Application Potential

Previously it has not been possible to monitor the course of a high-pressure extraction directly during the extraction of products and during workpiece cleaning in liquid or supercritical CO2. However, this would be advantageous, in order to know the current status of the process and to be able to terminate it efficiently. As part of a technology transfer project, the Research Centre in Karlsruhe in cooperation with Fa. Sitec Sieber Engineering AG has developed a mobile module ready for marketing, which allows NIR in-line monitoring of both extractions and reactions in CO2 up to 180°C and 1000 bar. The application potential for the new NIR in-line monitoring module was seen in particular in the industrial extraction with CO2 (natural products, parts cleaning) since in this area the process can be controlled and thus the selectivity increased, efficient termination criteria established for the extraction, and the production costs lowered.

The reaction sequence of fast reactions in CO2 is also monitored by NIR in-line monitoring. These data will be used for the elucidation of reaction mechanisms, for kinetic modelling and for process optimization.


Application potential of SINASCO for industrial CO2 extraction

  • Coffee and tea decaffeination: Extraction control, termination
  • Hops: Extraction control, termination
  • Cacao: Extraction control, termination
  • Nutmeg: Selective extraction for detecting the transition from oil to butter
  • Spices, rapeseed oil: Selective extraction, extraction control and termination
  • Paprika: Selective extraction, extraction control and termination
  • Essential oils: Selective extraction, extraction control and termination
  • Cork: Extraction of trichloroanisole (TCA) on a ppm scale
  • Rice: Removal of pesticides, extraction of starch (huge potential in Asia)
  • Parts cleaning and abrasive slurry cleaning


General application

  • Determination of contamination curves of unknown systems: Previously only possible off-line through cumbersome manual testing, but now automatable in-line