The overall goal of the project was to determine if alternative batching methods used in the commercial glass industry would result in improved melting behavior. Improved melting would enable shorter residence times in the glass tank, translating into direct reductions in energy intensity. We proposed that initial reactions between alkali and alkaline earth salts in the batch create a low viscosity melt that segregates from quartz particles, resulting in a mixture that must then be re-homogenized in the tank. The thesis of this project was that segregation can be limited by narrowing the range of melting points of the batch constituents and by increasing the viscosity of the molten phase formed during heating. To control the melting sequence and consequently the viscosity of the molten phase(s), we proposed to selectively batch raw materials into mixtures that form higher viscosity 'endpoints,' instead of simply mixing all of the batch components together prior to charging. Specifically, the idea is to eliminate high-temperature segregation within the batching process, rather than allow the melting process to dictate the composition of the segregated species. The general approach taken to control segregation was to separate the reactions between quartz and soda ash and quartz and limestone/dolomite. This was tested using two methods: (I) selectively batch and pelletize to form small ''reaction volumes'' that react initially to form an intermediate reaction product; or (II) pre-react selective batch components to form an intermediate feedstock. Feedstocks for Approach I were produced by spray drying two binary mixtures of quartz+ soda ash and quartz+ limestone/dolomite. For Approach II, the same mixtures were heated to 1400 C and quenched. Feedstocks for both approaches were then mixed and heated in crucibles until a batch-free melt was produced. The current mixed batch formulation from Guardian Glass was used as a control.
PDF
download
878987797
028-85220240
