Glossary · Automation fundamentals, platforms and components
Spectroscopy
German: Spektroskopie
In analytics and process measurement, spectroscopy is the family of methods that analyze how matter absorbs, emits or scatters electromagnetic radiation as a function of wavelength, in order to identify substances or determine their concentration. Common process methods include UV-Vis, near-infrared (NIR), mid-infrared and Raman spectroscopy.
- Automation components
- Process industry
In one sentence
Spectroscopy analyzes how matter absorbs, emits or scatters light by wavelength to identify substances and measure concentrations.
Example
An NIR spectrometer on the conveyor measures the moisture content of grain continuously and sends the value to the dryer control.
How it applies
- Engineering: Process spectrometers are installed in-line (probe in the process), on-line (bypass) or at-line (near the line). Quantitative results rely on calibration models, often built with chemometrics or Machine learning (ML) methods.
- Validation: The model is as important as the instrument. Model changes, reference methods and performance checks must be managed, especially in Process analytical technology (PAT) under GMP.
- Documentation: Manuals should cover sample interface, cleaning, reference measurements and model versions. Where lasers or UV sources are used, include the corresponding safety information.
Spectroscopy vs. single-parameter sensors
A pH or conductivity sensor measures one property directly. A spectrometer measures a whole spectrum and derives one or more values through a model, which adds flexibility but also model maintenance.