We present a new molecular simulation technique for determining partial molar enthalpies in mixtures of gases and liquids from single simulations, without relying on particle insertions, deletions, or identity changes. The method can also be applied to systems with chemical reactions. We demonstrate our method for binary mixtures of Weeks-Chandler-Anderson particles by comparing with conventional simulation techniques, as well as for a simple model that mimics a chemical reaction. The method considers small subsystems inside a large reservoir (i.e., the simulation box), and uses the construction of Hill to compute properties in the thermodynamic limit from small-scale fluctuations. Results obtained with the new method are in excellent agreement with those from previous methods. Especially for modeling chemical reactions, our method can be a valuable tool for determining reaction enthalpies directly from a single MD simulation. © 2014 AIP Publishing LLC.

We show how we can find the enthalpy of a chemical reaction under non-ideal conditions using the Small System Method to sample molecular dynamics simulation data for fluctuating variables. This method, created with Hill's thermodynamic analysis, is used to find properties in the thermodynamic limit, such as thermodynamic correction factors, partial enthalpies, volumes, heat capacities and compressibility. The values in the thermodynamic limit at (T,V, μj) are then easily transformed into other ensembles, (T,V,Nj) and (T,P,Nj), where the last ensemble gives the partial molar properties which are of interest to chemists. The dissociation of hydrogen from molecules to atoms was used as a convenient model system. Molecular dynamics simulations were performed with three densities; ρ = 0.0052 g cm(-3) (gas), ρ = 0.0191 g cm(-3) (compressed gas) and ρ = 0.0695 g cm(-3) (liquid), and temperatures in the range; T = 3640-20,800 K. The enthalpy of reaction was observed to follow a quadratic trend as a function of temperature for all densities. The enthalpy of reaction was observed to only have a small pressure dependence. With a reference point close to an ideal state (T = 3640 K and ρ = 0.0052 g cm(-3)), we were able to calculate the thermodynamic equilibrium constant, and thus the deviation from ideal conditions for the lowest density. We found the thermodynamic equilibrium constant to increase with increasing temperature, and to have a negligible pressure dependence. Taking the enthalpy variation into account in the calculation of the thermodynamic equilibrium constant, we found the ratio of activity coefficients to be in the order of 0.7-1.0 for the lowest density, indicating repulsive forces between H and H2. This study shows that the compressed gas- and liquid density values at higher temperatures are far from those calculated under ideal conditions. It is important to have a method that can give access to partial molar properties, independent of the ideality of the reacting mixture. Our results show how this can be achieved with the use of the Small System Method.

Summary Cement-sheath integrity is important for maintaining zonal isolation in the well. The annular-cement sheath is considered to be one of the most-important well-barrier elements, both during production and after well abandonment. It is well-known, however, that cement sheaths degrade over time (e.g., from repeated temperature and pressure variations during production), but the link between leak rate and the cause of cement-sheath degradation has not yet been established. In this paper, we have studied fluid flow through degraded cement sheaths. The degree of degradation of the cement sheaths varied from systematically connected cracks to real microannuli. The leak paths, created by thermal-cycling experiments, were imported into a computational-fluid-dynamics (CFD) simulation software. The pressure drop over the cement sheath was used as a boundary condition, and the resulting pressure-driven flow was studied using methane gas as the model fluid. The Forchheimer equation was used to estimate the effective permeability of the cement sheaths with defects. Our results show that the pressure-driven flow is complex and greatly affected by the geometry of the flow paths. A nonlinear pressure-buildup curve was observed for all experimental cases, indicating that Darcy's law was not validated. For homogeneous microannuli, the pressure-buildup curve was linear. The estimated effective permeability for all cases was observed to be orders of magnitude larger than that of a good cement sheath.

We have developed a classical molecular dynamics model for the hydrogen dissociation reaction, containing two- and three-particle potentials derived by Kohen, Tully and Stillinger. Two fluid densities were investigated for a wide range of temperatures, and 11 fluid densities were considered for one temperature. We report the temperature range where the degree of reaction is significant, and also where a stable molecule dominates the population in the energy landscape. The three-particle potential, which is essential for the reaction model and seldom studied, together with the two-particle interaction lead to a large effective excluded volume diameter of the molecules in the molecular fluid. The three-particle interaction was also found to give a large positive contribution to the pressure of the reacting mixture at high density and/or low temperatures. From knowledge of the dissociation constant of the reaction and the fluid pressure, we estimated the standard enthalpy of the dissociation reaction to be 430 kJ mol(-1) (ρ = 0.0695 g cm(-3)) and 380 kJ mol(-1) (ρ = 0.0191 g cm(-3)). These values are in good agreement with the experimental vaule of 436 kJ mol(-1) under ambient pressure. The model is consistent with a Lennard-Jones model of the molecular fluid, and may facilitate studies of the impact of chemical reactions on transport systems.

The cement sheath is one of the most important well barrier elements in the well, both during production and after abandonment. However, normal production operations which involve temperature variations in the well, such as steam injection, stimulations and shut-down periods, may damage the integrity of the cement sheath. Temperature increase and decrease, i.e. thermal cycling, cause the casing to expand and contract, which creates debonding and cracking of the cement sheath and thereby loss of zonal isolation. This paper presents novel results from an experimental study of cement sheath integrity during thermal cycling. The temperature was cycled repeatedly from 5°C to 125°C in a controlled manner from inside the casing, and Portland cement with silica additive was tested with both sandstone and shale as surrounding rock. Debonding and cracking of cement were quantified and visualized by X-ray computed tomography (CT), and it was found that cracking and debonding occurred for the sandstone sample, whereas the shale sample remained almost unaffected. There were some initial defects in the cement sheath in the sandstone sample, and these small and scattered defects grew together during thermal cycling into a continuous leak path; i.e. resulting in a loss of zonal isolation.The digitalized 3D geometry of this leak path was imported into Computational Fluid Dynamics (CFD) software, thereby enabling a unique visualization of fluid flow through an actual leak path in degraded cement and an estimation of leak rates for different pressure differences. It is seen that microannuli are not homogeneous or uniform, and that fluid flow through microannuli and cracks is complex and not easily predictable.