You have been redirected to your local version of the requested page

Five myths about online dispersive NIR spectroscopy, FT-NIR, and FT-IR – Part 2

Apr 4, 2022

Article

This article is Part 2 of a series.

Go to

Part 1

In the first part of this series, a brief historical overview was given for both infrared (IR) and near-infrared (NIR) spectroscopy, as well as for Fourier transformation (FT) and dispersive spectroscopy. A few myths were discussed and put to rest, and we showed that Fourier transformation spectroscopy (FT-NIR) is not necessarily the only nor the best way to integrate reproducible spectroscopic measurements into industrial processes. On the contrary—dispersive instruments are a robust possibility with ideal opportunities for model transfer, high resolution, and high light throughput even for sensitive applications. Dispersive NIR is at least as good as FT-NIR.

Now, two more misconceptions will be cleared up. Here we’ll go more into detail comparing the IR and NIR wavelength ranges. Furthermore, we will show that most IR applications can also be realized with NIR spectroscopy, and that this results in many economical benefits for plant operators. In the rest of this article, we will compare NIR and IR spectroscopy directly from a process integration point of view and show a real case study of application development with an IR replacement strategy. With this, we conclude that dispersive NIR is better for process integration than FT-IR.
 

Click below to jump directly to a section:

Myth 4: Many IR applications cannot be implemented with NIRS due to its lower specificity and higher detection limits

From theory, it is known that just as for UV-VIS spectroscopy, NIR and IR spectroscopy also follow Lambert–Beer's law. Here the measured extinction depends on the optical pathlength, the substance-specific extinction coefficients, and the concentration of the analyte. If you’re interested in seeing Lambert’s original publication, you can find it below.
 

Photometria sive de mensura et gradibus luminis, colorum et umbrae


Due to the high extinction coefficients of organic components in the IR range, even low concentrations can be reliably determined. However, either a strong dilution of the sample is necessary (which is hardly possible in a production process), or the optical pathlength is drastically reduced. Usually 50–200 µm cuvettes are used for the IR wavelength range.

However, this has significant drawbacks within the process: the sample streams might be dirty or form deposits on the optics from time to time, which means cleaning is very difficult and can lead to accidental misalignment. If the optics must be disassembled, a reproducible measurement is hardly possible afterwards since the application has been created for a highly accurate fixed pathlength. This necessitates costly and time-consuming recalibration procedures to readjust the calibration models, with the associated downtime for the instrument(s). Operational reliability is jeopardized because measurements cannot be performed during this time. In this case, the application should be transferred to a sampling solution with higher pathlength, such as to the NIR wavelength range.


Method development: what is it all about? Have a look at our related blog articles on this topic below.

How to implement NIR spectroscopy in your laboratory workflow

NIR spectroscopy pre-calibrations: Immediate results


In the NIR wavelength range, immersion probes and flow cells with significantly longer pathlengths (0.5–20 mm) are used. These are either adjusted by spacers or by threaded screws so that an extremely reproducible adjustment can be made. If contamination occurs, cleaning is also much easier.

Industrial flow cells used for fast loop, bypass pipelines, and harsh industrial environments.