Nesse contexto, introduzimos um método multinível hierárquico top-down, que combina simulações de dinâmica molecular com cálculos ab initio de transporte eletrônico, para abordar fenômenos de multiescala. O potencial dessa implementação foi demonstrado em um estudo de caso envolvendo o fluxo de água e o transporte de íons através de um nanotubo de carbono tipo (6,6). Mostramos que o traço iônica pode representar uma mudança na condutância elétrica do nanocanal, e levar a uma medida indireta da corrente iônica. Também implementamos uma versão modificada da análise de rede de ligações de hidrogênio baseada em teoria de grafos, a fim de fazer o estudo das propriedades estruturais e dinâmicas em diferentes regiões do poro. Com essa abordagem, nós fomos capazes de explorar sistematicamente os efeitos de interface em fluidos espacialmente confinados. Combinando-se simulações de dinâmica molecular com a análise da rede de ligações de hidrogênio em camadas, nós pudemos avaliar a extensão dos efeitos de superfície nas propriedades dinâmicas e os detalhes da interface calcita/salmoura. Com a abordagem desenvolvida, conseguimos isolar os efeitos específicas dos íons da solução aquosa na rede de ligações de hidrogênio. Mostramos que a camada superficial exibe uma topologia de rede semelhante à observada em água pura, uma vez que a barreira eletrostática e física exibida por essa região, inibe a adsorção de íons na superfície da calcita. Fora dessa faixa, os íons influenciam consideravelmente a rede de ligações de hidrogênio: observamos a formação caminhos geodésicos mais extensos em relação àqueles observados em água pura. Esses ramos, que são formados por ligações de hidrogênio contíguas, podem conectar moléculas de baixa a alta dinâmica. Tal estrutura, pode explicar as propriedades mecânicas adesivas observadas em fluidos altamente confinados. Nossas principais contribuições decorrem na descrição da estrutura do solvente, dos íons da solução aquosa na interface calcita/fluido; e suas indicações físicas, e seu potencial significado nos processos de crescimento e dissolução de cristais. Nossas implementações fornecem contribuições interessantes para a compreensão atual dos processos que ocorrem em meios porosos. Especialmente, podendo contribuir para um desenvolvimento racional de novos dispositivos nanofluidicos. / The unusual physical properties exhibit by fluids within nanoscopic porous media play an important role in the plethora of chemical, geochemical and environmental processes. Currently, many aspects of the structure and dynamics of the spatially constrained fluids are still poorly understood. Additionally, the interfacial phenomena considerably influences the processes occurring in nanoporous media, which can have a major effect on nanofluidics devices. These multiphase systems and multi-physics phenomena occurs at solid/solution interfaces, with electronic and dynamic effects taking place across size and time scales. Currently, a single methodology is not capable to disentangle all the complexity find in such systems because it is restricted to a specific scale or computationally demand. In addition, the usual computational modeling methodologies applied to investigate bulk phases, they are, in general, not suitable to systematically access the surface effects occurring at solid/fluid interfaces. The challenges imposed by the nanofludics-based systems within the molecular modeling framework require innovative initiatives (among the available methodologies) to correctly access the interface properties. In this thesis, we develop and apply novel computational approaches to properly design and characterize nanofluidics-based systems at atomic level. In this context, we introduced an hierarchical top-down multilevel method by combining molecular dynamics simulations with first principles electronic transport calculations to address the multiscale phenomena problem. The potential of this implementation was demonstrated in a case study involving the water and ionic (Na, Li, and CL) flow through a (6,6) carbon nanotube. We showed that the ionic trace, observed on the electronic transmittance, it may handle an indirect measurement of the ionic current that is recorded as a sensing output. We implemented also a layered version of hydrogen bond network analysis based on graph theory. With this approach, we were able to properly explore interface effects arising on spatially confined fluids. By combining molecular dynamics simulations with the layered hydrogen bond network analysis, we evaluated the extension of surface effects on the fluids dynamics properties and the interaction details at calcite/brine interface. With the developed approach, we have been able to isolate the specific features of the aqueous solutions ions on the hydrogen bond network. We showed that the surface layer near the calcite/brine interface displays similar network topology as observed in pure water, since the electrostatic and physical barrier displayed by this layer inhibit the adsorption of ions on the calcite surface. Outside that region, these ions affect the hydrogen bond network. We observed a more extended geodesic paths with respect to that observed in pure water. Such hydrogen bond branches may connect low to high dynamics molecules across the pore and hence, it may explain the glue-like mechanical properties observed in confinement environment. Our main contributions in this work relies on describing the structure of solvent and electrolyte aqueous solution at calcite/fluid interface and their physical indications and potential significance on the crystal growth and dissolution processes. Our implementations provide interesting contributions to the current understanding of processes occurring in porous media. Specially, it may contribute on the rational design of novel nanofluidics devices.

<p>Water is one of the most common compounds on earth and is essential for all biological activities. Water has, however, been a mystery for many years due to the large number of unusual chemical and physical properties, e.g. decreased volume during melting and maximum density at 4 °C. The origin of the anomalies behavior is the nature of the hydrogen bond. This thesis will presented an x-ray absorption spectroscopy (XAS) study to reveal the hydrogen bond structure in liquid water.</p><p>The x-ray absorption process is faster than a femtosecond and thereby reflects the molecular orbital structure in a frozen geometry locally around the probed water molecules. The results indicate that the electronic structure of liquid water is significantly different from that of the solid and gaseous forms. The molecular arrangement in the first coordination shell of liquid water is actually very similar as the two-hydrogen-bonded configurations at the surface of ice. This discovery suggests that most molecules in liquid water have two-hydrogen-bonded configurations with one donor and one acceptor hydrogen bond compared to the four-hydrogen-bonded tetrahedral structure in ice. This result is controversial since the general picture is that the structure of liquid water is very similar to the structure of ice. The results are, however, consistent with x-ray and neutron diffraction data but reveals serious discrepancies with structures based on current molecular dynamics simulations. The two-hydrogen-bond configuration in liquid water is rigid and heating from 25 °C to 90 °C introduce a minor change in the hydrogen-bonded configurations. Furthermore, XAS studies of water in aqueous solutions show that ion hydration does not affect the hydrogen bond configuration of the bulk. Only water molecules in the close vicinity to the ions show changes in the hydrogen bond formation. XAS data obtained with fluorescence yield are sensitive enough to resolved electronic structure of water molecules in the first hydration sphere and to distinguish between different protonated species. Hence, XAS is a useful tool to provide insight into the local electronic structure of a hydrogen-bonded liquid and it is applied for the first time on water revealing unique information of high importance.</p>

Water is one of the most common compounds on earth and is essential for all biological activities. Water has, however, been a mystery for many years due to the large number of unusual chemical and physical properties, e.g. decreased volume during melting and maximum density at 4 °C. The origin of the anomalies behavior is the nature of the hydrogen bond. This thesis will presented an x-ray absorption spectroscopy (XAS) study to reveal the hydrogen bond structure in liquid water. The x-ray absorption process is faster than a femtosecond and thereby reflects the molecular orbital structure in a frozen geometry locally around the probed water molecules. The results indicate that the electronic structure of liquid water is significantly different from that of the solid and gaseous forms. The molecular arrangement in the first coordination shell of liquid water is actually very similar as the two-hydrogen-bonded configurations at the surface of ice. This discovery suggests that most molecules in liquid water have two-hydrogen-bonded configurations with one donor and one acceptor hydrogen bond compared to the four-hydrogen-bonded tetrahedral structure in ice. This result is controversial since the general picture is that the structure of liquid water is very similar to the structure of ice. The results are, however, consistent with x-ray and neutron diffraction data but reveals serious discrepancies with structures based on current molecular dynamics simulations. The two-hydrogen-bond configuration in liquid water is rigid and heating from 25 °C to 90 °C introduce a minor change in the hydrogen-bonded configurations. Furthermore, XAS studies of water in aqueous solutions show that ion hydration does not affect the hydrogen bond configuration of the bulk. Only water molecules in the close vicinity to the ions show changes in the hydrogen bond formation. XAS data obtained with fluorescence yield are sensitive enough to resolved electronic structure of water molecules in the first hydration sphere and to distinguish between different protonated species. Hence, XAS is a useful tool to provide insight into the local electronic structure of a hydrogen-bonded liquid and it is applied for the first time on water revealing unique information of high importance.

La précipitation et la dissolution du carbonate de calcium (CaCO3) sont des processus clés dans les systèmes naturels en raison de leur association intime avec le cycle du carbone terrestre. La précipitation se produit généralement sur des substrats étrangers en abaissant les barrières énergétiques qui contrôlent la nucléation. Ce processus appelé nucléation hétérogène résulte d'une interaction entre la sursaturation du fluide et les différentes énergies d’interface entre substrat-noyau-fluide. Malgré l’importance des énergies d’interface sur le devenir de la nucléation hétérogène, la littérature actuelle reste rare dans leurs valeurs absolues, limitant la précision de la modélisation du transport réactif. La formation des biominéraux constitue un réservoir majeur des carbonates dans la lithosphère. Des études récentes ont révélé des nucléations par multi-étapes impliquant la formation du carbonate de calcium amorphe (ACC), un intermédiaire métastable durant les premiers stades de la formation des biominéraux. De tels précurseurs amorphes permettent de réaliser les formes complexes des biominéraux, tandis que leur stabilité et leur cinétique de cristallisation sont contrôlées par de multiples facteurs. L'élucidation des mécanismes sous-jacents est bénéfique pour le développement de matériaux biomimétiques.Le premier objectif est de développer une compréhension prédictive des valeurs d'énergie d’interface régissant la nucléation hétérogène du CaCO3 en fonction des propriétés physico-chimiques spécifiques des substrats, comme l'hydrophobicité. Cette dernière est étudiée en utilisant de la phlogopite avec et sans substitution par le fluor produisant des substrats hydrophobes et hydrophiles. La technique de diffusion des rayons X aux petits angles en incidence rasante a été employée in situ pour obtenir des valeurs d’énergie effective d’interface. Il est intéressant de noter que les valeurs extraites pour les deux substrats sont similaires, et thermodynamiquement les deux fournissent un bon modèle pour la nucléation, alors que leurs mécanismes sont différents. La caractérisation ex situ par microscopie à force atomique a montré que le substrat hydrophile favorise la formation et la stabilisation d’ACC, tandis que le substrat hydrophobe favorise la formation de calcite. Ces résultats soulignent la flexibilité structurelle intrinsèque du CaCO3 et son avantage dans les processus de nucléation hétérogènes.Le deuxième objectif est de fournir une description atomistique de l'hydrophobicité du substrat. L'adsorption d'eau sur la phlogopite a été réalisée in situ par spectroscopie de photoélectrons à pression ambiante pour étudier l'effet de la substitution par le fluor et de différents types de contre-ions (K+, Na+ vs. Cs+). Ces résultats ont été interprétés par des simulations de dynamique moléculaire et la théorie de bond-valence. La combinaison de ces techniques montre que l'hydrophobicité du substrat provient d'une compétition entre deux facteurs: l'hydratation des contre-ions par rapport à celle du substrat.Le but final est d'étudier les mécanismes moléculaires par lesquels Mg2+, une impureté chez les précurseurs amorphes biogéniques, augmente la persistance cinétique d’ACC. La technique de diffusion inélastique incohérente des neutrons a été combinée avec la spectroscopie de corrélation de photons X pour élucider la dynamique à l'échelle nanométrique de l'eau et des ions dans les ACC. Les résultats montrent que la présence de Mg2+ augmente la diffusion atomique dans le solide tout en amplifiant la rigidité du réseau des liaisons hydrogène. Ces résultats contre-intuitifs sont abordés en considérant différents facteurs cinétiques inclus dans l’équation décrivant le taux de nucléation au sein de la théorie classique de la nucléation. Dans l'ensemble, ces résultats indiquent l'importance de l'eau comme stabilisant cinétique de la structure amorphe et de l'existence de barrières stériques qui abaissent le taux de cristallisation. / Precipitation and dissolution of calcium carbonate (CaCO3) are key processes in both natural and engineered systems due to their intimate association with the Earth’s carbon cycle. Precipitation usually occurs on foreign substrates since they lower the energetic barriers controlling nucleation events. This so-called heterogeneous nucleation results from the interplay between the fluid supersaturation and the interfacial free energies present at the substrate-nucleus-fluid interfaces. Despite the relevance of interfacial energies for the fate of heterogeneous nucleation, the current literature remains scarce in their absolute values, which limits the accuracy of reactive transport modelling. Of particular relevance to the carbon cycle, the formation of biominerals accounts for a major reservoir of the carbonate minerals in the lithosphere. Recent studies have revealed the existence of multistep nucleation pathways that involve formation of amorphous calcium carbonate (ACC), a metastable intermediate during the early stages of biomineral formation. Such amorphous precursors allow molding of the intricate shapes of biominerals, while their stability and crystallization kinetics are effectively controlled by multiple factors. Elucidating the underlying mechanisms is beneficial for the development of biomimetic materials.The first goal of this dissertation is to develop a predictive understanding of interfacial energy values governing CaCO3 heterogeneous nucleation as a function of specific physico-chemical properties of the substrates, such as hydrophobicity. This last was investigated using phlogopite, a common mica, with and without fluorine substitution yielding hydrophobic and hydrophilic substrates. In situ time-resolved Grazing-Incidence Small Angle X-ray Scattering experiments were performed to obtain effective interfacial energy values. Interestingly, the extracted values for both substrates were similar, and thermodynamically these substrates provide a good template for nucleation, but the pathways differ. By ex situ Atomic Force Microscopy characterization, the hydrophilic substrate was shown to promote the formation and stabilization of ACC, whereas the hydrophobic one favored the formation of calcite. These results point to the intrinsic structural flexibility of CaCO3 and its advantage in heterogeneous nucleation processes.The second goal is to provide an atomistic description of the substrate hydrophobicity/hydrophilicity. Water adsorption on phlogopite was studied in situ using Near-Ambient Pressure X-ray Photoelectron Spectroscopy to investigate the effect of fluorine substitution and the influence of different types of counterions (K+, Na+ vs. Cs+). The results of the spectroscopy experiments were further interpreted using molecular dynamics simulations and bond-valence theory. The combination of these techniques shows that the substrate hydrophobicity stems from a competition between two factors: hydration of counterions vs. that of substrate.The final goal is to study the molecular mechanisms by which Mg2+, a common impurity in biogenic amorphous precursors, increases the kinetic persistence of ACC. Inelastic Incoherent Neutron Scattering and X-ray Photon Correlation Spectroscopy were combined to elucidate the nanoscale dynamics of water and ions within ACC. The presence of Mg2+ was shown to enhance the atomic diffusion within the solid while simultaneously increasing the stiffness of the hydrogen bond network. These counter-intuitive results are addressed by considering the different factors included in the pre-exponential term of the nucleation rate equation within the framework of the classical nucleation theory. Overall, the results point to the importance of water as a kinetic stabilizer, and to the existence of steric barriers that lower the crystallization rate.