Your browser does not support JavaScript!

Home    Μελέτη ρόφησης υδρογόνου σε νανοδομημένα ανθρακικά υλικά, με συνδυασμό πειραματικών μεθόδων και προσομοίωσης Monte Carlo  

Results - Details

Add to Basket
[Add to Basket]
Identifier 000374595
Title Μελέτη ρόφησης υδρογόνου σε νανοδομημένα ανθρακικά υλικά, με συνδυασμό πειραματικών μεθόδων και προσομοίωσης Monte Carlo
Alternative Title Study of hydrogen adsorption in nano-structured carbon materials, with a combination of experimental methods and Monte Carlo simulations
Author Γκότζιας, Αναστάσιος Δημ.
Thesis advisor Φρουδάκης, Γεώργιος
Στεριώτης, Θεόδωρος
Θεοδώρου, Δώρος
Abstract Sorption processes in porous materials are of high importance for chemical technology since they are often used to replace energy-intensive or environmentally noxious separation processes such as distillation, extraction etc. In addition, porous materials are widely used as gas stores (e.g. for H2, CO2, CH4), catalysts or catalyst substrates, sensors and membranes. They also comprise the basis of radically new technologies (e.g. depollution, energy storage, new reactions etc.), mainly due to their unique physicochemical properties which moreover can be tuned at will. In the last decade, tremendous efforts have been devoted to the research and development of materials that can hold sufficient quantities of H2 (in terms of gravimetric and volumetric densities) and, at the same time, possess suitable thermodynamic and kinetic properties, in an attempt to contribute to the establishment of hydrogen as energy carrier. Porous materials have been shown to be quite important candidate hydrogen stores and as a result a wide range of structures have been explored. The materials that appear to have significant advantages and have thus received much attention are carbon-based porous solids, in view of their high surface areas, low cost, available bulk synthesis routes, good recycling characteristics, low densities, wide diversities of pore structures, reasonably good chemical stability, and amenability to synthesize a wide range of variants by post-treatment/functionalisation routes. Although such types of materials have been studied extensively the details of the respective storage mechanisms are still poorly understood, and as such they are subjects of intense research worldwide. The mechanism of a static or a dynamic process taking place within the pores of a solid is significantly affected by the topology - geometry of the porous system and the physicochemical properties of the surface, and as a result the characterisation of porous materials is based on the study of a range of parameters that are connected (directly or indirectly) with the mechanism of each process. The apparent "random" nature of porous materials requires the development of computational models as a first step towards the enhanced understanding of the phenomena taking place within the pores. In this respect a great number of approaches referring to different length scales of the porous structure have been reported up to now. The detailed description of the sorption phenomena requires the study of the respective mechanism at the pore level and as such the most commonly used model is that of single, independent pores. Although this approach overlooks the pore connectivity, a critical parameter for dynamic phenomena (diffusion, permeability etc.), it is considered adequate for the description of equilibrium processes such as sorption. Nevertheless the typical models of unstructured, homogeneous planes or cylinders do not take into consideration key aspects for sorption such as the surface roughness and curvature, the possible structural defects as well as any kind of heterogeneity resulting from the tilt of the pore walls, the chemical composition of the surface, the presence of surface functional groups etc. The present work has been aiming at the thorough investigation of the effect of the energetic and structural heterogeneity of porous carbon structures on their sorption properties by studying the potential fields within single pores and the simulation of sorption therein by adopting statistical mechanics methodologies such as grandcanonical Monte Carlo. The main focus has been the study of Η2 (and D2) due to its particular physicochemical properties. Apart from the apparent technological interest (hydrogen economy, isotopes separation etc.), H2 is an ideal probe for the characterisation of porous materials based on sorption or temperature programmed desorption (TPD) data. H2 molecules have a very small size (giving rise to quantum behaviour at low temperatures), are supercritical at usual pressures/temperatures and are practically inert at cryogenic temperatures. More specifically, a novel grand canonical Monte Carlo computer code has been developed for the reliable simulation of adsorption processes in porous structures. The source code has been systematically tested and assessed against both literature data and commercial software suites. As a first step the simple geometrical model of slit pores with parallel single graphene walls (atomically described) and varying width was adopted as it is considered the most representative for the majority of carbon porous materials. The study of the H2 interactions with the porous structures was extended for the case of deuterium in order to investigate quantum effects at 77K. The simulation results were combined with gas adsorption measurements (N2, CO2, H2 and D2) on commercial microporous carbon molecular sieves not only for the calculation of micropore size distributions but for evaluating the predictive power of the simulation models as well. Indeed it was proven that the D2 adsorption isotherm at 77K can be predicted quite accurately based on H2 measurements, proving thus the soundness of the models adopted and the approaches followed. Moreover desorption processes were simulated under conditions similar to temperature programmed desorption (TPD) experiments. It was discovered that even for the narrowest pores hydrogen storage is not satisfactory at temperatures higher than 150 K. Hydrogen adsorption was also studied in slit pores, after inducing inhomogeneity elements by decorating the graphene walls with different concentrations of epoxy or hydroxyl functional groups. Depending on the type and concentration of the functional groups the interaction potential landscapes inside the slit pores change significantly, while the calculated adsorption isotherms prove that the density of the adsorbed phase is increasing with increasing oxygen concentration. The oxygen decorated pore models were also used for the study of the effect of functional groups in adsorption based pore size distributions (PSD). Based on simulation results it has been shown that the choice of pore surface model has a profound effect in the GCMC based pore size distribution and thus the chemical properties of the pore wall surface should be taken into account. Beyond the chemically (or energetically) inhomogeneous pore structures hydrogen adsorption was studied for the first time inside novel extremely inhomogeneous geometries, namely carbon cones (CCs). CCs were recently produced and are made of convex graphenes that have 1-5 pentagons (found at the cone tip). CCs are bridging the topology gap between graphene (0 pentagons) and fullerene (6 pentagons for half fullerene), they have distinct differences from conventional carbon structures and have been suggested as possible hydrogen stores. It should be mentioned that this work was part of the European Commission funded (FP6) project HYCONES – “Hydrogen Storage in Carbon Cones”. In this context atomically-described conical pore models with five different tip declinations have been constructed. The structures are made of approximately 2000 carbon atoms and were used to study H2 adsorption inside them by means of GCMC simulations. The simulation results were further compared with carbon nanotube models. It has been shown that the structural inhomogeneity that is directly related with the carbon cone geometry has a significant effect on adsorption density as well as the isosteric heat of adsorption. The results that were briefly described above strongly support that both structural (geometry, size) and energetic (chemistry, size) plays a major role in the microscopic and macroscopic adsorption properties of porous materials. It is thus obvious that the conventional homogeneous pore models that are abundantly used for porous materials characterization (pore size distributions) are oversimplified approaches that cannot describe adequately the porous structure and thus have very limited predictive potential. This work perhaps is a small first step towards the development of a more realistic methodology for the characterization of inhomogeneous porous materials and the prediction of their sorption properties.
Language Greek
Subject Carbon nano-cones
Graphene oxides
Heterogeneity
Ανομοιογένεια
Νανο-κλώνοι άνθρακα
Οξειδωμένα γραφένια
Προσομοίωση
Ρόφηση υδρογόνου
Issue date 2012-03-20
Collection   School/Department--School of Sciences and Engineering--Department of Chemistry--Doctoral theses
  Type of Work--Doctoral theses
Views 417

Digital Documents
No preview available

Download document
View document
Views : 38