Gas hydrates are molecular host-guest mixtures where guest gas species are encapsulated in host water networks. They play an important role in gas storage in aqueous environments at relatively low pressures, and their stabilities are determined by weak interactions of the guest species with their respective host water frameworks. Thus, the size and the amount of the guest species vary, depending on the size of the empty space provided by the host water structures. The systems studied here are noble gas (He, Ne, Ar) and diatomic (H2) hydrates. Because of the similarity of the guests' sizes between the noble gases and the di-atomic gases, the noble gas hydrates act as simple models for the di-atomic gas hydrates. For example, He, Ne and H2 have approximately the same size. Density functional theory calculations are used to obtain the ground state formation enthalpies of each gas hydrate, as a function of host network, guest stoichiometry, and pressure. Dispersion effects are investigated by comparing various dispersion corrections in the exchange-correlation functionals (semi-local PBE, semi-empirical D2 pair correction, and non-local density functionals i.e. vdW-DF family). Results show that the predicted stability ranges of various phases agree qualitatively, although having quantitative difference, irrespective of the methods of the dispersion corrections in the exchange-correlation functionals. Additionally, it is shown in gas-water dimer interaction calculations that all DFT dispersion-corrected functionals overbind significantly than the interaction acquired by the coupled-cluster calculations, at the CCSD(T) level, which is commonly accepted to provide the most accurate estimation of the actual interaction energy. This could lead to an overestimation of the stability of the hydrate mixtures. Further study in the gas-water cluster indicates that less overbinding effect is found in the cluster than in the dimer. This implies that the overbinding energy caused by DFT might become less pronounce in the solid phase. Graph invariant topology and a program based on a graph theory are used to assign protons based on the 'ice rule' to fulfill the incomplete experimental structural data such as unknown/unclear positions of protons in the host water lattices. These methods help constructing host water networks for computational calculations. Several configurations of the host water structures are tested. Those configurations having lowest enthalpies are used as the host water networks in this research. Furthermore, the enthalpic spread between the configurations having the highest and the lowest enthalpy in the pure water ice network is very small (about 10 meV per water molecule). Nevertheless, it is still unclear to conclude that this protonic effect is also trivial in the gas-water compound. Therefore, this study also calculates the enthalpies of the gas-water mixtures having various proton configurations in the host water networks. Results indicate that very small enthalpic distributions among the proton configurations are found in the compounds as well. Furthermore, the enthalpic spread is almost constant as pressure increases. This suggests there is no pressure effect in the enthalpy gap amoung the proton distributions in both pure water ice and the gas-water compounds. Predicted stable phases for the noble gas compound systems are based on four host water networks, namely, ice Ih, II and Ic, and the novel host water network S!. The He-water system adopts ice Ih, II and Ic network upon increasing pressure. In the Ne-water system, a phase sequence of Sx/ice-Ih, II and Ic with a competitive hydrate phase in the S! host network at very low pressure is found. This is similar to the phase evolution of the H2-water system. For the Ar-water mixture, only a partially occupied hydrate in the Sx host network is found stable. This Sx phase becomes metastable if taking the traditional clathrates (sI and sII) into account. This result agrees very well with the experiment suggesting only two-third filling is found the large guest gases i.e. CO2. For the diatomic guest gas compound systems, the traditional clathrate structure (sII) that found to be existed experimentally in the H2-H2O system is also included in this study together with those four host water networks. Predicted phase stability sequence as elevated pressure is as follows: Sx, ice-Ih, II and Ic. This computationally prediction agrees very well with experiment. Results in this work suggest that the compound based on the traditional clathrate structure II (sII) host water framework is found to be metastable with respect to the decomposition constituents - in this case, they are pure water ice and the S!. The metastability of the hydrogen hydrates based on the sII structure might due to zero-point motions or other dynamic/entropic mechanisms uncovered in this research. Dynamic studies concerning the transition states of the hydrogen guest molecules in three competitive phases at very low pressure (less than 10 kbar), based on Sx, ice-Ih, and ice-II host water network, are considered. The energy barriers required by the hydrogen guest molecules in those three host frameworks are calculated by using Nudged Elastic Band (NEB) method. Results suggest that the hydrogen molecules are more mobile in the Sx than the other two host structures significantly. In the S! host water network, the energy barrier is about 25 meV/hydrogen molecule. This energy is about the room temperature suggesting that the hydrogen guest molecules are easily mobile in the Sx host water network if there is an empty site adjacent to them.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:756639 |
| Date | January 2018 |
| Creators | Teeratchanan, Pattanasak |
| Contributors | Hermann, Andreas ; Ackland, Graeme |
| Publisher | University of Edinburgh |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://hdl.handle.net/1842/31359 |
Page generated in 0.0026 seconds