Mineral crystallization is a fundamental and ubiquitous phenomenon that occurs in various natural and industrial processes, including bio mineralization, rocksphere cycling, drug purification, membrane filtration, CO2 geological sequestration, scale formation in oil and gas industry and cooling towers, to mention a few. Extensive studies have been done in the past century to better understand, predict, and control the mineral crystallization process, of which the importance can never be overstated. To achieve these goals, three questions need to be answered: what is the mineral solubility? what is the kinetics and mechanisms of mineral crystallization? and what is the kinetics and mechanisms of the inhibition of mineral crystallization? To answer these three questions, this thesis has done research in three corresponding aspects: (1) The thermodynamic property predictions for water-mineral-gas systems, for example, mineral solubility (e.g. halite, barite, calcite, gypsum, anhydrite, and celestite), CO2 solubility, and aqueous solution density, are important for various disciplines. With the advances of technology, a need for a comprehensive and accurate thermodynamic model to predict such properties at high temperature (up to 250 oC), high pressure (up to 1,500 bar), and high concentration of total dissolved solids (more than 300,000 mg/L) in the presence of mixed electrolytes has emerged. This thesis has established a Pitzer theory based thermodynamic model to fulfill these requirements for the Na-K-Mg-Ca-Ba-Sr-Cl-CO3-HCO3-SO4-CO2(aq) system. Based on a thorough review of previous studies, a set of consistent virial coefficients, standard partial molar volumes, and equilibrium constants developed in previous models were adopted in this thesis. The temperature and pressure dependences of other required virial coefficients were derived by simultaneously fitting the solubility data of the minerals and CO2 as well as solution density data, which were measured in literature and by this thesis. With this model, the 95% confidence intervals of the estimation errors for solution density predictions are within 4×10-4 g/cm3. The relative errors of CO2 solubility prediction are within 0.75%. The estimation errors of the SI mean values for barite, calcite, gypsum, anhydrite, and celestite are within ± 0.1, and that for halite is within ± 0.01, most of which are within experimental uncertainties. This model has been incorporated into ScaleSoftPitzer, an EXCEL based geochemistry software package developed by the authors. (2) In the past century, many mechanisms and models have been proposed to explain observations in different crystallization stages. However, most models only focus on a certain step or mechanism (e.g. nucleation, aggregation) and lack a comprehensive view. Incorporating nucleation, aggregation, and surface reaction together, this thesis has developed an analytical two-stage crystallization model to simulate the particle size and number concentration versus time and correlate them with the measured solution turbidity. Through measuring solution turbidity in real time, this model can reproduce the crystallization process by predicting the key parameters: nucleation rate, particle size, number concentration, surface tension, induction time, and particle linear growth rate. Most of these values for barite crystallization match with literature data and our direct cryo-transmission electron microscopy (cryo-TEM) measurements. Moreover, the established relationships of these key parameters versus temperature and supersaturation enable this model to predict barite crystallization kinetics based only on the initial supersaturation and temperature. (3) This thesis has developed a new theoretical model to analyze the kinetics and mechanisms of crystallization inhibitions based on the classical nucleation theory and regular solution theory. The new model assumes that inhibitors can impact the nucleus partial molar volume and the apparent saturation status of the crystallization minerals. These two impacts were parameterized to be proportional to additive concentrations and vary with inhibitors. This new model has been used to predict barite induction times without inhibitors from 4 to 250 oC and in the presence of eight different inhibitors from 4 to 90 oC, and calcite induction times with or without ten different inhibitors from 4 to 175 oC. The predicted induction times showed close agreement with the experimental measurements. Such agreement indicates that this new theoretical model can be widely adopted in various disciplines to evaluate mineral formation kinetics, elucidate mechanisms of additive impacts, predict minimum effective dosage (MED) of additives, and guide the design of new additives, to mention a few.