VITA-SCALE: Vitrification AI for Scale-Bridging Prediction and Process Translation
Vitrification is a key process for safely immobilizing certain waste materials in glass, but results from laboratory and pilot-scale tests do not always predict how a full-scale melter will perform. Even when the feed chemistry and glass properties are similar, operational outcomes can change with scale because equipment geometry, heat transfer, melt circulation, and agitation interact in complex ways. This gap makes process planning more difficult, can require additional testing and conservative operating margins, and leaves uncertainty when translating promising small-scale results to full-scale operations.
VITA-SCALE—Vitrification AI for Scale-Bridging Prediction and Process Translation—will develop a Super Intelligence (SI)-enabled workflow to help close this gap. The project aims to predict how important vitrification outcomes vary with feed chemistry, operating conditions, and equipment scale. These outcomes include melt rate, the time needed for the glass to become homogeneous, cold-cap behavior, redox evolution, off-gas response, and overall throughput.
The workflow will bring together multiple kinds of vitrification data and connect them with physical descriptions of chemistry, heat transfer, and material transport. Its agentic SI system is designed to organize and analyze these inputs, create chemistry representations grounded in established scientific principles, link them to thermal and transport models, and generate predictions across scales. By automating this process, VITA-SCALE aims to make scale translation more consistent and less dependent on manual analysis at each step.
Led by Savannah River National Laboratory, the project builds on work at SRNL that uses SI and digital-twin methods to understand and address operational challenges in vitrification systems. Previous efforts have applied tools such as digital twins, Random Forest analysis, and convolutional neural networks to investigate equipment and process issues, including condenser overheating, temperature behavior in an off-gas condensate tank, and solids plugging in a melter quencher. VITA-SCALE extends this foundation from diagnosing specific operational problems toward predicting how vitrification processes may behave at different scales.
The long-term goal is to support more confident decisions about testing, process design, and full-scale deployment. By combining operational evidence with physics-guided SI, the project seeks to reduce uncertainty in vitrification planning and improve the translation of laboratory and pilot-scale knowledge into reliable full-scale performance predictions.
Partners
- Savannah River National Laboratory
