Experimental determinations of new phase diagrams of molten salts are difficult, costly, and time consuming. To go around this problem, it is possible and necessary to use thermodynamics to predict the phase and temperature relations in almost any system under consideration, provided there is enough data available. After finding the areas of interest in each system in which trade-offs of physical and chemical properties can be made, one can again return to the experiment to verify the theory in a very short time. After checking the validity of our calculation method on two systems known from the literature, several new ternary and quaternary reciprocal systems were calculated. A continuous literature search for the basic thermodynamic data was the key to generate those systems. Several equations that help to estimate the missing data are given, and the importance of some particular data is emphasized. Contaminants, mainly water vapor (but also metal oxides, carbon dioxide and others), can have a profound effect on the properties of the salts. Some contaminants can be treated as reactive gases which interact with the molten salt solution. The theory, the references to experimental methods necessary, and the interpretation of interactions are the topics of the final part of the paper, and might constitute a contribution towards avoiding or reversing the contamination.