Raman versus NIR Spectroscopy – Practical considerations for PAT analyzer selection

When it comes to the collection of analytical techniques, there are a large number which exploit
spectroscopy in one form or another and provide a range of tools for both the identification and
measurement of atoms, ions, and molecules. Vibrational spectroscopy employs a wavelength range that
interrogates the vibrational energies of molecular bonds. Two of the main tools within the realm of
vibrational spectroscopy are Near Infrared (NIR) spectroscopy and Raman spectroscopy. Both are used
in industry because of their non-destructive nature, fast acquisition times and accurate/precise
informational content. However, NIR’s poorer selectivity and the impact of physical attributes such as
particle size, density, temperature, and moisture on results pose a significant practical challenge for
cost-effective implementation of the technique. It particularly imposes heavy requirements on
quantitative model maintenance since changes over time in the factors cited above require model

Raman spectroscopy, which provides users with more specific in situ measurements of complex
chemical matrices compared to other real-time measurement modalities, features many of the same
favorable attributes found with NIR with the additional critical advantage of chemical specificity, which
facilitates process understanding. Because of the excellent selectivity, reliability, robustness, and
flexibility, Tornado’s HyperFluxTM PRO Plus Raman analyzer is often a better choice when deciding which
technique to implement based on the need to monitor the critical quality attributes and properties of
users manufacturing process.

These vibrational spectroscopic technologies each possess unique strengths and limitations, and each
have their place. This blog briefly compares the two techniques, discusses their similarities and
differences, and perhaps more importantly, is intended to be used as a practical guide to allow PAT
scientists, process development chemists, or process control engineers to understand the main
differences and capabilities of the two techniques.

Near Infrared (NIR) Spectroscopy
NIR is a vibrational spectroscopic method that uses the near-infrared region of the electromagnetic
spectrum (from 780 nm to 2500 nm). Since bands in this region are weak, no dilution of materials in
matrices is needed to obtain the spectral information. This presents a great advantage for the analyst
since no sample preparation is required eliminating sample preparation errors, analyst labor and sample
destruction. Typical applications include medical and physiological diagnostics and research including
blood sugar, urology (bladder contraction), and neurology (neurovascular coupling). There are also
applications in other areas as well such as pharmaceutical, food and agrochemical quality control,
atmospheric chemistry, and astronomy.

NIR systems can be principally used to measure NIR radiation that has been collected in two modes,
either via diffuse reflectance or transmission. The instrument’s configuration such as detector, detector
position and power will change depending on the desired mode.

A large selection of benchtop and process NIRS instruments can be found in the market as these systems
are common analytical tools in quality and process development laboratories. Process NIRS may employ
invasive or non-invasive probes. Probe selection is application dependent and will depend on the
sampling interface. Furthermore, process NIRS can also incorporate wireless communication
technologies allowing easier process monitoring set-ups.

Raman Spectroscopy
Whereas NIR is accomplished via an absorption measurement, Raman is a scattering technique that is
low energy in nature. It requires superior photon management to derive maximum benefit. Irradiation
of a sample with an intense single wavelength light source results in light scatter, most of which is at the
same wavelength as the source and is called Rayleigh scatter. However, a small amount will scatter at a
different frequency. This is known as the Raman Effect. This process is not the same as the NIR
absorption process,. Many absorptions which are weak in the mid-IR are strong in Raman, and in
molecules with a centre of symmetry the techniques are mutually exclusive. FTIR and Raman are
therefore said to be complementary. Symmetric vibrations give rise to intense Raman lines;
nonsymmetric ones are usually weak and sometimes unobservable.

Raman spectrometers are not as simple as NIR devices. Given the low number of photons produced,
lasers are used to generate the monochromatic radiation needed to produce enough inelastically
scattered photons. Raman systems also require a series of notch and/or edge filters to block or
attenuate the Rayleigh contribution from the sample, which could otherwise be overwhelming if
allowed to reach the detector.

Raman spectrometers (like NIR systems) can also be found in dispersive or Fourier Transform (FT)
configurations. The former design is far more common as the advantages of the FT design are not as
pronounced in Raman instrumentation. Dispersive instruments usually utilize diode lasers in the visible
and near infrared region (532-980 nm) with multidimensional and/or one dimensional charged coupled
device (CCD) cameras as detectors. Systems using excitation sources in the visible region are more
efficient, since the intensity of the Raman signal is proportional to the bi-quadratic of the laser

One of the major criteria in selecting a Raman analyzer is speed of measurement. A powerful laser (≈500
mW) combined with a high-throughput spectrometer delivers the fastest spectral acquisition time by
provided exposures in milliseconds. This leads to lower limits of detection and provides actionable
information in real time about the process. In the specialty chemical and pharmaceutical space, finer
quantitative resolution (accuracy) does make a significant difference where measurements are made in
parts per million. For example, one of our customers needed to know that their reactant was specifically
at 400 parts per million. We were able to make such a measurement, while the next best Raman
instrument on the market could not measure to that resolution fast enough. The difference for the
customer in being able to measure more precisely is vast. In the biopharmaceutical industry, fast,
precise measurements in downstream bioprocessing, for example, can assist companies during the
purification process and enable them to achieve reliable real-time results.

Practical similarities between and NIR and Raman
Both NIR and Raman are used in chemistry to provide a structural fingerprint by which molecules can be
identified. Multivariate spectroscopic data analysis of Raman or NIR signals can also be used to achieve
quantitative analysis of organic samples and some inorganic materials. In addition, both methods are
non-destructive which enable direct analysis of intact samples (i.e. no sample preparation). Both
spectrometer types may be/can be interfaced with probes equipped with long fibers to improve sample
flexibility, which is useful for process analytical technology applications.

As for which technique is optimal for the analysis, it will depend on the application and the logistics of
the measurement. By better understanding all of the key points of your process, your team will be able
to make a better decision of which sensor should be employed in the desired environment.

Although NIR spectroscopy offers some advantages, such as determining differences in particle size,
Raman spectroscopy often makes it easier to evaluate spectra. Fluorescence is one of the major issues
that we encounter and NIR is certainly a good alternative where this is a debilitating problem. Even with
fluorescence, however, if the Raman signal can be seen over the fluorescence signal, a successful
measurement can often be made (with Tornado’s PRO Plus Raman analyzer). Raman cannot effectively
measure water content, whereas NIR can. NIR can also be employed for wireless measurements on a

moving vessel, which is a difficult situation for Raman. For practically any other circumstance, however,
Raman is often a strongly favorable option. This is because of the advantages cited throughout this blog.
Raman can be used more effectively for measurements in an aqueous matrix, such as in bioprocessing.
Raman also provides specific molecular information. Each peak can be definitively related to a functional
group in a particular environment. This leads to better quantitative and qualitative specificity. While NIR
practitioners rely heavily on chemometric interpretations of their data, those employing Raman are
allowed an additional dimension of molecular interpretability. Those using Raman are not only afforded
empirical quantitative insights, but also have the added benefit of discerning molecular changes (or
molecular stasis, as the case may be). This leads to extraordinary process understanding, the value of
which cannot be overestimated. This is one of the truly game-changing benefits of using Raman for
process development and monitoring.

If you would like to discuss your application and are considering adopting Raman spectroscopy as a
process analytic technology, please contact us at info@tornado-spectral.com