The Use of Mathematical Models and Digital Twins to Improve Scale up. What Is the Role of a Systematic Scale Down for the Experimental Design of Lab Experiments for Smart Scale up?

For direct scale up of a chemical process it is important to master the performance as well of involved chemical reactions as of the used manufacturing equipment.. Comprehensive understanding how a chemical process will behave in production-scale equipment is the key for success to make predictions for process safety, quality and economics in scale-up. Various mathematical models, which describe reaction kinetics, unit operations, mixing properties assist us in process development.

Based on the case study of a complex heterogeneous hydrogenation, it is described, how mathematical models were used in our scale down / scale up approach. Digital twins of manufacturing and lab reactors were established to design laboratory experiments - generating meaningful data for direct scale up from lab to plant. A new concept to ensure comparable heat transfer and geometrical equivalence to lab experiments was introduced. 3D printing of stirrers and baffles is used to vary and study the effect of hydrogen transfer, mixing properties and dispersion of solids. A new concept for heat transfer enables us to employ identical rate of heating surface to reaction volume over various scales. Heat transfer rates as well as surface temperature can be adjusted to mimic the targeted equipment. A special designed colling finger and a new cryostat proptotype hat to be created.

Experimental design, mathematical models for kinetics, mass transfer and heat transfer are combined in a "tool box". A scale down reactor (lab scale twin) was manufactured to validate the model and to test optimized process parameters. Based on the lab experiments the chemical process parameters were optimized. Results and predictions were succesfully compared with data from manufacturing scale and for further refinement of the model.

Publications: https://doi.org/10.2533/chimia.2021.948; OPRD publication planned