Dietrich Wienke

On the use of recent developments in vibrational spectroscopic instrumentation in an industrial environment: quicker, smaller and more robust

Abstract In recent years, considerable developments have taken place in Raman, mid-infrared (mid-IR) and near-infrared (NIR) instrumentation. This has had a major influence on the number and type of applications these techniques have in industry. In the current paper, we illustrate this by three applications from our laboratory. The first example relates to the UV curing of coating resins. By using rapid-scanning mid-IR in combination with a reflection–absorption cell, it became possible to follow the kinetics of the curing reaction in thin resin layers on the 50–1000-ms scale. Several examples are presented. In addition, it was shown that multichannel near-infrared spectroscopy (NIRS) is a powerful tool to study curing reactions in thick layers (100–500 μ) and/or on a time scale under 50 ms. For in situ reaction monitoring, NIRS, mid-IR–attenuated total reflection (ATR) spectroscopy or laser Raman spectroscopy are currently the favoured choices of vibrational techniques. An example is given describing the formation of melamine formaldehyde (MF) resins followed by in-line Raman spectroscopy. It was shown that modern high-throughput fibre optic Raman spectroscopy readily provides real-time information on solvation behaviour, the conversion of the reaction and the end point. The availability of high-resolution (10 nm) band-pass filters and high-resolution (10 nm) acousto–optical tunable filters (AOTF) enabled us to develop both a portable and a rapid carpet identifier. In order to recycle valuable face fibres from carpet waste material, one has to sort the carpets according to the face fibre type. For this purpose, a cheap portable identifier (identification in 2 s, battery-powered, 3 kg) was developed based on NIR filter technology. In addition, a rapid carpet identifier was developed based on the AOTF, also working in the NIR. With the latter instrument, one can identify the face fibre of the carpets in ca. 25 ms.