The development of thermochemical hydrogen cycles for the centralized production of hydrogen has been identified as an area of primary interest among various international energy research groups. A leading candidate for the production of hydrogen via water splitting is the sulfur-iodine (S-I) cycle, where water is converted into hydrogen and oxygen via a three-step process: (1) the Bunsen reaction to form hydriodic acid and sulfuric acid, (2) decomposition of sulfuric acid, and (3) decomposition of hydriodic acid. However, this cycle involves complex, highly nonideal phase behavior that is poorly understood, so performance projections and efficiency calculations are currently being made with little confidence. Only with quality experimental data, along with carefully constructed property and process models that correctly describe the process, can the true potential of the S-I cycle be assessed. This, in essence, is the goal of our DOE-funded project team.

In our talk, we present phase-behavior measurements relevant to the HI decomposition section (i.e., the reactive distillation column) of the S-I cycle, specifically for mixtures of hydriodic acid (HI), molten iodine (I₂), and water at elevated temperatures and pressures. Initial process simulations of the S-I cycle by our project team have identified liquid-liquid equilibrium (LLE) measurements for the iodine-water binary as a top priority, as both the phase compositions and the UCST for this system are seen to have a significant impact on the predicted ternary HI–I₂–water phase behavior for the column. Only one study (F. C. Kracek, J. Phys. Chem., 1931, 417-422) has ever been reported for this system, and the results extend only up to ~200 °C, well below the expected operating temperatures of the S-I process.

A continuous-flow apparatus with a windowed equilibrium cell has been designed and constructed to measure equilibrium phase compositions for HI-I₂-H₂O mixtures at temperatures to 350 °C and pressures to 200 bar. To our knowledge, the apparatus is the first of its kind capable of making measurements for such highly corrosive systems at elevated temperatures and pressures, as all wetted parts are made of tantalum. In addition to presenting results for the iodine–water binary, we also expect to be able to discuss our initial measurements for the HI-I₂-H₂O ternary.