Analytische Chemie
Trace element concentrations in gold grains from various geological units in South Africa were measured in situ by field emission-electron probe microanalysis (FE-EPMA), laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS) and synchrotron micro X-ray fluorescence spectroscopy (SR-µ-XRF). This study assesses the accuracy, precision and detection limits of these mostly non-destructive analytical methods using certified reference materials and discusses their application in natural sample measurement. FE-EPMA point analyses yielded reproducible and discernible concentrations for Au and trace concentrations of S, Cu, Ti, Hg, Fe and Ni, with detection limits well below the actual concentrations in the gold. LA-ICP-MS analyses required larger gold particles (> 60 µm) to avoid contamination during measurement. Elements that measured above detection limits included Ag, Cu, Ti, Fe, Pt, Pd, Mn, Cr, Ni, Sn, Hg, Pb, As and Te, which can be used for geochemical characterisation and gold fingerprinting. Although LA-ICP-MS measurements had lower detection limits, precision was lower than FE-EPMA and SR-µ-XRF. The higher variability in absolute values measured by LA-ICP-MS, possibly due to micro-inclusions, had to be critically assessed. Non-destructive point analyses of gold alloys by SR-µ-XRF revealed Ag, Fe, Cu, Ni, Pb, Ti, Sb, U, Cr, Co, As, Y and Zr in the various gold samples. Detection limits were mostly lower than those for elements measured by FE-EPMA, but higher than those for elements measured by LA-ICP-MS.
One important aim of precision agriculture (PA) is the optimization of crop growth by means of site-specific measures, e.g. fertilizer application. Thus, PA should contribute to a resource efficient and ecofriendly soil management. Due to the expenses associated with traditional methods of soil analysis, requiring sample collection and laboratory analysis, PA technologies are still not in widespread use. Therefore, the aim of the project “I4S – Integrated System for Site-Specific Soil Fertility Management” is the development of a field-deployed, sensor-based analysis system offering rapid, cost-effective and spatially-resolved fertilizer recommendations.
In this system, laser-induced breakdown spectroscopy (LIBS) could be ideally suited to assessing elemental nutrient contents of soils. In addition to low cost, durability and small size, a reliable quantification procedure is a crucial requirement for such a system. However, the texture as well as the composition of the soil can affect the spectra. This matrix dependence is the key challenge to be addressed in the application of LIBS for soil evaluation. The focus of this work was the establishment of a LIBS method for soil analysis in a laboratory environment for future field application.
Natural soil samples of various origins, textures and compositions were used to characterize the matrix dependence of the LIBS spectra. Reference samples were prepared by adding defined amounts of the target elements to the soils (standard addition). Signals not superimposed by peaks of other elements were identified for each element. The reference samples also provided calibration curves for the respective soil type when the initial concentrations in the soils were taken into account. Additionally, the common laboratory method ICP-OES following aqua regia extraction was used to obtain reference values. Various approaches of calibration free evaluation of the data were also evaluated. In addition to traditional single-pulse experiments, the advantages of dual-pulse LIBS in relation to signal intensity, reproducibility as well as overcoming the matrix dependence of soil spectra were investigated. These methods were subsequently applied to validation samples collected on a dense grid within a field.