Three materials containing Ni 2 P, Ni 12 P 5 , and Ni 3 P phases on silica gel with surface area 320 m 2 /g at loadings of 32-37 wt % and the crystal size of Ni x P phases 30, 9, and 13 nm, respectively, were prepared by a combination of impregnation and TPR methods and tested in hydrodesulfurization (HDS) and adsorptive desulfurization (ADS) of diesel fuel. There were established opposite trends in changing the DS efficiency in two processes: The HDS rate constant decreased while the ADS sulfur capacity (breakthrough at 1 ppmw) increased with increasing the Ni to P ratio in Ni x P from 2 to 3. The observed behavior was attributed to the specific features of the densities of states (DOS) obtained from the density functional theory calculations of total and partial DOS for Ni and P in Ni x P phases and revealed in XPS measurements of binding energy of Ni 2p 3/2 -and P 2p-electrons. This attribution was consistent with the analysis of the relative part of d-electrons of Ni participating in bonding with p-electrons of phosphorus in these phases.

A significant increase of the sulfur uptake by Ni x P with an increasing Ni to P ratio from 2 to 3 was reported recently in ultradeep adsorption desulfurization (ADS) of diesel fuel. It corresponded to increased Ni x P surface coverage, while no bulk Ni−P−S phases were identified. Density functional theory is applied in this communication to calculate the adsorption energies for S adatoms adsorbed on the surface of Ni2P, Ni12P5, and Ni3P phases. The preferable (001) surface-terminated slabs are considered to link the coverage tendency with the peculiarities of the electron density of states at the Fermi energy. The electron charge redistribution due to the adsorption of S atoms is calculated, and the nature of bonding of S with the substrates is discussed. The combination of ab initio calculations performed in the ground state for the considered systems with statistical thermodynamics renders the possibility to obtain the coverage at the temperatures beyond T = 0 K in good agreement with experimental data. The increase of surface coverage of Ni x P phases with an increase in the x value is the result of fine interplay between the number of possible sites for S adsorption and the difference in the adsorption energies for these sites.

The increase of the surface basicity-acidity of MgO material by factors of 1.8-3.0 due to consolidation of its nanocrystals was demonstrated by the indicator titration. It was shown that the parallel increase of surface acidity and basicity is attributed to the formation of grain boundaries (GB) after MgO aerogel densification. A simple model predicting the increase of surface acidity-basicity of MgO that correlates with the results of direct measurements was proposed. The model is based on the study of the fine atomic structure at GB surface areas in consolidated MgO nanocrystals in the framework of Density Functional Theory. It is found that the displacements of coordinatively unsaturated surface ions near the GB are significant at the distances ~3-4 atomic layers from the geometrical contact plane between nanocrystals. The detailed analysis of atomic positions inside GB demonstrated the coordination deficiency of surface atoms at the GB areas leading to the formation of stretched bonds and to creation of low coordinated surface ions due to splitting of coordination numbers of surface atoms belonging to GB areas. Density of states for electrons shows the existence of additional states in the band gap close to the bottom of the conduction band. The adsorption energy of CO2 molecules atop oxygen atoms exposed at surface GB areas is of the same order of magnitude as that reported for oxygen atoms at crystallographic edges and corners of MgO crystals. It provides additional options for bonding of molecules at the surface of nanocrystalline MgO increasing the adsorption capacity and catalytic activity.

Direct measurements of the dissociation of I2 molecules at the optical axis of a supersonic chemical oxygen-iodine laser (COIL) as a function of I2 flow rate were carried out. This enabled us to determine the number of consumed O2(Δ1) molecules per dissociated I2 molecule. The number depends on the experimental conditions: it is 4.2±0.4 for typical conditions and I2 densities applied for the operation of the COIL, but increases at lower I2 densities. Possible dissociation mechanisms consistent with our results are discussed and the importance of dissociating I2 prior to its mixing with O2(Δ1) is stressed.

No abstract