Aerogels, produced by sol-gel technologies, have several applications in sensors, high energy particle physics, catalysis, heat insulation, supercapacitors, heat storage devices, high efficiency windows, among others. These applications take advantage of the outstanding properties these materials present as a result of their structure. However, the low mechanical properties that these materials present as result of the process, limits their commercial applications. In this dissertation, it is investigated the relationship between the processing conditions and mechanical properties of these materials computationally. The prediction of the effective properties for these materials is a daunting task because of their complex structure. Aerogels's structure is not homogeneous nor periodic, but rather amorphous, nanostructured, and highly porous, making the traditional techniques used to study other materials inapplicable. This dissertation presents the prediction of mechanical properties of aerogels calculated by a novel Multifractal Multidimensional Multiscaling Approach (MMMA) developed here. MMMA consists on recursively calculating the effective properties of the material along several scales. Since aerogels and structures produced by sol-gel technologies present a multifractal character, it is shown that MMMA is applicable to predict the effective properties of these materials. The implementation of MMMA requires a fractal characterization of the structure. For this, computational scattering experiments were performed on structures resembling aerogels. The structures resembling aerogels were produced computationally incorporating the chemistry and the physical phenomena involved in the formation process. MMMA was used to predict the mechanical properties of silica aerogels for different processing conditions. Thus, mechanical properties, scattering experiments, and processing conditions were investigated and correlated in this work. I want to thank many people that contributed directly or indirectly with this dissertation and the completion of my PhD. First, I want to thank Dr. Ever Barbero, that more than my advisor gave me the opportunity to dig into fundamental problems as the subject of this dissertation. From the office, Sandro Rivas who suffered major long talks about a diversity of problems related to the dissertation, as well as Joaquin Gutierrez, Adi Adumitroaie, and Juan Cruz. I also want to thank Marco Maurier who helped writing Fortran codes. I also want to thank the committee members for their comments and assistance. Last but not least I want to thank Dr. Boyd Edwards who besides introducing me into Non-linear dynamics, chaos and fractals, helped me as a committee member as long as the logistics allowed. iv Contents Acknowledgement iv List of Figures ix List of Tables xi List of Symbols xii D P property scaling exponent for decoupled case κ t thermal conductivity