(52e) Isobaric Heat Capacity of Supercritical Fluid Mixtures: Experimental Measurements and Molecular Simulations

Ishmael, M., Cornell University
Stutzman, L., Cornell University
Lukawski, M., Cornell University
Tester, J. W., Cornell University
Isobaric heat capacity of supercritical fluid mixtures: experimental measurements and molecular simulations


Mitchell Ishmael, Lauren Stutzman, Maciej Lukawski, and Jefferson Tester

Cornell Energy Institute and CBE and MSE

Supercritical fluids offer advantageous, tunable properties, making them attractive candidates for a diverse set of applications. Our study was motivated by the potential of using supercritical pure fluids and fluid mixtures as media for thermal energy storage and as working fluids in thermodynamic power cycles for low enthalpy heat sources. We hope to take advantage of the thermophysical property changes that occur near the critical point, in particular the large isobaric heat capacities in the near critical and supercritical region. For a small set of pure fluids and very few fluid mixtures, sufficiently accurate equations of state (EOS) exist for predicting heat capacities in the supercritical region. In order to obtain accurate isobaric heat capacity values for supercritical pure fluids and mixtures, our group has built an accurate (±1%) high temperature (20-150 °C), high pressure (1-300 bar) calorimeter. We have made measurements on a number of binary mixtures (X + carbon dioxide), including methanol, decane, and various refrigerants. As experimental measurements can be costly and time intensive, we also used molecular simulations to estimate heat capacities of supercritical fluid mixtures. A thorough comparison was conducted between predicted and experimental values over a wide range of temperatures, pressures, and compositions. For the majority of these comparisons, quantitative agreement was achieved. As a result, molecular simulation has been demonstrated as a powerful complement to direct experimental measurements and as a distinct alternative to using equations of state for predictions.