Today there are no models to predict the precipitation of salts without experimental data. The fundamental purpose of this project is to develop a predictive means to assess the possibilities for precipitation of salts, without the need for experimental observations.
Many applications with salts in medicine, energy storage, geology and material science, as well as in engineering applications for chemical and biochemical production, involve solid-liquid equilibria.
However, today there are no models to predict the precipitation of these salts without experimental data. Many salts form mixed solids, where neutral molecules (such as water) get embedded in the crystal lattice.
Examples of these include desiccants like CaSO4*2H2O which can e.g. be used as a drying agent for organic synthesis or air moisture. Other examples include MgCl2*6NH3 which can be used to store ammonia in a safe form.
In mixed salt systems, we may also see complex combinations of salts, such as polyhalite (K2SO4 *MgSO4 *2CaSO4 *2H2O). Today, there are no models to predict the precipitation of these salts without experimental data. This severely limits the predictive capabilities of electrolyte models.
The fundamental purpose of this project is to develop a predictive means to assess the possibilities for precipitation of salts, without the need for experimental observations, using the latest advances in computational chemistry and machine learning. The development of a truly predictive thermodynamic model in the form of an electrolyte equation of state for such systems is the ultimate target of the project.
The project is carried out as a part of the ERC Advanced Grant (“New Paradigm in Electrolyte Thermodynamics”) with the goal of gaining a fundamental understanding of electrolyte thermodynamics to enable the development of physically sound and robust models for electrolyte solutions.
The project is carried out under partial supervision from Dr. Bjørn Maribo-Mogensen from Hafnium Labs, a startup specializing in physical property prediction, located in Copenhagen. The software Q-props and insights from Hafnium Labs may be used in the project.
Main supervisor:
Associate Professor Xiaodong Liang
Co- supervisor:
Professor Georgios M. Kontogeorgis
Dr. Bjørn Maribo-Mogensen (external)