Real-time principal component analysis of in-line NIR spectroscopic data as applied to heterogeneous catalysis research Original Research Article

A new methodology for monitoring a catalytic reaction in a continuous bench scale reactor is presented. It uses an in-line near infrared spectrometer to analyse the product. This is followed by principal component analysis which converts a spectrum to a point on a three-dimensional score plot. With a new spectrum being processed every few minutes, a picture of what is happening to the product in real-time evolves. This is a qualitative assessment of the reaction and enables the identification of periods of stability and process disturbances. No time-consuming calibrations or prior knowledge of the ideal near infrared product spectrum is required.The paper discusses this new methodology and illustrates its success with three case studies. It is shown how the method can be used to identify process changes or disturbances quickly, to recognise periods of stability and to detect problems such as rapid catalyst deactivation. Catalyst test runs can now be conducted much faster and more efficiently.