Simulated Natural Gas Hydrate Storage: Experimental and Modeling Approach

Abstract

Natural gas hydrate emerged as a hope of an enormous energy source for the future, primarily formed during the encapsulation process of natural gas inside the cages formed by hydrogen bonding of the water molecules. The research has been going on to develop gas hydrates as an energy storage and transportation source for the existing natural gas value chain, being a relatively safe and sustainable medium. However, the significant limitation of natural gas hydrate is the mass transfer limitation due to interfacial resistance during hydrate growth. Therefore to resolve such issues, we have studied low molecular weight hydrocarbons having carboxylic acid (-COOH) groups by varying their concentrations. The acid molecules can be easily dissolved in the water due to their polar nature, thus making them more suitable as gas hydrate additives. Further, this study helps to determine the effect of acid additives in the formation kinetics of the gas hydrates, i.e., enhancement of natural gas storage in the form of a gas hydrate or inhibition of natural gas hydrate formation. An enhancement in the formation of gas hydrates was observed when a relatively higher concentration of succinic acid was applied. On the other hand, a lower concentration of oxalic acid resulted in the comparatively higher inhibitory property of the gas hydrate formation. In short, the results achieved from this study may be helpful in scaling up the process of natural gas storage in the form of gas hydrates with the help of chemical additives (usually non-foaming), eliminating the formation of the excessive amount of foam usually observed with conventional kinetic hydrate promotors, e.g., sodium dodecyl sulfate (SDS). The Artificial Neural Network (ANN) modeling has been used to validate the experimental results of the mole consumption of natural gas in the gas hydrate. The predicted model from ANN was in good agreement with the experimental results. The Adaptive Neuro-Fuzzy Inference System (ANFIS) was further applied to predict the mole consumption data, having the synergistic effect of ANN with fuzzy logic to provide more extensive outcomes, including all the variability lying within the system. ANFIS predicted data also followed a similar trend to the ANN outcomes. The modeling data provided more insights into the gas hydrate formation kinetics, which can be further applied to screen the additives in the future for screening purposes.