Infinity Supercritical Review: Near-Infrared Spectroscopy: Quantitative Analysis According to ASTM E1655

PDF: 201889-infinity-supercritial-article-review-near-infrared-spectroscopy

Near-Infrared Spectroscopy: Quantitative Analysis According to ASTM E1655

1.  Near-Infrared Spectroscopy (NIRS) is a commonly used analytical tool used for material inspection, because, quote – it enables the possibility to verify rapidly the identity and the quality of … materials – unquote.

2. Its primary use relates to production quality control.  Flexibility of process design allows the analysis to occur at the end of production in a separate lab environment or be integrated into production for in-line quality testing.

3. As with other analytical tools such as, quote – gas chromatography and X-ray fluorescence… NIRS needs to be calibrated using samples with known analyte content – unquote.

4. Calibration allows for the creation of mathematical models which relate chemical properties of the samples to the observed absorbance and spectra of the samples.

5. When unknown samples are analyzed, their chemical properties can be calculated using the mathematical formulas.

6. Unlike other forms of spectroscopy, no single peak in wavelength absorbency is used to define a chemical trait, rather large portion of spectrum are analyzed as a whole. This method effectively deals with spectra with multiple wavelength absorbency peaks, it also necessitates a more concerted calibration effort.

7. ASTM E1655 offers, quote – standard practices for infrared multivariate quantitative analysis – unquote a defines an approach to process design which will guide a user through calibration analysis.

8. More basic analytic processes may be able to standard chemical samples for calibration purposes, but processes with unique materials or greater product variance should use known product samples for calibration.

9. The following influence factors should be incorporated into calibration, quote – geographical origin, manufacturer, particle size, sample temperature, moisture content, sample presentation (quality), and process conditions – unquote.

10. Once calibration has taken place and the analytical process is in use, the equipment performance , quote – should be monitored periodically using a well-defined procedure – unquote. The author suggests ASTM E275 as a possible monitoring procedure.

11. During the process development phase is may be beneficial to determine which spectral ranges are most effected by the changes to the chemical characteristics of interest.  Limiting the spectral range of may improve the specificity of predictive models and reduce the analytical range of product testing.

12. Three primary mathematical model algorithms are common: quote – multilinearregression analysis (MLR), principal component regression (PCR), and partial-least squares regression (PLS) – unqoute. PLS is often the best choice.

13. PLS can be developed to define the spectral differences observed by changing a specific chemical attribute or it can be used to define the ways in which multiple chemical attributes jointly influence the spectral observance.

14. It is very important to correctly identify and remove outliers from the calibration set.  Outliers may be due to incorrect chemical presence or an improperly prepared sample, but their singular impact can have a significant erroneous effect on the overall model.

15. Fortunately, for most users, software is available that will perform PLS and indicate the presence of possible outliers.

16. Once calibration has been completed, a validation process should be undertaken using a selection of known samples from across the possibility of routine analysis.

17. Once validation has been successfully achieved, routine analysis of the target product can be put into action.  It is important to monitor the performance of the equipment and overall analytical process in order to explain any variance in production.  It may also become necessary to expand the calibration set if it becomes apparent that variance in the target product is not what was originally expected.  Successful analysis requires ongoing process refinement.

Source: Kadenkin A. Near-Infrared Spectroscopy: Quantitative analysis according to ASTM E1655. Near-Infrared Spectroscopy: Quantitative analysis according to ASTM E1655.

https://www.metrohm.com/en-us/applications/WP-029EN?fromApplicationFinder=true

More Information: For more information on eco-extraction systems, or the Fast Filter dewaxing machine for removing waxes from winterized solutions for botanical oil, please visit Infinity Supercritical.

 

Leave a Reply