Statistical Analysis of Supercooling in Fuel Gas Hydrate Systems

Abstract

The recently developed high pressure automated lag time apparatus (HP-ALTA) was applied to the study of the formation and growth of interfacial gas hydrate films for three gases; methane (C1), 90% methane/10% propane gas mix (C1/C3), and synthetic gas mix (SGM). The effects of gas pressure and cooling rate were studied for each gas. Some degree of supercooling was observed in all cases. The probability distributions of formation temperature (T f) were often found to be bimodal, due to the formation of either gas hydrate or ice in sequential experimental runs. The width of the T f distribution of ice was about 3–4 K. In contrast, the width of the T f distribution of gas hydrates was about 20 K which reflects the importance of mass transfer (gas diffusion) processes in nucleation. Differences in hydrate and ice nucleation probability distributions were observed for different gases, reflecting differences in both thermodynamic equilibrium phase behavior and hydrate formation mechanisms. For all gases studied, T f generally increased with increasing gas pressure. A minimum threshold pressure for hydrate formation was observed, with magnitude decreasing in the order C1 > SGM > C1/C3. The effect of cooling rate on gas hydrate nucleation probability was also studied. The median of the distribution of T f (T f50) was found to decrease with an increased cooling rate, consistent with the increases in effective induction time as samples were cooled more slowly. Our results clearly highlight the value in collecting large data sets which can be used to assemble probability distributions when studying intrinsically stochastic processes such as gas hydrate nucleation and growth.

Probability distributions of gas hydrate formation

“…Since mass transfer of gas into the aqueous phase is required as an additional step in the nucleation and growth processes, this additional kinetic step causes the width of the probability distribution to be much larger in temperature than that in ice. 12 It is, therefore, not clear whether the correlation obtained by Heneghan et al 5 for ice remains valid or applicable at all to gas hydrates because of this intrinsic difference in the mechanisms of nucleation and growth between ice (single-step process) and gas hydrates (two-step process).…”

“…12 They obtained probability distributions of the hydrate formation temperatures (T f ) at different gas pressures and cooling rates, and found that T f generally increased with gas pressure. They also showed that a minimum threshold pressure for hydrate formation exists for each gas studied.…”

Experimental Study on Hydrate Induction Time of Gas-Saturated Water-in-Oil Emulsion using a High-Pressure Flow Loop

“…The authors cautioned that this formula should not be used outside the range of the experimental data correlated by them. Maeda et al (2012) experimentally studied the induction time under different supercooling degrees, and obtained a longer induction time at the slower cooling rate.…”

The TTT Curves of the Heterogeneous and Homogeneous Crystallization of Lithium Disilicate – A Stochastic Approach to Crystal Nucleation

“…Later probability distribution functions in the formation of the first supercritical nucleus have been studied for other liquid metals, such as niobium and zirconium (Morton et al, 1994), aluminum (Uttormark et al, 1997), gold and copper (Wilde et al, 2006(Wilde et al, , 2009, and tin (Yang et al, 2009(Yang et al, , 2011(Yang et al, , 2013 but also in water (pure and seeded) (Barlow and Haymet, 1995;Heneghan et al, 2001Wilson et al, 2005) and in gas hydrates (Maeda et al, 2011(Maeda et al, , 2012. In all these experiments, a single sample was repeatedly undercooled from above the melting temperature, while either the time (at an isothermal hold) or the temperature (at an isochronal cooling) of crystallization was detected.…”