RESUMEN
We conducted molecular dynamics (MD) simulations to calculate the density and surface tension of concentrated ammonium nitrate (AN) solutions up to the solubility limit of ammonium nitrate in water, by combining the SPC/E, SPCE/F and TIP4P/2005 water models with OPLS model for ammonium and nitrate ions. This is the first time that the properties of concentrated solutions of nitrates, especially AN, have been studied by molecular dynamics. We effectively account for the polarisation effects by the electronic continuum correction (ECC), practically realised via rescaling of the ionic charges. We found that, the full-charge force field MD simulations overestimate the experimental results, as the ions experience repulsion from the interface and prefer to remain in the subsurface layer and the bulk solution. In contrast, reducing the ionic charges results in the behaviour that fits well with the experimental data. The nitrate anions display a greater propensity for the interface than the ammonium cations. We accurately predict both the density and the rise in the surface tension of concentrated solutions of AN, recommending TIP4P/2005 for water and the scaled-charge OPLS model (OPLS/ECC) for the ions in the solutions. We observe that, the adsorption of anions to the interface accompanies their depletion in the subsurface layer, which is preferentially occupied by cations, resulting in the formation of the electric double layer. We demonstrate the ion deficiency for up to 3 Å below the surface and establish the requirement to include the polarisability effects in the OPLS model for AN. While these results confirmed the findings of the previous studies for dilute solutions, they are new in the solubility limit. Concentrated solutions exhibit a strong effect of the abundance of solute on the coordination numbers of ions and on the degree of ion pairing. Surprisingly, ion pairing decreases significantly at the interface compared with the bulk. The present study identifies OPLS/ECC, along with TIP4P/2005, to yield accurate predictions of physical properties of concentrated AN, with precision required for industrial applications, such as a formulation of emulsion and fuel-oil explosives that now predominate the civilian use of AN. An application of this model will allow one to predict the surface properties of supersaturated solutions of AN which fall outside the capability of the present laboratory experiments but are important industrially.
RESUMEN
Catalysts of iron oxide on γ-alumina and silica which were prepared by an incipient wetness impregnation technique have been investigated in an effort to understand how the surface chemical properties are influenced by the nature of the supports. Surprisingly, this is the first study to compare in depth the influence of the supports on physicochemical parameters such as acidity, site nuclearity, and reducibility. In this study, surface characterisation techniques including N2 physisorption at -196 °C, ammonia temperature-programmed desorption, inductively coupled plasma optical emission spectrometry, temperature-programmed reduction with hydrogen, CO-chemisorption, scanning electron microscopy, transmission electron microscopy, and NO adsorption by in situ Fourier transform infrared spectroscopy have been performed to understand the different surface reactions occurring over the two different supports. The aim of this study is to ascertain the primary differences between these two catalysts using several catalyst characterization techniques and correlate their chemical and structural differences to their catalytic activity in the conversion of 2-chlorophenol. The results disclose a higher density of acid sites, a smaller particle size of iron oxide, stabilization of Fe(II) aluminate after reduction on the alumina surface, and finally, the formation of isolated iron cations on the surface of alumina which are notably absent on the silica-supported catalyst.