Statistical thermodynamics model and empirical correlations for predicting mixed hydrate phase equilibria

• A computationally-tractable model is presented for mixed hydrate phase equilibria.
• Predicted phase equilibria of mixed hydrates without any adjustable parameters.
• Structural transitions of simple hydrates with temperature and mixed hydrates with relative gas composition are predicted.
• The data is incorporated into reservoir simulators like STOMP and HydrateResSim.

Natural gas hydrate deposits contain CH4 along with other hydrocarbon gases like C2H6, C3H8 and non-hydrocarbon gases like CO2 and H2S. If CH4 stored in natural gas hydrates can be recovered, the hydrates would potentially become a cleaner energy resource for the future producing less CO2 when combusted than does coal. The production of CH4 from natural gas hydrate reservoirs has been predicted by reservoir simulators that implement phase equilibrium data in order to predict various production scenarios. In this paper two methods are discussed for calculating the phase equilibria of mixed hydrates.In the first method, the phase equilibrium is predicted using a ‘cell potential’ code, which is based on van der Waals and Platteeuw statistical mechanics, along with variable reference parameters to account for lattice distortion, and with temperature-dependent Langmuir constants proposed by Bazant and Trout. The method is validated by reproducing the existing phase equilibrium data of simple and mixed hydrates and the structural transitions that are known to occur, without the use of any fitting parameters. A computationally-simple method is to use empirical correlations of gas hydrate dissociation pressure with respect to temperature and gas-phase composition as they are easy to implement into the simulators. The parameters for the empirical expression were determined for the CH4–C2H6 mixed hydrate system by non-linear regression analysis of available experimental data and data obtained from the first method.