(7a) Prediction of Hydrogen Solubility In Heavy Hydrocarbons Over a Range of Temperatures and Pressures Using Molecular Dynamics Simulations

The solubility of gases in liquids can range over many orders of magnitude. Although fundamental principles of thermodynamics governing solubility are quite well understood, it is still not possible to make accurate predictions of solubility for many industrially important systems using thermodynamics only. Molecular-based studies of gas-fluids system can play an important role in meeting the physical property-related needs of the chemical industry, as a feasible alternative to experimental measurements, especially at high temperatures and high pressures. Possible applications include safe storage of organic liquids, recovery of chemical wastes, and property data estimation for process design. In particular, hydrogen solubility is needed for the modeling of hydrogenation processes, where such data is extremely crucial but traditional equation of state (EOS) methods do not provide accurate enough estimates for such systems.  Using the method of molecular dynamics (MD) simulation, we have estimated the solubility of hydrogen in heavy hydrocarbons for a range of temperatures and pressures. Heavy hydrocarbon systems are known to be challenging not only for experimental measurements but also for reliable estimations using traditional equations of state. The simulation system used was designed with semi-permeable membranes to mimic actual experimental studies of gas solubility. Binary interaction parameters between the solute gas and the solvent (heavy hydrocarbons) components were adjusted when necessary to improve agreement with experimental results and then used in subsequent multi-component studies. Temperature and pressure ranges studied included higher temperatures (up to 685 K) and pressures (up to 11.0 MPa) which are especially difficult to investigate experimentally. Simulation results were finally used to adjust the binary interaction parameters (BIP) in simulation packages (e.g. Aspen) to enable quick and reliable predictions.