Spectroscopic techniques and advanced analysis methods for characterization of bio-based fertilizers
Spectroscopy in combination with chemometrics is considered a precise analytical method for finding constituents of materials with unknown compositions. The spectroscopic analysis mainly involves the investigation of the light interaction with matter. This interaction of light with matter is unique for each material, and useful information about the composition of the material can be derived from such interaction. Within the FertiCycle project, we aim to establish a fast and rapid bio-based fertilizers characterization framework based on non-destructive spectroscopic sensing technologies. The developed framework will help to quantify the nutrient content in bio-based fertilizers and their plant-available forms. In order to be able to analyze the complex nature of spectral information obtained from bio-based fertilizers, advanced artificial intelligence techniques (machine learning and deep learning) are being investigated. The analysis techniques are investigated to not only find the correlation between the spectral information and the nutrient of interest, but also to finf the characteristic wavelengths where the correlation is maximum. Finding this position characteristic will help us understand how a particular element interacts with light in different chemical and physical states, and if it is possible to assign a fingerprint region.
![](/english/research/plant_soil/climate-change-mitigation-and-sustainable-soil-management/ferticycle/news/newsletter4/images/khan_photo_300_x_150.png)
Workflow diagram: samples to be analyzed, nutrients of interest, methods of determination, analysis methods.
Another aspect of the project is to investigate the potential of these sensor technologies to quantify the plant's available forms of nutrients (nitrogen and phosphorus). Currently, the near-infrared (NIR) and mid-infrared (MIR) spectroscopies are investigated for a diverse set of bio-based fertilizers. The results suggest that both these sensor technologies have the potential to estimate certain nutrient (N, P, Al, FAA-N, and NH4-N) in bio-based fertilizers while fail to predict (Na, Cu, Cr, Co, Cd, As, and Mn). To improve the accuracy of prediction, a spectral fusion technique (combining NIR and MIR) is realized to find if a multi sensors array is feasible for the accurate estimation of nutrient contents. In future work, the potential of X-ray fluorescence (XRF) and fourier transformed infrared photoacoustics (FTIR-PAS) sensor will be investigated for the prediction of nutrient contents in bio-based fertilizers.
By Khan Wali