Spectroscopie NIR

In addition, we also offer pre-calibration models. Pre-calibrations are robust, ready-to-use operating procedures for certain applications based on real product spectra. They allow you to immediately start your NIR analysis without any time-consuming sample sourcing, sample preparation, or prediction model development needed. Pre-calibrations are available for quality control of isocyanates, polymers, polyols, fuels, wood pulps, palm oil, food and feed samples, cannabis testing, and more.

Learn more about our pre-calibrations

The advantages of NIR spectroscopy (NIRS) over IR spectroscopy are:

1. The intensity of the bands are lower with NIRS. Thus, there is less detector saturation. This means that for solids you can use pure samples and for liquids the standard method is to use 8 mm vials.
2. The energy of the light is higher with NIRS. This leads to deeper sample penetration. This means you can obtain information about the bulk of the sample and not just the surface.
3. NIRS can be used for quantification and identification, wheras IR is almost exclusively used for identification.
4. NIRS works with fiber optics, which means it can also be used directly in the process.

Learn more about the differences between NIRS and IR in our blog:

Blog: NIR vs IR: What is the difference?

A large selection of NIR pre-calibrations are availabe with the OMNIS NIR Analyzer. These pre-calibrations are based on numerous real product spectra which means you do not need to perform any sample analysis or model creation. You can use the Metrohm pre-calibrations directly and start using your OMNIS NIR Analyzer immediately. Just unpack and you are ready to be successful.

Metrohm NIR pre-calibrations are available for different products and applications in the areas of polymer, chemicals and food & feed. Examples include:

• PE
• PP
• PET
• PA6
• polyols
• isocyanaytes
• cannabis
• edible oil
• palm-oil
• coffee
• flour
• milk powder

Go to pre-calibrations

The OMNIS Software includes algorithms such as PCA, PLS, and SVM, enabling users to develop models with high accuracy and robustness. Visualization of sample properties in score, loading, and influence plots simplifies method development and allows users to extract the maximum value from their data, all while maintaining an intuitive workflow.

Furthermore, the OMNIS Software allows the combination of multiple model types to further increase accuracy and expand application possibilities. For example, identification models can be structured into multiple sub-identification models to distinguish even more similar substances, while sub-quantification models can be built for specific concentration ranges to improve accuracy.

Creation of prediction models can be a challenging task for two reasons.

1. It requires samples with known reference values for the training set, to serve as a basis for the model development. These samples should cover the expected range for the parameter of interest. 
2. In order to create a prediction model from these samples, a technique called chemometrics, or multivariate data analysis, needs to be applied. This usually requires that the user has knowledge of chemometrics.

With the OMNIS NIR Analyzer series, everybody can develop prediction models with the OMD (OMNIS Model Developer) and no knowledge of chemometrics is required. The OMD is started with one click in the software and it creates prediction models automatically and without any further user interaction. These models can be saved and immediately used for routine measurements. When more samples have been measured, these models can also be easily updated at a later stage.