Supercritical Water Desalination: Model-Predicted NaCl Concentration Comparison

Freshwater availability is increasingly becoming a concern for various parts of the world. Seawater desalination is becoming more commonplace as a source for freshwater in water-stressed regions such as California and the Middle East. Reverse osmosis is the most commonly employed technology for seawater desalination thanks to its ability to operate at a large scale, producing freshwater from seawater with relative ease. There are other sources of water which need to be desalinated, however, which reverse osmosis systems cannot effectively treat. Unconventional oil and gas extraction, commonly known as fracking, produces vast amounts of wastewater with high TDS levels (approximately 100,000-300,000 mg/L) and dissolved organics, known as produced water, which cannot be effectively treated by conventional desalination technologies. Supercritical water desalination is currently being explored as a solution produced water desalination. Supercritical water desalination takes advantage of waters unique properties beyond its critical pressure and temperature which result in substantially lower solubility for inorganic salts. Designing a supercritical desalination system requires extensive knowledge of fluid properties as well as salt solubilities across a wide temperature and pressure range. Obtaining this information experimentally is expensive and time-consuming. Utilizing a high fidelity model to produce key system properties can improve desalination system design in an efficient manner. This research aims to evaluate various concentration models, and thus their underlying formulation methods, and determine which model yields the most accurate concentration results for a produced water sample across the temperature range 25-450 °C at 240 bar. It is hypothesized that an empirically-derived model will outperform a conventional thermodynamic-based model for concentration determination at these elevated conditions. This research was accomplished by comparing the predicted concentrations of a NaCl-H2O solution produced by the concentration models: HSC, PHREEQC, AspenPlus, and SoWat to experimental data across the aforementioned process conditions. The predicted NaCl concentration produced by each model was evaluated to determine its ability to accurately predict concentration at elevated conditions. The empirically-derived SoWat model predicted NaCl concentration curve outperformed the concentration curves produced HSC, PHREEQC, and AspenPlus when comparing with experimental data. This model can be confidently utilized to develop a supercritical water desalination system as its predicted results are accurate. The employment of a high fidelity model such as SoWat will drastically reduce the cost and time required to develop an effective supercritical water desalination system