No description available.

Adsorption is widely used for gas storage and separation for industrial and environmental applications. With appropriate adsorbents adsorption represents a low-energy alternative to conventional separation technologies. Hybrid organic-inorganic materials offer the ability to tailor the pore geometry as well as the surface chemistry, which are the two governing properties of an adsorbent with respect to its adsorption behaviour. Thus they enable the design of specific adsorbents for gas separation processes. By means of molecular simulation, the adsorption behaviour can be predicted prior to manufacturing the material, reducing the effort of designing and/or screening of adsorbents significantly. This work focuses on two groups of hybrid materials. On the one hand, we studied aluminium methylphosphonates which are crystalline, organic-inorganic microporous materials. They can exhibit molecular sieving effects and as they are crystalline the atomic co-ordinates, which are necessary to perform molecular simulations of the adsorption process, can be derived accurately from XRD studies. On the other hand, organically modified, non-crystalline periodic mesoporous silicas (PMSs) can be prepared with different pore geometries and their mesopores are sufficiently large to accommodate even large surface groups. We have developed new methods to generate atomic models for the amorphous wall structure of PMSs by means of computer simulation of their synthesis. This so called kinetic Monte Carlo simulation is presented in combination with Grand Canonical Monte Carlo simulation of adsorption as a tool in a design procedure for adsorbents for gas separation. The application of these design tools is demonstrated by designing an MCM-41-type hybrid adsorbent for the removal of carbon dioxide from power-plant flue gas, and by optimising the process conditions for the separation of air into nitrogen and oxygen using microporous A1MePO-α.

No description available.

The adsorption of mixtures of polar and non-polar species is influenced by both the structure and the surface chemistry of the adsorbent. Several molecular simulation techniques have been employed to study the effect of these factors on the extent and mechanism of adsorption of water and light hydrocarbons. In particular, the effect of pore size as well as the concentration, distribution and type of polar surface groups have been examined in a detailed and systematic way. It is shown in this thesis that the presence of polar sites on the surface of the carbon enhances the affinity of the adsorbent towards water. However, in most situations of practical interest, the most important variable was seen to be the concentration of polar sites, with their distribution and type playing only a minor part. This fundamental study formed a basis for the development of a model for activated carbon that can be used in an industrial context. The model for activated carbon includes a representation of the structure of the adsorbent using a distribution of slit-shaped pores, and of the surface chemistry, using a distribution of polar sites. The adsorption in single pores was calculated by grand canonical Monte Carlo simulation. The pore size distribution was obtained from an analysis of pure-ethane adsorption isotherms, while the polar site distribution was calculated by analysing pure-water adsorption. The performance of the model was assessed by comparing the simulated results for binary water/ethane adsorption with experimental adsorption data on a commercial activated carbon. The sensitivity of the predictions to changes in the surface chemistry of the model was examined. The model proposed here was seen to perform much better than the more widely used classical methods, opening good prospects for its use in industrial applications.

No description available.

Unsteady state measurements of oxygen dissolution rates in water were carried out in a large baffled aerated mixing vessel (0.6 m8 capacity) at various gas flow rates and agitator power inputs for a variety of impeller types. A mathematical model accounting for the gas and liquid phase dynamics was formulated to evaluate the mass transfer product ()La). Measurements of the dispersion bubble properties (diameter, interfacial area and gas bold-up) were also made and correlation, for gas hold-up, interfacial area and aerated power dissipation were proposed. It is believed that valid values of kLa have been determined for the air-water system, using a correct mathematical model; most of the earlier studies using physical methods, reported in the literature, have made several unjustified assumptions in determining kLa. The order of merit of several different agitators in regard to effective use of power for gas-liquid mass transfer was established and a new type (the comb turbine) was found to be superior, under certain circumstances, to the conventionally used "Rushton turbine". The effect of scale-up on kLa in aerated mixing vessels was also experimentally examined by carrying out some experiments in a geometrically similar smaller tank (0.049 m). A deterioration in mass transfer rate is found on scale-up at constant power per unit volume of liquid and constant superficial gas velocity, but further work is required to establish the causes.

No description available.

No description available.

An experimental investigation of heat transfer in packed beds of low tube/particle diameter ratio is reported and the most widely-used two-dimensional homogeneous continuum models are thoroughly tested by statistical methods using the experimental data obtained. It is shown that the omission of axial dispersion effects leads to significant lack-of-fit in such models and to parameter estimates which vary systematically with bed depth. A model including axial dispersion is considered for each of two possible simplified downstream boundary conditions. This axially-dispersed model shows no lack-of-fit and yields depth-independent parameter estimates when the boundary condition is placed at infinity; when the alternative condition at bed exit is used this model shows little improvement over the model which omits axial dispersion. A new theory for predicting the axial and radial effective thermal conductivities and the apparent wall heat transfer coefficient is derived from a two-phase continuum model containing the essential underlying and independently measurable heat transfer processes. The theory gives good agreement with the results obtained in this work, in contrast with previously existing theory, explains much of the confused literature data, and pin-points the remaining major areas of uncertainty.

No description available.