Artificial neural network for on-site quantitative analysis of soils using laser induced breakdown spectroscopy

Josette El Haddad (1) , M. Villot-Kadri (2) , Amina Ismael (2) , G. Gallou (2) , Karine Michel (3) , Delphine Bruyère (3) , Valérie Laperche (3) , Lionel Canioni (1) , Bruno Bousquet (1)

1 LOMA - Laboratoire Ondes et Matière d'Aquitaine
2 IVEA Solution
3 BRGM - Bureau de Recherches Géologiques et Minières (BRGM)

Nowadays, due to environmental concerns, fast on-site quantitative analyses of soils are required. Laser induced breakdown spectroscopy is a serious candidate to address this challenge and is especially well suited for multi-elemental analysis of heavy metals. However, saturation and matrix effects prevent from a simple treatment of the LIBS data, namely through a regular calibration curve. This paper details the limits of this approach and consequently emphasizes the advantage of using artificial neural networks well suited for non-linear and multi-variate calibration. This advanced method of data analysis is evaluated in the case of real soil samples and on-site LIBS measurements. The selection of the LIBS data as input data of the network is particularly detailed and finally, resulting errors of prediction lower than 20% for aluminum, calcium, copper and iron demonstrate the good efficiency of the artificial neural networks for on-site quantitative LIBS of soils.

Mots clés: Laser-induced breakdown spectroscopy (LIBS); Soil; Quantitative analysis; Artificial neural network

Domaines: Optique [physics.optics]

Josette El Haddad, M. Villot-Kadri, Amina Ismael, G. Gallou, Karine Michel, et al.. Artificial neural network for on-site quantitative analysis of soils using laser induced breakdown spectroscopy. Spectrochimica Acta Part B: Atomic Spectroscopy, 2013, 78-79, pp.51-57. ⟨10.1016/j.sab.2012.11.007⟩. ⟨hal-00788952⟩