Experimental and Thermodynamic Modeling of Hydrate Inhibitors for Sour Gas Systems in South Pars Gas Fields

Summary Dealing with phenomena such as gas hydrate formation, which threatens sustainable production by blocking transmission lines, is a priority during field operations. The presence of laboratory data on gas hydrate thermodynamic equilibria, as well as models that can estimate it with acceptable approximation within the range of the field temperature and pressure profile, seems essential for safe and sustainable production by mitigating the risks associated with hydrate formation. The aim of the present work is to examine the pressure-temperature region of the hydrate zone, at which stable hydrate forms thermodynamically, for the South Pars gas field with natural gas composition including hydrogen sulfide (H2S) in the presence of monoethylene glycol (MEG) and methanol (MeOH), commercially used with the intention of performing thermodynamic inhibition. First, static experimental high-pressure tests were performed at levels of 10 wt%, 20 wt%, and 30 wt%. The presence of two thermodynamic inhibitors shifted the hydrate equilibrium curve (i.e., the border of the hydrate zone) to lower temperatures. The average suppression temperature in the presence of MEG with the concentration of 10 wt%, 20 wt%, and 30 wt% is 1.1°C, 4.9°C, and 9.5°C, respectively. In return for the presence of MeOH, it was observed as 4.0°C, 8.7°C, and 15.2°C, representing a more potent inhibitory effect of MeOH compared with MEG. Then, a thermodynamic model was provided exclusively to anticipate the South Pars gasfield hydrate equilibrium conditions based on the fugacity difference of water in the hydrate and aqueous phases. The model incorporated the van der Waals-Platteeuw (vdW-P) model, Peng-Robinson (PR) equation of state (EOS), and nonrandom two-liquid (NRTL) activity coefficient model related to the hydrate phase, vapor phase, and alcohol/glycol solutions, respectively. The optimized parameters of the model were determined using a genetic algorithm (GA) technique. The prediction of the model showed acceptable accuracy, with average absolute pressure deviations below 7% compared with the experimental results.