TY - GEN A1 - Chao, Madlen T1 - Influence of moisture and grain sizes on the analysis of nutrients in agricultural soils using LIBS N2 - Over the last few years, there has been a growing interest to apply spectroscopic methods to the agricultural field for better understanding of soil properties and for efficient, sustainable management of arable land. Within the project I4S (intelligence for soil), funded by the BMBF, an integrated system for site-specific soil fertility management is developed, consisting of different sensors like X-Ray fluorescence analysis (XRF), near-infrared spectroscopy (NIR) and laser-induced breakdown spectroscopy (LIBS). LIBS provides a fast and simultaneous multi-element analysis with little to no sample preparation, which makes it a suitable method for real-time analysis on the field. The quantification of macro and micro nutrients in soils with LIBS is challenging due to matrix effects, different levels of moisture content and varying grain sizes. First studies revealed that the problems with matrix effects can be overcome by using well characterised soils as reference materials and chemometric tools like Partial Least Squares Regression(PLSR) for calibration. The next step was to investigate the influence of moisture and grain sizes on the LIBS signal, which is a big issue when measuring directly on the field. The results showed that the LIBS signal decreases exponentially with increasing moisture content, as most of the laser energy is used for vaporising the water. With moisture contents of 30 % or higher almost no signal can be detected. This decrease is more severe for sandy soils than for clay soils. First tests of different grain size distributions indicate that the variation of the LIBS signal increases with growing amounts of larger grains. This results in a higher standard deviation, because of a poorer reproducibility of the plasma formation and plasma characteristic. With the help of chemometric tools the influence of moisture and grain sizes should be implemented in the calibration model for accurate analysis of nutrient composition in agricultural soils. T2 - 14. Kolloqium Ak Prozessanalytik CY - Hannover, Germany DA - 03.12.2018 KW - Soil KW - LIBS KW - Moisture KW - PLSR KW - Grain size PY - 2018 UR - https://opus4.kobv.de/opus4-bam/frontdoor/index/index/docId/47056 AN - OPUS4-47056 AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany