Search

Improving and developing simulation techniques are key to obtaining higher efficiency and accuracy in molecular simulations of dense liquid systems. The methodology development introduced in this thesis is relevant both for academia and industrial applications. In this thesis, the methods developments/improvements for molecular simulations are introduced followed by applications for realistic systems and systems of industrial relevance....

One of the most important steps we can take to minimize global warming is to become more energy efficient. An important example with potential for improvement is the large amount of energy wasted through waste heat streams in industry. For a long time there has already existed a solution that can efficiently upgrade waste heat streams to useful levels – heat pumps. However, heat pumps are rarely applied in industry. There are various reasons for this, including lack of knowledge and increased complexity. Another major reason is the long payback periods. This work focuses on a promising option that, for certain applications, can outperform traditional technologies by having a higher Coefficient of Performance (COP) and reaching higher temperatures: compression-resorption heat pumps (CRHP) using wet compression.

Undoubtedly, materials that can be tuned on a molecular level offer tremendous opportunities. However, to understand and customize such materials is challenging. In this context, molecular simulation can be helpful. The work presented in this thesis deals with two types of materials, Metal-Organic Frameworks and Ionic Liquids, and the study with molecular simulation to determine their potential for specific gas separations. For the prediction of their behavior and relevant materials properties with molecular simulation, force fields of sufficient quality are required..

Network materials, foams, and emulsions are ubiquitous in our daily life. We have a good intuition about how they respond as we handle them, but our theoretical understanding is poor. One of their most interesting features is that they are unusually fragile and appear to switch between solid and liquid state seamlessly. In fact, foams and emulsions undergo a non-equilibrium phase transition as their packing fraction increases - this is the jamming transition. Networks show a similar transition as their connectivity increases, where the material switches from sloppy to rigid.The fact that these materials undergo a phase transition, opens up the theoretical toolset of statistical mechanics. An important part of current research is therefore dedicated to finding diverging length and time scales and investigating the critical behavior of the systems in detail. Because the systems in question are highly disordered, analytical modeling is challenging. At the same time there are significant experimental obstacles to approaching the critical point closely. For this reason, the development of simulation software plays an important role - all data presented in this thesis is generated through simulations. As the subtitle of this dissertation suggests, our findings concern length, strain, and time scales which can be found in the linear response to external forces.

Improving efficiencies of thermal energy conversion systems is an important way to slow down global warming and mitigate climate change. Vapor absorption heat pump and refrigeration cycles are highly efficient ways of heating and cooling. These thermally activated systems also provide opportunities for the integration with a wide spectrum of low-grade and renewable heat sources, such as district heating networks, exhaust industrial heat, concentrated solar thermal energy and biomass. New fluids - ionic liquids - have been introduced into the absorption refrigeration/ heat pump field as absorbents to overcome drawbacks of traditional working fluids and to improve the energetic efficiency of systems. Some ionic liquids show high boiling points, superior thermal and chemical stabilities and strong affinities with refrigerants. Ammonia (NH3) is an environmentally friendly refrigerant with favorable thermodynamic and transport performance. Thus, studies in this thesis placed emphasis on the ammonia/ionic liquids working pairs. Studies in this thesis focus on exploring applications of ammonia/ionic liquid based vapor absorption refrigeration cycles, from a practical point of view in the refrigeration and heat pump field. By applying multi-scale evaluations covering thermodynamic and heat and mass transport aspects, it is intended to further understand the fundamentals of applying ionic liquids in heating and cooling systems. The highlights include: Assessments of equilibriummodels applied for ammonia-ionic liquid working fluids; Prediction of properties of ammoniaionic liquid fluids using molecular simulation; Collection and modeling of relevant thermophysical properties; Evaluation of the heat and mass transfer performance. Besides, concepts of using ionic liquids as absorbents with ammonia as the refrigerant in various thermodynamic cycles are analyzed and evaluated for applications in the built environment and industry...

Separation and selective production of branched paraffins are among the most important and still challenging processes in the oil and gas industry. Addition of branched hydrocarbons can increase the octane number of a fuel without causing additional environmental concerns. Conversion of linear hydrocarbons into branched ones also improves the performance of lubricants at low temperatures. Zeolites are commonly used for separation of branched hydrocarbons and selective conversion of linear long chain hydrocarbons into shorter branched ones...

This thesis provides an overview of the development of the Pair Configuration to Molecular Activity Coefficient (PAC-MAC) model. PAC-MAC is a computational method to predict thermodynamic miscibility properties of various molecular solutions. Examples of calculated thermodynamic miscibility properties are vapor-liquid equilibrium (VLE) diagrams, mixing free energies and Flory-Huggins χ-interaction parameters. Within PAC-MAC, a unique approach is used by combining the quasi-chemical approximation of Guggenheim with the use of classical force fields. The quasi-chemical approximation is used in foregoing models for miscibility prediction (UNIFAC, COSMO-RS) whereas classical force fields are used in methods preserving the molecular 3D structure (molecular dynamics and Monte Carlo simulations). The combination of the quasi-chemical approximation with classical force fields results in a quick and accurate method containing, besides the force field parameters, only two empirically optimized parameters.

The research presented in this thesis focuses on the use of hydrate slurries in the air conditioning and refrigeration areas. Both experimental and mathematical methods have been used. Hydrate slurries have been suggested as promising cold storage materials that can be used in air conditioning systems due to their high latent heat (193 kJ/kg and 387 kJ/kg for the hydrates studied in this thesis) and positive phase change temperature (12.5 °C and 8.0 °C for the hydrates studied in this thesis). However, large scale industrial applications of hydrate slurries are still very limited. This suggests that more research efforts should be devoted to the demonstration of its advantages.

Monte Carlo simulations were performed in the NPT ensemble and in an expanded version of the Gibbs ensemble. The phase behavior of single-phase hard-sphere chain fluids was determined using NPT ensemble simulations, while the isotropic-nematic phase equilibrium of single-component and binary mixtures of hard-sphere and Lennard-Jones fluids was calculated using expanded Gibbs ensemble simulations. Fluid properties such as packing fractions (hard-sphere fluids), densities and temperatures (Lennard-Jones fluids), order parameters, pressures, and gas solubilities (Henry coefficients) were obtained for chain fluids with different degrees of elongation and flexibility. The study of single-component athermal (hard-sphere) systems showed that larger chain lengths increase the potential for the formation of nematic phases, decreasing monotonically the pressure and packing fractions at which the isotropic-nematic phase transition takes place. However, a maximum in the isotropic-nematic packing fraction difference was observed as a function of chain length. Flexibility reduces the anisotropy of the molecule, which increases the isotropic-nematic transition pressure and decreases the packing fraction difference between both phases. It was shown that the solubility of a single-segment hard-sphere gas molecule does not depend on molecular order, being determined only by the packing fraction of the system. The solubility difference at the isotropic-nematic phase transition is therefore determined by the packing fraction difference at coexistence. The effect of connectivity was studied by considering relative Henry coefficients, defined as the Henry coefficient of the gas in the chain fluid divided by the Henry coefficient of the gas in the monomer fluid at same density. A linear relationship between relative Henry coefficients and packing fraction was obtained. Binary mixtures of hard-sphere chain fluids showed a phase split and fractionation between a nematic phase concentrated in the longest component and an isotropic phase richer in the shortest component. Fractionation is a consequence of the larger tendency to align of the long chains compared to the short chains. This tendency increases with the difference in chain length between both components forming the mixture. A larger fractionation leads to a larger isotropic-nematic packing fraction difference, which was found to be always larger for the mixture than for the constituting pure components. Flexibility of the longest component reduces fractionation, increasing the pressure and packing fraction at which the isotropic-nematic transition takes place for a specific composition of the fluid. As in the case of pure components, a linear relationship between relative Henry coefficients and packing fraction independent of the molecular order was observed. The isotropic-nematic phase equilibria of Lennard-Jones chains was determined for: linear chain molecules with different lengths, a partially-flexible chain molecule, and a binary mixture of linear chains. Increasing chain lengths displaces the isotropic-nematic coexistence of linear Lennard-Jones chain fluids to larger equilibrium densities and lower equilibrium pressures, increasing at the same time the isotropic-nematic density difference. Flexibility reduces the anisotropy of the molecule shifting the isotropic-nematic equilibrium to larger pressures and densities, reducing at the same time the density difference between both phases. In a binary mixture of linear Lennard-Jones chains, fractionation between an isotropic phase richer in the short chains and a nematic phase richer in the long chains was observed. The isotropic-nematic density difference was found to be always larger in the mixture than for pure components, with a maximum at a certain mole fraction. The solubility of a single-segment Lennard-Jones gas molecule was calculated for linear Lennard-Jones chain fluids and for a binary mixture of them. A linear relationship between solubility difference and density difference at coexistence was identified for all linear chain systems. Furthermore, it was shown that the solubility of a Lennard-Jones gas in linear Lennard-Jones chain fluids is independent of the molecular order of the fluid at coexistence. The isotropic-nematic solubility difference in the binary mixture was observed to be always larger in the mixture than for the pure components, with an observed maximum at the mole fraction corresponding to the maximum in the isotropic-nematic density difference. Simulation results for the isotropic-nematic transition were compared to theoretical predictions obtained from the analytical equation of state developed by van Westen et al.. Excellent agreement between simulation and theoretical results were observed. A rescaled Onsager theory was used in the description of hard-sphere systems showing an accurate prediction compared to simulation results. A perturbed theory approach was used in the treatment of Lennard-Jones fluids, whereby orientation dependent attractions were not considered explicitly in the development of the dispersion term. A reliable description of the isotropic-nematic phase equilibrium of Lennard-Jones fluids was obtained using this approach. The simulation data presented in this work represents a large contribution to the liquid crystal phase behavior of chain fluids. The effect of molecular properties on the phase behavior of liquid crystals was determined using linear and partially-flexible molecules with different elongations and flexibility, for purely repulsive (hard-sphere) and soft-attractive (Lennard-Jones) interactions. Predictions obtained from the equation of state of van Westen et al. showed an excellent agreement with simulation data, validating the assumptions made in the development of this theory.

The aim of the study described in this thesis is to obtain a profound understanding of transformations in NCs at the atomic level, by performing molecular simulations for such transformations, and by comparing the simulation results with available experimental high resolution transmission electron microscopy (HRTEM) data to validate the simulations and to reveal underlying physical mechanisms. These transformations include structural and morphological transitions and cation exchange processes in ionic nanocrystals (II-VI and IV-VI semiconductors). The main simulation method used is classical Molecular Dynamics (MD) simulation. First principles density functional theory (DFT) calculations were used to develop empirical force fields that are able to accurately reproduce phase transitions. Using these newly developed force fields, large scaled classical MD simulations were carried out and linked to HRTEM experiments. The partially charged rigid ion model (PCRIM) was chosen for the force fields. This PCRIM approach has a simple functional form with a few number of parameters and has a clear physical meaning for ionic crystals. To simulate cation exchange in colloidal NC systems at the NC/solution interface, we used a combination of all-atom force fields and a coarse-grained model. In Chapter 2, an ab-initio based force field for ZnO is developed within the framework of the PCRIM approach. The values of the partial charges were determined by Bader charge analysis of DFT calculations on various ZnO phases. Beside Coulombic interactions, only short-ranged pairwise interatomic interactions were included. An initial guess of the parameters of the short-ranged pair potentials were first obtained by the lattice inversion method. The parameters were further adjusted by an ab-initio potential surface fitting procedure. The new ZnO force field has a very simple functional form is able to accurately reproduce several important physical properties of ZnO materials. These physical properties include the lattice parameters and phase stability of several ZnO polymorphs, as well as the elastic constants, bulk moduli, phonon dispersion, and melting points of wurtzite ZnO. The transition pressure of the wurtzite-to-rocksalt transition calculated with the force field equals 12.3 GPa, in agreement with experimental measurements and DFT calculations. A wurtzite-to-honeycomb phase transition is predicted at an uniaxial pressure of 8.8 GPa. We found a rational and effective way to derive force fields with simple functional forms for accurate simulations of phase transitions in ionic crystals. In Chapter 3, we developed a transferable force field for CdS-CdSe-PbSPbSe solid systems. The selection of the force field and the fitting procedure are similar to that of the ZnO force field in Chapter 2. The challenges when developing this force field were to maintain the transferability of this force field for four materials (CdS, CdSe, PbS, and PbSe) and to describe their mixed phases. This was solved by assuming that different cations/anions have the same values of the partial charges, and that shortranged interatomic interactions between two cations/anions are the same in different materials. For the mixed phases, DFT calculations of the mixed phases were included in both the training and validation sets. This work is the first step for further simulation studies of these II-VI and IV-VI semiconductor NCs and heteronanocrystals (HNCs). In Chapter 4, a thermally induced morphological and structural transition of CdSe NCs was investigated using MD simulations. In MD simulations, a CdSe nanosphere with the ZB structure transforms into a tetrapodlike morphology at 800 K. In a CdSe tetrapod, four WZ legs attach to the {111} facets of a tetrahedral ZB core. This transformation is achieved by a layer-by-layer slip of the ZB-{111} bilayer. Simulations show that the slips are mediated by the formation of Cd vacancies on the surface of the NCs to overcome the potentially large energy barriers associated with slip. The morphology of the annealed NCs is found to be temperature and size-dependent. An octapod-like morphology is found in NCs with a relatively large NC size and in a certain range of the heating temperature. Surprisingly, nanoscale transformations of CdSe NCs have been directly observed in HRTEM in situ heating experiments. Our findings provide a simple method to modify the morphology of ionic NCs and can potentially be used in the synthesis of branched NCs. The cation exchange process of PbSe/CdSe HNCs has been investigated by HRTEM in situ heating experiments in combination with MD simulations and DFT calculations in Chapter 5. In the HRTEM experiments, we bserved that Cd atoms in PbSe/CdSe nanodumbbells (CdSe rods with one or two PbSe tip(s)) are replaced by Pb atoms. The exchange rate depends on the heating temperature and the amount of Pb atoms present in the system. Sometimes, fully converted PbSe nanodumbells can be observed. MD simulations were performed to investigate the mechanism of this cation exchange process. It was found that the the CdSe domains near the PbSe/CdSe interfaces have significant structural disorder. These findings are in line with the experimental observation that the exchange process proceeds in a layer-by-layer fashion along the WZ-direction. We concluded that cation exchange in PbSe/CdSe HNCs is mediated by the local structural disorder which enables the formation of vacancies and accelerated the motion of cations. In Chapter 6, a coarse-grained psuedoligand model was introduced to simulate cation exchange in PbS colloidal NCs taking into account the cation-solvent interactions. Modelling colloidal NC systems including interactions with the solvent has long been a challenge due to the large system size and long time scales. Here, we incorporated the effects of ligands and solvents into negatively charged large spherical coarse-grained psuedoligands. MD simulations combining coarse-grained and all-atom models can successfully reproduce the cation exchange process in PbS colloidal NCs. Simulations show that the exchange rate and system equilibrium can be controlled by the temperature and by changing ligands. The exchange process is directly related to vacancy formation and the high mobility of Cd ions at the PbS/CdS interface. Our simulations also predict that high-pressure conditions will be beneficial for achieving fast exchange at elevated temperatures. Our coarse-grained model can be easily extended to other systems for the computational investigation of transformations in nanostructures.

The increase in concentrations of greenhouse gases is responsible for global warming over the past few years. A major portion of the emitted greenhouse gases contains carbon dioxide (CO2). The capture of carbon dioxide from the effluent sources, its transport, and storage has been identified as the most promising method to mitigate global warming by reducing the carbon footprint in the atmosphere. Post-combustion CO2 capture processes mainly use chemical solvents like monoethanolamine (MEA) to capture CO2 from flue gas streams. The regeneration of CO2 from the rich solvents is an energy intensive process which decreases the overall efficiency. Other issues like solvent volatility/corrosiveness, toxicity, and solvent costs are critical for choosing the best solvent for post-combustion capture. There is a need for designing new solvents which are less energy intensive, less toxic and corrosive than the ones existing in the industry. Solvent design is extremely challenging, since there are potentially millions of molecules that can be used as a solvent for post-combustion CO2 capture and to find the best solvent is extremely difficult from an experimental point of view. Understanding the thermodynamic processes taking place during chemisorption help in studying the effects of solvents on the absorption/desorption of CO2 in the solvents and their constituent reactions. Ultimately, these aid in developing new solvents that have high capacity to absorb CO2 and low energy penalties to regenerate the CO2 rich solvents. Some of the thermodynamic processes that take place are absorption/desorption from the gas phase, the reaction of CO2 with the aqueous solvent and the diffusion of CO2 molecules into the aqueous solvent mixture. Understanding these processes can help in developing a screening tool that can predict the solubilities, equilibrium constants and conversions, diffusivities of CO2 in different solvents. Molecular simulations can be used in this regard to great effect. Monte Carlo simulations can be used to study equilibrium conversions of the different species in the reacting mixture with great efficiency and accuracy. They can also measure the absorption/desorption of CO2 from the gas phase into the solvent. Molecular Dynamics simulations can help calculate diffusion coefficients of CO2 in the reacting mixture. In this thesis, we develop advanced techniques and methods that can potentially be used to screen large number of solvent molecules and to potentially select the most promising ones. Existing molecular methods are insufficient to efficiently describe the chemisorption and diffusion of CO2 in liquid solvents. We have developed new methods to study the different thermodynamic processes taking place between the CO2 and solvent molecules. The high accuracies and efficiencies of molecular simulations make it an attractive tool in designing new solvents which are more efficient, less toxic and less corrosive than the ones that are available in the market.

In this thesis I aim to contribute to a molecular understanding and -description of the phase behaviour of liquid crystalline materials. In particular, I aim at the development of a molecular-based equation of state (EoS) for describing nematic (only orientationally ordered) liquid crystals (LCs) and their mixtures. Special emphasis is put on the role of intra-molecular flexibility on the liquid crystalline phase behaviour. Also, the solubility of small gases in nematic solvents is studied—an area that could be important for potential applications of LCs as novel solvents in gas-absorption processes. In the first part of this introductory chapter, the reader is provided with some background on the liquid crystalline state of matter, the status of LC research, and common and potential applications of LCs (Sections 1.1 and 1.2). Subsequently, available theories for the nematic state are briefly reviewed, and the foundations of the perturbation methodology which is at the basis of the EoS developed in this work are discussed (Section 1.3). Finally, an outline of the work performed in this thesis is presented (Section 1.4).

Global greenhouse gas emissions, especially of CO2, have been increasing tremendously over the past century. This is known to cause not only an increase of temperature, but also a change in our climate. Along with a shift to renewable sources of energy, Carbon Capture and Storage is necessary to mitigate climate change. Power plants are the largest point source of CO2 emissions and therefore, capture of CO2 from such sources is a must. Post Combustion CO2 Capture (PCCC), and specifically absorption-desorption based technology is the preferred choice of technology for CO2 capture from flue gases. It has been extensively used in the oil and gas industry for gas treatment. Its application for CO2 capture from flue gases is not straightforward, mainly due to different flue gas composition and operating conditions. Other aspects such as solvent degradation, solvent emissions and corrosion become even more critical. In Chapter 1, the state-of-the-art in PCCC is explained with further details on the current knowledge and understanding of these aspects. In Chapter 2, the aspect of solvent ageing is studied over two test campaigns in a CO2 capture pilot plant using 30 wt.% MEA. Solvent degradation occurs via thermal and oxidative routes, with the latter being more prominent. Ammonia is known to be a major oxidative degradation product, while the remaining degradation products are known to be corrosive. Therefore, solvent degradation is expected to have a significant impact on the corrosion in the plant and the resulting emissions of ammonia. The link between these three parameters was confirmed using online monitoring probes. Moreover, an autocatalytic behaviour was observed resulting in an rapid increase of the solvent metal content and ammonia emissions. The solvent iron content was above 500 mg/kg, while the ammonia emissions exceeded 150 mg/m3 STP (STP; 0°C and 101.325 kPa). By correlating the process conditions to the underlying degradation and corrosion mechanisms, online monitoring tools can be used to assess and manage the lifetime of the solvent. Even if the state of the solvent is kept in check by reclaiming methods, there could be instances where ammonia emissions could increase. Therefore, it is necessary to have an end of pipe countermeasure for such emissions. Chapter 3 presents the results from a test using an acid wash scrubber for ammonia emissions in a pilot plant test campaign. Several parametric tests were conducted in order to test the efficiency of the acid wash. Moreover, the ammonia concentration in the gas inlet to the acid wash was increased artificially (~150 mg/m3 STP). The acid wash scrubber reduced ammonia emission to very low levels, below 5 mg/m3 STP and mostly below 1 mg/m3 STP. Moreover, the MEA content was also reduced to mostly below 1 mg/m3 STP. A comparison between a model made in Aspen Plus and the experimental results showed good agreement, with deviations only at pH above 5 to 6. Aerosol based emissions are known to be a concern in PCCC. In Chapter 4, the impact of flue gas particles such as soot and sulphuric acid aerosol droplets on solvent emissions was studied in a mobile CO2 capture plant. These tests confirmed that solvent emissions can be in the order of grams per m3 STP, which is several orders of magnitude higher than volatile emissions. The number concentration of these particles had a direct relation to the extent of emissions. Particle number concentrations in the range of 107-108 per cm3 led to emissions of MEA in the range of 600-1200 mg/m3 STP. In Chapter 5, further tests were performed on the same setup in order to assess the impact of operating conditions of the CO2 capture plant on aerosol based emissions. Increasing the temperature of the lean solvent resulted in lowering of the aerosol based emissions. However, the total solvent emission increased as a result of increased volatile emissions. Aerosol based emissions were observed also for AMP-Pz as the capture solvent. The pH of the lean solvent was decreased by lowering the stripper temperature and thereby, changing the CO2 loading of the solvent. This resulted in an increase in the aerosol based emissions as the activity of the amine increased in the solvent. As the CO2 content in the flue gas was reduced from 12.7 vol.% to 0.7 vol.%, a maximum in the emissions was observed between 6 and 4 vol.%. When a mixture of a slow reacting volatile amine, AMP, with a fast reacting non-volatile, taurate, was used, no aerosol based emissions were observed. This led to the important conclusion that in reactive absorption, along with supersaturation and particle number concentration, the reactivity of the amine plays an important role in aerosol based emissions. A Brownian Demister Unit (BDU), consisting of multiple polypropylene fibre elements, can be potentially used as a countermeasure for aerosol based emissions. This was tested in a pilot plant campaign using MEA and is discussed in Chapter 6. The BDU reduced emissions from about 85–180 mg/m3 STP to about 1–4 mg/m3 STP. A water wash was found to be effective against vapour based emissions, while the BDU was effective against aerosol based emissions. A BDU is not effective against ammonia emissions, as they are present in the vapour form. From the measured nitrosamines, NDELA was found to be in the solvent in the order of 2000 ng/ml, while in the water wash it was ca. 1 ng/ml. Gas phase nitrosamine concentrations were in the range of tens of ng/m3 STP. The BDU results in a significant additional pressure drop of ca. 50 mbar, for the configuration and type of BDU used here. This translated to an additional consumption of electricity by the blower in the range 26–52%. A system containing three distinct phases, gas, liquid and aerosol droplets, are complex to understand and model. In Chapter 7, a methodology is presented with which such a complicated system can be modelled in commercially available software such as Aspen Plus. The mass and energy exchanges are split into two distinct interactions, gas-liquid and gas-aerosol. Aerosol droplets are considered to be as bulk liquid without any direct interaction with the solvent. The different parameters that were varied were the CO2 concentration in the flue gas, temperature of the lean solvent and CO2 loading of the lean solvent. The resulting trends were in good agreement with the experimental results presented in Chapter 5. The model did not predict a maximum in the emissions as the CO2 content in the flue gas was varied. Although absorption desorption based process for CO2 capture is well known, several operating issues needs to be addressed for its application in PCCC, as evident in this thesis. It is important to monitor the degradation of the solvent and deploy appropriate methods at the right time, to minimize its detrimental effect on the corrosion of the plant and avoid high emissions of ammonia. Aerosol based emissions in a PCCC process is a serious issue. The experimental results, proposed mechanism and modelling methodology will enable the design of appropriate counter-measures against aerosol based emissions. It is recommended to devise appropriate strategy and innovative solutions based on the understanding of the various operational aspects of absorption-desorption based PCCC as presented in this thesis. This will increase the confidence level in the technology and lead to its successful deployment for mitigating climate change in the short term.

A novel class of materials, i.e. Metal-Organic Frameworks (MOFs), has successfully been developed that is extremely suited for application in heat pumps and chillers. They have a superior performance over commercial sorbents and may potentially contribute to considerable energy savings worldwide. Globally about 33 % of the energy consumption is used for heating and cooling of e.g. houses and buildings. Adsorption driven heat pumps and chillers are very well suited to reduce this energy consumption and can even use low-grade waste heat or sustainable solar energy in combination with environmentally benign working fluids (e.g. water). MOFs are porous crystalline materials built up from inorganic clusters connected by organic ligands in 1, 2 or 3 dimensions, and display a rich variety of topologies and can be functionalized in many different ways. They offer the materials scientist an outstanding platform to design new materials with superior properties. The described research has identified MOFs with sufficient stability against water, that show the desired adsorption behavior of water. These MOF-water pairs possess higher energy efficiency and working capacity than benchmark materials and may operate with a lower driving temperature. The selected MOFs can be coated (without binder) directly on heat-exchanger surfaces for a fast response. In short, there is a bright future for the application of MOFs in adsorption heat pumps and chillers with a large energy savings potential.

In this thesis, we investigated the potential of physical ILs for CO2 capture at pre-combustion and natural gas sweetening conditions. The performance of ILs with respect to conventional solvents is assessed in terms of gas solubilities and selectivities. The work discussed in this thesis consists of two parts. The first part deals with experimental determination of gas solubilities in ILs, while in the second part molecular simulations are used to predict gas solubilities in physical solvents. In Chapter 2, a comprehensive review of CO2 capture with ILs is presented. In Chapter 3, the experimental results of pure CO2 and CH4 solubilities in many different kinds of ILs are reported. Ideal CO2/CH4 selectivities are derived from the experimental data and a comparison with conventional solvents is provided. In Chapter 4, the experimental results on the solubility of CO2/CH4 gas mixtures in ILs is discussed. Real CO2/CH4 selectivities are derived from this mixed-gas solubility data. In Chapter 5, Monte Carlo (MC) molecular simulations are used to predict the solubility of natural gas components in ILs and Selexol. In Chapter 6, MC simulations are used to compute the bubble points of CO2/CH4 gas mixtures in ILs. In Chapter 7, MC simulations are used to compute the solubility of the pre-combustion gases CO2, CH4, CO, H2, N2 and H2S in an IL. Separation selectivities relevant for the pre-combustion process are derived from the MC data and a comparison with experimental data is provided. In Chapter 8, a novel Monte Carlo method is developed to study the reactions of CO2 with aqueous monoethanolamine (MEA). The so-called Reaction Ensemble Monte Carlo method in combination with the Continuous Fractional Component technique (RxMC/CFC) is used to compute the equilibrium speciation of all relevant species formed during the chemisorption process of CO2 with aqueous MEA solutions. The computed speciation results are compared with available experimental data. Finally, Chapter 9 provides a detailed comparison of gas solubilities in ILs with respect to the conventional solvents Selexol, Purisol, Rectisol, propylene carbonate, and sulfolane.

In this thesis the technical and economic performance of compression-resorption heat pumps has been investigated. The main objective of this thesis was to improve the performance and reduce the investment costs of compression-resorption heat pumps applied in process industry. A model that is able to capture the performance of most heat pumps based on the Carnot and Lorentz COP by only knowing the temperature driving forces and estimating the compressor isentropic efficiency has been developed. By including an economic model for the investment costs of compressors and heat exchangers and including costs for electricity and heating, the model was capable of predicting the payback period for conventional systems. For larger systems of 10 MWth applied to a standard distillation column where compression-resorption heat pumps could use 50% of the lift as temperature glide, predictions show that such a heat pump will have a minimum payback period of approximately 3 years, while systems with a capacity of 2.8 MWth require about 5 years, clearly demonstrating the effect of system size. A more detailed model was implemented to investigate the performance of a compression-resorption heat pump applied to distillation processes in the Dutch industry. In this case, the heat pump could only make use of the temperature glide available in reboilers and condensers, which is much smaller than using the temperature glide in the columns. This is limiting the possible performance of the heat pump. The results showed that for most cases the performance was such that both energetic and cost advantages can be obtained with the implementation of such heat pumps. Further the model showed that in case of wet compression the inlet of the absorber should be a saturated vapor to reach maximum efficiency. To increase the performance of the compression-resorption heat pump and decrease the investment cost, the performance of mini-channel heat exchangers operating with a two-phase ammonia-water mixture was experimentally investigated. Initial research focused on the absorber performance in a mini channel annulus with a hydraulic diameter of 0.4 mm and a length of 0.8 m. Absorption side heat transfer coefficients in the range of 1000 to 10000 W m-2 K-1 were obtained for mass fluxes between 75 and 350 kg m-2 s-1 while the average vapor quality ranged from 0.2 to 0.6. The pressure drop varied between 0.2 and 1.6 bar under the given conditions and correlated with literature models within +25% / -25%. The heat transferred from shell to tube side ranged between 50 and 300 W. At low vapor qualities the heat transfer coefficient increases sharply between mass fluxes of 100 and 175 kg m-2 s-1. This behavior was less profound during experiments at higher vapor qualities. The tube side of the same heat exchanger was also investigated using the ammonia-water mixture during a desorption process. The tube side had a diameter of 1.1 mm and a length of 0.8 m. The obtained desorption side heat transfer coefficients lie in the range between 5500 and 10500 W m-2 K-1. The mass fluxes ranged from 150 to 300 kg m-2 s-1 and the average vapor quality ranged from 0.2 to 0.5. The heat transfer performance was well predicted by a model from literature after one of the empirical constants was adjusted. Due to ongoing deposition of debris in and in front of the channel the pressure drop increased over time such that a clear trend in pressure drop as function of mass flux and vapor quality could not be derived. The heat transferred in the heat exchanger under the given conditions ranged from 50 to 250 W. 116 One of the problems with mini-channel heat exchangers is upscaling. One tube can deliver up to 250 W, so for a system of 10 MWth 40 000 tubes are required. Further problems arise with flow distribution: one wants to distribute the flow such that each tube gets the same amount of liquid and vapor. As an intermediate step a heat exchanger was designed comprising of 116 tubes with a diameter of 0.5 mm. The shell side has a hydraulic diameter of 1.8 mm and an inner diameter of 21 mm. The flow distribution for single phase flows was first analyzed using a model which could capture the effects of contraction and expansions in the distributor and a similar design of the collector. Modeling the pressure drop in the shell side of a heat exchanger with the given geometry is complex. To simplify the approach, the Chilton-Colburn method has been chosen to be able to predict the friction factor in the shell side. The results from the model using water as working fluid showed that the flow is distributed evenly over all the tubes, deviations from the average were smaller than 0.1%. The heat transfer coefficients obtained with water as the working fluid on both sides lie between 750 and 850 W m-2 K-1 for mass fluxes between 5 and 30 kg m-2 s-1. The heat transfer model predicts 800 W m-2 K-1 for all heat transfer experiments, and the variation between the experiments can be merely seen as the deviations possible in the measurements and data reduction. Heat transfer experiments using the ammonia-water mixture have been conducted on this heat exchanger. When using this mixture in the shell side of the heat exchanger, it becomes clear that the heat transfer performance is lower compared to the same unit working with water in both shell and tube side. The shell side heat transfer coefficient is the limiting factor during the experiments. The overall heat transfer coefficient during the experiments ranged from 150 to 600 W m-2 K-1 for shell side mass fluxes from 2.7 to 8.1 kg m-2 s-1. The overall heat transfer coefficient now shows an increasing trend with mass flux, while during the water experiments the trend of increasing heat transfer coefficient with increasing mass flux remained within the error of the measurements. During all measurements the flow condition in the tube side was in all cases laminar flow. The ammonia-water mixture has also been put as the working fluid in the tube side of the heat exchanger in such a way that both shell and tube sides operate within the two-phase region. The overall heat transfer coefficients ranged between 300 and 1800 W m-2 K-1. The maximum attainable heat transfer coefficient increased because the mass flux could be increased from 8.1 kg m-2 s-1 to 16 kg m-2 s-1. Again the trend of increasing heat transfer coefficient with increasing mass flux was obtained. By increasing the inlet temperature of the absorber, the average vapor quality increases and the heat transfer coefficients also. Pressure drop ranged from 0.01 bar to 0.3 bar for tube side mass fluxes between 25 and 200 kg m-2 s-1, while the pressure drop on the shell side varied between 0.04 to 0.5 bar at mass fluxes between 2 and 16 kg m-2 s-1. During all two-phase measurements oscillations in flow rate and pressure drop have been identified, while they were stable during single phase flow. These oscillations are most likely caused by Taylor- and hydro-dynamic instabilities. During the desorption processes in the tube side of the heat exchangers the oscillations were larger, however, experiments

The aim of this thesis is to study multicomponent diffusion in liquids using Molecular Dynamics (MD) simulations. Diffusion plays an important role in mass transport processes. In binary systems, mass transfer processes have been studied extensively using both experiments and molecular simulations. From a practical point of view, systems consisting more than two components are more interesting. However, experimental and simulation data on transport diffusion for such systems are scarce. Therefore, a more detailed knowledge on mass transfer in multicomponent systems is required. The presence of multiple components in a system introduces difficulties in studying diffusion in experiments. Investigating the concentration dependence of diffusion coefficients seriously increases the required experimental effort. In this thesis, we will use MD simulation based on classical force fields to study multicomponent diffusion in liquids. Diffusion can be described using both Fick and Maxwell- Stefan (MS) diffusion coefficients. Experiments provide Fick diffusion coefficients while simulations usually provide MS diffusion coefficients. Fick and MS diffusivities are related via the matrix of thermodynamic factors. A brief survey on methods for studying liquid diffusion and their limitations is presented in chapter 1 In chapter 2, we study the diffusion in the ternary system n-hexane-cyclohexanetoluene. The existing models for predicting MS diffusivities at finite concentrations (i:e: the Vignes equation) as well as the predictions at infinite dilution (i:e: predictions of Ðxk!1 i j using the so-called WK, KT, VKB, DKB and RS models) are tested using MD simulations. We find that (1) the Vignes equation only results in reasonable predictions for MS diffusivities yielding differences of 13% compared to the actual diffusion coefficients; (2) the best predictive model (the KT model) for calculating MS diffusivities at infinite dilution results in differences of 8% compared to the actual diffusion coefficients. It is important to note that the differences of 8% can be a coincidence since KT model is empirical and does not have a theoretical basis. This limitation makes KT model unreliable for other systems. To overcome the difficulties in predicting ternary MS diffusivities at infinite dilution (i:e: Ðxk!1 i j ), we derive the so-called LBV model based on the Onsager relations. MS diffusivities at infinite dilution can be expressed in terms of binary and pure component self-diffusivities and integrals over velocity cross-correlation functions. By neglecting the latter terms, we obtain the LBV model. In chapter 3, the LBV model is validated for WCA fluids and the ternary systems n-hexane-cyclohexane-toluene and methanol-ethanol-water. We find that: (1) for ideal mixtures i:e: the WCA system, as well as the n-hexane-cyclohexane-toluene system, the LBV model is accurate and superior compared to the existing models for predicting ternary MS diffusivities at infinite dilution (i:e: the WK, KT, VKB, DKB and RS models); (2) in mixtures containing associating components, i:e: the ethanol-methanol-water system, the LBV model indicates that in this system the integrals over velocity cross-correlation functions are important and cannot be neglected. Moreover, the LBV model provides an explanation why the MS diffusivity describing the friction between adsorbed components in a porous material is usually very large. In chapter 4, we focus on describing the values of MS diffusivities at finite concentration. A multicomponent Darken model for describing the concentration dependence of MS diffusivities is derived from linear response theory and the Onsager relations. In addition, a predictive model for the required self-diffusivities in the mixture is proposed leading to the so-called predictive Darken-LBV model. We compare our novel models to the existing generalized Vignes equation and the generalized Darken equation. Two systems are considered: (1) ternary and quaternary WCA systems; (2) the ternary system n-hexane-cyclohexane-toluene. Our results show that in all studied systems, our predictive Darken-LBV equation describes the concentration dependence better than the existing models. The physically-based Darken-LBV model provides a sound and robust framework for prediction of MS diffusion coefficients in multicomponent mixtures. In chapter 5, diffusion in more complex ionic liquid (IL) systems are investigated. Previous research reported in literature has largely focused on self-diffusion in ILs. For practical applications, mutual (transport) diffusion is by far more important than self-diffusion. We compute the MS diffusivities in binary systems containing 1-alkyl- 3- methylimidazolium chloride (CnmimCl), water and/or dimethyl sulfoxide (DMSO). The dependence of MS diffusivities on mixture composition are investigated. Our results show that: (1) For solutions of ILs in water and DMSO, self-diffusivities decrease strongly with increasing IL concentration. For the system DMSO-IL, an exponential decay is observed for this; (2) For both water-IL and DMSO-IL, MS diffusivities vary by a factor of 10 within the concentration range which is still significantly smaller than the variation of the self diffusivities; (3) The MS diffusivities of the investigated IL are almost independent of the alkyl chain length; (4) ILs stay in a form of isolated ions in CnmimCl-H2O mixtures, however, dissociation into ions is much less observed in CnmimCl-DMSO systems. This has a large effect on the concentration dependence of MS diffusivities; (5) The LBV model for predicting the MS diffusivity at infinite dilution described in chapter 3 suggests that velocity cross-correlation functions in ionic liquids cannot be neglected and that the dissociation of ILs into ion pairs has a very strong influence on diffusion. In experiments, Fick diffusion coefficients are measured and molecular simulation usually provides MS diffusivities. These approaches are related via the matrix of thermodynamic factors which is usually known only with large uncertainties. This leaves a gap between theory and application. In chapter 6, we introduce a consistent and efficient framework for the determination of Fick diffusivities in liquid mixtures directly from equilibrium MD simulations by calculating both the thermodynamic factor and the MS diffusivity. This provides the missing step to extract Fick diffusion coefficients directly from equilibrium MD simulations. The computed Fick diffusivities of acetone-methanol and acetone-tetrachloromethane mixtures are in excellent agreement with experimental values. The suggested framework thus provides an efficient route to model diffusion in liquids based on a consistent molecular picture. In chapter 7, we validate our method for computing Fick diffusivities using equilibrium MD simulations for the ternary system chloroform - acetone - methanol. Even though a simple molecular model is used (i:e: rigid molecules that interact by Lennard-Jones and electrostatic interactions), the computed thermodynamic factors are in close agreement with experiments. Validation data for diffusion coefficients is only available for two binary sub-systems. In these binary systems, MD results and experiments do agree well. For the ternary system, the computed thermodynamic factors using Molecular Dynamics simulation are in excellent agreement with experimental data and better than the ones obtained from COSMO-SAC calculations. Therefore, we expect that the computed Fick diffusivities should also be comparable with experiments. Our results suggest that the presented approach allows for an efficient and consistent prediction of multicomponent Fick diffusion coefficients from MD simulations. Now, a tool for guiding experiments and interpreting multicomponent mass transfer is available.

The term carbon dioxide capture and storage (CCS) refers to a range of technologies that can reduce CO2 emissions from fossil fuels enabling the continued use of this fuel type without compromising the security of electricity supply. The technologies applicable to CCS differ in many key aspects; the stage of the electricity generation process at which the CO2 is captured, the CO2 capture process, efficiency, availability and matureness of the technology. The integration of these technologies into power plants results in a reduction in power generation efficiency, which remains one of the major issues for the commercial implementation of CCS. Among the possible technologies, the focus of this thesis is on post-combustion capture as it is a known technology, is readily available and it can be retrofitted to existing power plants. This thesis is concerned with the development of new carbon capture processes that require less energy for CO2 separation and are, at the same time, more environmentally friendly. Prior to the development of any new process, the current state of the art needs to be analysed and updated in order to set realistic targets for the new technology and benchmark the potential of the newly developed processes. Therefore, part of the work of this thesis is a thorough benchmarking exercise in which updated baselines for the performance of conventional post-combustion capture processes are given. The new process concepts developed in this thesis are based on the combination of enhanced absorption and enhanced desorption, two effects encountered in capture processes that are based on precipitating amino acid solvents. For this purpose, the conceptual design methodology has been followed with a specific target of energy reduction set to (at least) 30% of a conventional MEA process.

The adsorption of water on hydrophobic zeolites such as silicalite and on hydrophilic MOF (metal-organic framework), Cu-BTC, is completely different, as described in chapters 2 and 4. While in hydrophobic materials water adsorption isotherms are very steep and difficult to measure, both experimentally and by simulation, in hydrophilic materials water adsorbs easily and its isotherms are similar to the isotherms of other molecules. The key property to understand these differences is the dipole moment of water. Water molecules prefer to stay in a bulk water phase rather than adsorbing in a microporous material. In bulk liquid water, molecules interact strongly, forming clusters via hydrogen bonds. Inside a hydrophobic zeolite pore, the formation of water clusters is restricted by the geometry of the pores. Only at high pressures water molecules are forced to enter the zeolite pores. When a few water molecules are adsorbed, new molecules adsorb in layers close to the already adsorbed molecules. This is the reason why the adsorption isotherm of water in hydrophobic zeolites is very steep. The adsorption properties of water in zeolites are difficult to measure both experimentally and by molecular simulations. Experiments are complicated by the fact that water is adsorbed at defects. Therefore, the inflection point in the isotherm of water is very sensitive to defects, as well as to the structural crystallographic positions of the zeolite framework atoms (in addition to pore blockage/collapse etc.). There are only a few water models that are suitably calibrated for studying water adsorption in zeolites. The Tip5pEw model is calibrated using the Ewald summation and reproduces the bulk properties of water properly. Therefore, it is a suitable candidate to describe water in a periodic porous environment, even though there is still much uncertainty in the proper values of the partial charges of the zeolite framework atoms. The dipole moment of water results in behavior that is completely different from other molecules with similar size but without dipole moment, so the partial charge of the zeolite atoms is a critical parameter that has to be chosen carefully. The adsorption of water is also very sensitive to small changes in the precise location of the zeolite atoms. We provided evidence that this sensitivity is directly related to the coupling of the dipole of the water molecules with the electric field induced by the zeolite. Therefore, one has to be cautious when computing the properties of water and highly polar molecules in these hydrophobic structures. The inclusion of framework flexibility considerably increases the required simulation time, and the error bars of the computed points of the isotherm are much larger than in the case of a rigid structure. It would be desirable to reproduce the experimental isotherm quantitatively, so further refinement of the force field parameters is needed. Fitting these parameters in the case of a flexible framework is quite time consuming, and in the rigid case it resulted to be impossible. Contrarily to hydrophobic zeolites, water adsorbs easily in the hydrophilic Cu-BTC MOF. Cu-BTC contains pores with open metal centers which interact strongly with water. Therefore, the adsorption isotherm of water in Cu-BTC is not as steep as in the case of zeolites, but linear up to relatively large loadings. Fitting the force field to reproduce the experimental adsorption isotherm was straightforward. It was only necessary to modify one decisive parameter of the interactions, i.e. the charge of the metal center (and the rest of the atom charges are scaled accordingly in order to keep the structure charge neutral). The strong electrostatic interaction of water with the metal centers is also responsible for the special behavior of water in Cu-BTC, compared to other molecules without dipole moment. At low loadings, water is preferentially adsorbed at the metal centers. Most of the molecules studied in Cu-BTC, such as alkanes, nitrogen, and carbon dioxide, prefer to adsorb at the side pockets of the structure. This property may be exploited for the separation of components from water. It is interesting to note that the qualitative behavior of water was the same at all the partial charges assigned to the structure during the study. Despite all the difficulties that occur when describing the adsorption of polar molecules in hydrophobic porous materials, we still can use suitable force fields to obtain a qualitative description of adsorption. In chapter 3 we studied the separation performance of water and alcohol mixtures in the hydrophobic, pure siliceous zeolite DDR. This is an example of a separation based on differences in diffusion coefficients rather than in adsorption. Adsorption isotherms for water, methanol, and ethanol on all-silica DDR were experimentally measured by single-component vapor-phase adsorption and calculated by GCMC simulations. The measured alcohol adsorption can be described by a single-site Langmuir adsorption isotherm. The Monte Carlo (MC) simulations were able to qualitatively reproduce the adsorption behavior of the experimental isotherms. The adsorption of water is of type II. The water loading is under predicted by the calculations at pressures up to 2.5kPa. The calculated saturation loadings of all the components considered are larger than the experimental values, although they are comparable to saturation loadings on LTA-type zeolites. The molecular models and force field parameters were taken directly from literature without further adjustment. Nevertheless, the order of magnitude and the shape of the pure-component isotherms give a good resemblance of the experimental data. Mixture adsorption isotherms and diffusivities were calculated and compared with permeation data measured under pervaporation conditions and using specific models. The calculated mixture isotherms show that the loading of both alcohols and water at constant partial fugacity increases as compared to pure-component adsorption. Moreover, the shape of the water isotherm changes from type IV to type I. The decrease in water and ethanol permeance in the mixture as compared to pure-component permeation is not caused by competitive adsorption. The increase in loading in mixture adsorption is significantly more profound for water than for alcohols, but this does not lead to adsorption selectivity for water at the feed conditions of the pervaporation experiments. Therefore, we studied the diffusion of molecules in DDR to find the influence of the dynamics in the separation of water and alcohols. The self-diffusivities calculated by Molecular Dynamics (MD) simulations showed that the water diffusivity is at least 1-3 orders of magnitude higher than the diffusivity of the alcohols. Although component fluxes calculated from the simulation data over predict the experimentally observed values by one order of magnitude, the permeate composition corresponds well with experimental data. The selective water transport through DDR-type zeolite membranes can, therefore, be explained by the higher diffusivity of water respect to alcohols in the hydrophobic DDR-type zeolite. In chapter 5 we studied with more detail the adsorption properties of the MOF Cu-BTC. Cu-BTC consists of two types of cages. One of the cages is commensurate with small molecules and the other is capable of adsorbing larger molecules. This characteristic may induce strong selectivity for mixtures. Understanding separation selectivity requires a proper description of the adsorption behavior of Cu-BTC, which was provided in this chapter. We also described the properties of the structure when as-synthesized water is removed. The observed negative thermal expansion for Cu-BTC has important implications for adsorption, because of the close match between small molecules and the small pockets. In chapter 6 we investigated the adsorption behavior of the main components of natural gas in Cu-BTC and IRMOF-1 using Monte Carlo Simulations. We computed adsorption isotherms at 298 K for pure components and mixtures, and analyzed the preferential adsorption sites on these two MOFs. The detailed study of the sitting of the molecules in both structures provided an explanation for the high adsorption capacity of IRMOF-1 and for the high adsorption selectivity towards carbon dioxide of Cu-BTC. On the basis of our observations, IRMOF-1 seems a good material for the storage of the different components of natural gas, whereas Cu-BTC could be a promising material for their separation. In chapter 7 we analyzed the reasons for separation of xylene isomers found experimentally in the MOF MIL-47. Both experiments and molecular simulations show a large adsorption selectivity of xylene isomers in MIL-47. Our simulations show that this selectivity is due to differences in the packing of xylene isomers. The Ideal Adsorption Solution Theory (IAST) correctly described the binary and ternary mixture adsorption isotherms of xylene isomers in MIL-47. The selectivity factors obtained at low temperatures are larger than the experimental ones. This may suggest that the experimental selectivity of xylenes in MIL-47 can still be improved. In our simulations we use pure crystals, while real MIL-47 crystals may contain defects that may reduce their separation efficiency. In addition, the saturation loading found in the simulations might be difficult to achieve experimentally.

The aim of this thesis is to study thermodynamic properties of nanocrystals (NCs) capped by organic ligands using molecular simulations. Nanocrystals are metallic or semiconductor crystallites of 2–10 nm size, consisting of hundreds to thousands of atoms. Due to their small size, they have unique properties that make them promising for various applications. Organic ligands are crucial for synthesis, stability and surface functionalization of nanocrystals. These ligands are, typically, linear molecules with a specific headgroup that binds to the NC and a hydrocarbon tail pointing outwards. Knowledge of microscopic properties such as interactions of capped NCs with each other and with the surrounding is essential for integration of NCs into novel materials and devices, especially when these devices are created by self-assembly…

Zeolites are microporous materials with pores that have about the same size as small molecules like water or benzene. They are important catalysts in the petrochemical industry, for example for catalytic cracking, and isomerization- and alkylation reactions. This thesis deals with molecular aspects of the application of zeolite catalysis for the production of cumene via benzene alkylation with propene using Mordenite (MOR-type zeolite). We will use classical force field based molecular simulations to study properties at the atomistic scale. In Chapter 1, an overview is given of the various properties of zeolites, as well as the simulation methods to study the adsorption and diffusion of guest molecules adsorbed in zeolites. In Chapter 2, it is shown that the adsorption and diffusion properties of alkanes in MOR strongly depend on the distribution of non-framework Al atoms. A model is presented to accurately describe this effect. A new method to efficiently compute the heat of adsorption for guest molecules adsorbed in zeolites with non-framework cations is presented in Chapter 3. It is found that this method is much more efficient than the existing methods. Furthermore, it is shown that for the simulation of polar molecules in zeolites, the Ewald summation is superior compared to the recently proposed Wolf-method. In Chapter 4, we study the adsorption selectivity of benzene/propene mixtures in various zeolites, and we evaluate their use as a potential catalysts for benzene alkylation. In Chapter 5, we propose an efficient method to accurately compute pore size distributions. This method is extensively used in Chapter 6, in which we study the dealumination mechanism of MOR-type zeolite using Kinetic Monte Carlo simulations. The results of these dealumination simulations are in very good agreement with various experiments. It is found that the pore size of the main channels is significantly increased. Finally, we propose several criteria for potential catalysts for the production of cumene from benzene and propene.

Granular materials such as sand have both liquid-like and solid-like properties similar to both liquids and solids. Dry sand in an hour-glass can flow just like water, while sand in a sand castle closely resembles a solid. Because of these interesting properties granular matter has received much attention from numerous physicists. Part of the research on these materials focuses on the statistics of contact forces between individual particles and how these statistics can be used to understand and predict material properties. Contact forces in granular materials are organized in so-called forces networks. A common quantity to characterize these force variations is the probability distribution of the contact force, P(f). A long-standing issue is the asymptotic behavior of this distribution. In particular, one discusses whether the tail of P(f) is exponential, Gaussian, or has a different form. Furthermore, its relation with material and system properties is unclear. In view of the technical difficulty to measure contact forces, especially in the bulk of the material, we used computer simulations in the so-called force network ensemble of Snoeijer et al. (Phys. Rev. Lett., 2004, 92, 054302). Unfortunately, the estimation of P(f) for large contact forces f is inefficient. The reason is that by far the largest fraction of generated force networks contains only small forces. For unambiguous conclusions on the asymptotic behaviour of P(f), extremely long (of the order of years or even centuries) computer calculations are needed. To obtain better statistics for large contact forces, we developed an umbrella sampling method for the force network ensemble. In Chapter 1, an overview is given of the different methods to study the statistics of force networks. We also explained the umbrella sampling method that we developed to obtain excellent statistics for large forces. In Chapter 2, we applied this method to study the tail of the force distribution P(f) for different systems. The average number of contacts of a particle and the packing configuration are shown not to be important for the asymptotic behavior of P(f). Only the dimensionality of the system has a significant influence: P(f)~exp[-c f a] with a?2.0 for two-dimensional systems, a?1.7 for three-dimensional systems and a?1.4 for four-dimensional systems. In Chapter 3, a possible explanation is presented for the Gaussian decay of large contact forces in two-dimensional systems. It was found that mechanical balance on each particle is essential for the tail of the contact force distribution, which throws serious doubts on the statement that exponential statistics are a generic property of static granular materials. In Chapter 4, we focused on several details of the contact forces and their distribution. We also investigated how well the force network ensemble describes systems with “real” interactions.