A Multiscale Approach for Gas Hydrates Considering Structure, Agglomeration, and Transportability under Multiphase Flow Conditions: I. Phenomenological Model

Abstract

A new topological model on how gas hydrates form, grow, and agglomerate for oil and water continuous flow, with and without surfactant additives, is presented. A multiscale approach is used to explain how the porous structure of gas hydrates and the affinity between the phases affect the particle morphology and their agglomeration. We propose that gas consumption due to hydrate growth happens mostly in the water trapped inside the capillaries of the hydrate structure near the outer surface of the particles. This approach is herein referred to as the “sponge approach” and is treated as a surface problem, instead of the volume problem often treated in literature (the “shell approach”). Affinity between phases (which in a macro point of view is interpreted as a wetted angle that gives rise to capillarity forces and that can be changed by the use of surfactant additives) describes the preferential entrapment of oil or water inside the hydrate sponge structure. Yet by splitting agglomeration into smaller processes and depending on the morphology of the particles and on the evolution of the porous structure of the hydrates, (i) capillarity bridges may form, causing particles to be sticky, and (ii) water may be available at the outer surface of the particles and may promote consolidation of particle–particle (agglomeration) or particle-wall (deposition). The settling of slurries is treated as a separated solid–liquid flow instability problem once mixture deceleration (due to phase consumption during crystallization) and particle size (due to growth and agglomeration) are known. We also propose a new explanation on how surfactants act as anti-agglomerants in oil continuous flow, differently from the common DLVO theory used in literature, which can only explain anti-agglomeration of particles much smaller than the ones formed over droplets of a very fine dispersion flow.

Advancing Laboratory Characterization and Qualification of Additives for Hydrate Slurry Flow in Multiphase Systems

“…While the apparent volume was not precisely estimated, one can infer from the visual observation that the hydrate bed and deposit absorb not only the water and gas but also a significant amount of oil in its porous structure (visually, less free oil is seen flowing). This observation is consistent with previous results reported in other papers. − It is thus important to realize that a low water conversion does not necessarily result in a low risk for hydrate disruption in terms of aggregation and deposition.…”

Investigation on Hydrate Growth at the Oil–Water Interface: In the Presence of Wax and Kinetic Hydrate Inhibitor

“…Therefore, the initially formed hydrate shell in the presence of wax is not complete with many fine holes where waxes exist distributing on it and consequently has a relatively high porosity. Moreover, during the growth of the hydrate shell, unconverted water inside the water droplet could gradually seep out through the waxexisting holes under the action of the capillary force 39 and the pressure difference between the inside and outside of the droplet, 43 as shown in Figure 7b-3. Subsequently, the water seeping out of the holes would soon spread out on the surface of the hydrate shell around the holes considering the hydrophilic nature of the hydrates and further convert to hydrates, as shown in Figure 7b-4.…”

“…In water-in-oil emulsions, hydrates usually first nucleate and grow at the oil−water interface. 38 Moreover, the properties of the initially formed hydrate shell covering the oil−water interface, such as morphology, roughness, hardness, porosity, and wetness, can greatly affect the flow behaviors 39 (collision, cohesion, adhesion, agglomeration, breakage, plugging, etc.) of the hydrate-coated droplets or the hydrate particles.…”

A Multiscale Approach for Gas Hydrates Considering Structure, Agglomeration, and Transportability under Multiphase Flow Conditions: III. Agglomeration Model

“…where agg K is the kernel of agglomeration (which modeling is the focus of this article), (7) where the average particle size and the average outer surface were written in terms of the moments . Notice that the shape and volume factors are neglected 23 1   , since the agglomerates are not necessarily spherical, as the particles of part II 2 . The geometrical sense is given by the agglomerate shape coefficients 12 , KK when dealing with the particle population.…”