High-level waste (HLW) glass compositions, processing schemes, limits on waste content, and corrosion/dissolution release models are dependent on an accurate knowledge of melting temperatures and thermochemical values. Unfortunately, existing models for predicting these temperatures are empirically-based, depending on extrapolations of experimental information. In addition, present models of leaching behavior of glass waste forms use simplistic assumptions or experimentally measured values obtained under non-realistic conditions. There is thus a critical need for both more accurate and more widely applicable models for HLW glass behavior, which this project addressed. Significant progress was made in this project on modeling HLW glass. Borosilicate glass was accurately represented along with the additional important components that contain iron, lithium, potassium, magnesium, and calcium. The formation of crystalline inclusions in the glass, an issue in Hanford HLW formulations, was modeled and shown to be predictive. Thus the results of this work have already demonstrated practical benefits with the ability to map compositional regions where crystalline material forms, and therefore avoid that detrimental effect. With regard to a fundamental understanding, added insights on the behavior of the components of glass have been obtained, including the potential formation of molecular clusters. The EMSP project had very significant effects beyond the confines of Environmental Management. The models developed for glass have been used to solve a very costly problem in the corrosion of refractories for glass production. The effort resulted in another laboratory, Sandia National Laboratories-Livermore, to become conversant in the techniques and to apply those through a DOE Office of Industrial Technologies project joint with PPG Industries. The glass industry as a whole is now cognizant of these capabilities, and there is a Glass Manufacturer's Research Institute proposal in this area that will likely be supported. In ...