What is NIR spectroscopy?
At Vibralytics, we use Near-Infrared (NIR) spectroscopy and chemometrics to obtain quantitative and qualitative information about the composition and properties of materials. Not only food products. We work with:
NIR stands for Near-Infrared Spectroscopy. It is an analytical technique based on the interaction between near-infrared radiation and matter. The NIR region extends approximately from 780 to 2500 nm, between the visible and mid-infrared regions of the electromagnetic spectrum.
When NIR radiation interacts with a sample, specific wavelengths are absorbed according to the molecular composition of the material. NIR absorption bands originate predominantly from overtones and combination bands of fundamental molecular vibrations, particularly vibrations associated with O–H, C–H, and N–H bonds. These chemical bonds are present in major food constituents such as water, lipids, proteins, and carbohydrates.
Limitations
- Not every molecule absorbs infrared light. Molecules whose vibrations do not change their dipole moment are invisible to NIR.
- No traces. This technology does not detect trace levels: concentrations below the ppm level are out of its reach.
The resulting spectrum contains information related to the chemical composition of the sample. However, NIR spectra are characterised by broad, overlapping absorption bands, meaning that individual constituents generally do not produce isolated, uniquely identifiable peaks. Instead, the analytical information is distributed across multiple wavelengths.
This is why chemometrics is an essential component of quantitative NIR analysis.
To transform the infrared signal into a concrete result, chemometrics or multivariate analysis is used.
What is Chemometrics?
Chemometrics is the science of applying mathematical and statistical methods to chemical data. The term was coined by Svante Wold in 1971, in response to the increasing amount and complexity of data generated by modern analytical instruments.
Chemometrics is used to develop mathematical models that describe relationships within chemical systems and predict properties or behaviours of unknown samples.
In descriptive applications, models are used to explore relationships, patterns, and underlying structure in chemical data. In predictive applications, models are developed to predict properties of new samples.
For example, a chemometric model can relate the NIR spectra of honey samples with known adulterant content to that reference information, allowing the adulterant content of new samples to be predicted.
From spectra to analytical results
A chemometric study involves several stages, including experimental design, data acquisition, preprocessing, model calibration, validation, and evaluation. These stages must follow appropriate modelling practices to obtain reliable and reproducible results.
Once calibrated and validated, a model can process a new NIR spectrum to provide a predicted value or classification. Depending on the sample, reference method, spectral range, experimental design, and calibration model, NIR spectroscopy can be used to predict parameters such as moisture, protein, fat, sugar, and other defined quality attributes.
NIR measurements are rapid and non-destructive, making them suitable for laboratory, at-line, on-line, and in-line applications.
At Vibralytics, we combine NIR spectroscopy, experimental design, and chemometrics to transform spectral measurements into validated analytical models for real-world food applications.
Explore our use cases to see how NIR spectroscopy and chemometrics can be applied to real analytical problems.
Want to learn more about chemometrics?
Request your free manual in Spanish or in English. You can also request 3 hours of specialized online training.