Spatially Resolved Reaction Calorimetry and Fouling Monitoring in a Continuous Tubular Emulsion Polymerization Reactor

Abstract

A modular flow reactor setup for the continuous emulsion copolymerization of vinyl acetate and vinyl-neodecanoate has been fitted with Distributed Optical Fiber Sensors (DOFS) based on Rayleigh backscatter and instream thermocouples. After segmentwise calibration of the DOFS for temperature measurement, they were used as sensors for online thermal analysis in the reactor. Calorimetric analysis and thermal conversion determination were conducted online during the reaction using the DOFS as spatially distributed temperature sensors, which span the length of the reactor. Three DOFS in the reactor wall at different diameters enable the quantitative calculation of the thermal energy transferred between the DOFS and therefore determine the heat transport through the reactor wall. In experiments with heated water, the radial heat loss for reference values from 2 to 9 W could be determined with an accuracy of 0.39 W. The sensor-equipped tubular reactor (length 1.95 m) was used for the emulsion copolymerization of vinyl acetate and vinyl-neodecanoate with a redox-initiator reactive at room temperature (RT). The quick and intense fouling process of the reaction could be reflected in the thermal measurements. From the immediate shifting of the reaction hotspot (length of highest monomer conversion) downstream, it could be deduced that the chosen reaction does not have an induction period pertaining to fouling. The broadness of the hotspot and the maximum hotspot temperature T_max could be monitored during the reaction. Due to the fouling constricting available reactor volume at constant feed rate, the hotspot became broader with time and T_max dropped. The drop of T_max could be tracked especially well with the DOFS in the wall, in comparison to the thermocouples instream, and it is deduced to be a useful parameter for fouling monitoring. The heat released from the reaction was calculated using the thermocouples instream for axially transported heat and the DOFS for radially transported heat. It could be seen clearly that during the times T_max was not aligned with one of the six thermocouples, the axial heat was estimated as too low. The radial heat measurement had a high spatial resolution of the DOFS (0.26 cm) and always captured T_max. During good alignment of T_max and one of the six thermocouples, the calculated total reaction heat was 35.3 W or a 79 ±3% monomer conversion. This correlates well with the gravimetrical conversion measurement, averaging 78 ±2% conversion for that time frame in the range of 8–16 dimensionless residence times.