In aqueous solutions, a charged surface causes the redistribution of nearby ions. The ion layers formed are known as the electrical double layer (EDL), and are widespread in many systems involving electrochemistry, colloidal science, biomedicine, and energy storage. Over the centuries, models have been developed to describe the structure of the EDL, including the Helmholtz, Gouy-Chapman, and Gouy-Chapman-Stern models. While these models have given fairly accurate descriptions, they have limitations in depicting the structure of the EDL at the molecular level. In this thesis, we focus on the EDL of colloidal silica nanoparticles, not only because of the wide applications of silica nanoparticles in drug delivery, cosmetics, food additives, etc, but also because of their high stability and tunable sizes. We employ angle-resolved second harmonic scattering (AR-SHS), a label-free nonlinear optical method, to characterize the EDL around silica colloidal nanoparticles. AR-SHS is a highly sensitive tool providing information on two interfacial properties: Surface potential Phi0 and interfacial water orientation Chi2. We first calculate the size dependence of the surface and electrostatic geometric form factor functions for nonlinear scattering, together with their relative contribution to the AR-SHS patterns. This theoretical part gives an insight into the scattering efficiency of NPs of different sizes. We then show to which extent the SH signal of different particle sizes is modulated by the material's surface properties and the composition of the aqueous environment. Finally, we propose a scheme to predict the dominating optical response (whether the surface or electrostatic) as a function of particle size. Our work establishes a quantified connection between experimental AR-SHS patterns and the material's surface properties of the particle. We then explore the evolution of the EDL thickness with the increase of ionic concentration using AR-SHS. Following previous work in our group, we characterize the behavior of the EDL at salt concentrations below the millimolar range, including inner-sphere adsorption, diffuse layer formation, and outer-sphere adsorption. Moreover, we show for the first time that, by appropriately selecting the nanoparticle size, it is possible to retrieve information also in the millimolar range. In this range, a decrease in Ί_0 is observed, suggesting a compression in the EDL thickness. Molecular dynamics simulations show that the EDL compression is primarily due to the compression of the diffuse layer rather than that of the Stern plane. Finally, we examine the influence of the EDL ion distribution on the AR-SHS signal. We derive the scattering form factors for the electrostatic response for five different ion distributions, each representing a different electrostatic potential decay in the EDL. Using the same ion distributions, we then calculate the integrated AR-SHS intensities and extract the surface parameters Phi0 and Chi2 through fitting of the AR-SHS data. The results indicate that, although there are deviations amongst the different electric potential functions, the trends of Phi0 and Chi2 extracted can be captured successfully irrespective of the ion distribution. The study successfully demonstrates the universality of the AR-SHS approach to establish a molecular-level picture of the EDL.