Quantitative multiple-element simultaneous analysis of seaweed fertilizer by laser-induced breakdown spectroscopy

Area of Science:

• Analytical Chemistry
• Agricultural Science
• Spectroscopy

Background:

• Seaweed-based fertilizers are valuable nutrient sources.
• Accurate elemental analysis is crucial for fertilizer quality control.
• Advanced chemometric methods can enhance spectroscopic analysis.

Purpose of the Study:

• To quantify multiple elements in seaweed-based fertilizer.
• To evaluate the performance of different chemometric models (PLSR, SVR, RF) for elemental analysis.
• To determine the optimal multivariate model for specific elements.

Main Methods:

• Laser-induced breakdown spectroscopy (LIBS) was employed for elemental detection.
• Partial Least Squares Regression (PLSR), Support Vector Regression (SVR), and Random Forest (RF) were used for chemometric analysis.
• Model performance was assessed using root mean square error (RMSE) and relative standard deviation (RSD).

Main Results:

• PLSR provided optimal results for Mg, K, and P.
• RF achieved the best predictions for Mn, Cu, Sr, and Ca.
• SVR was most effective for Ba quantification.
• Low RMSE and RSD values were reported for all analyzed elements.

Conclusions:

• The choice of multivariate model significantly impacts the accuracy of elemental quantification in seaweed fertilizer.
• LIBS coupled with appropriate chemometrics offers a rapid and effective method for multielement analysis.
• This approach supports quality control and optimization of seaweed-based fertilizers.