『Abstract
　The increase of traffic and the rising energy consumption mean a challenge to the air pollution control and to environmental protection. Measures of air pollution control concentrated primarily on the reduction of gaseous pollutants. However, in the field of air hygiene in Central Europe, especially the load of near-surface atmospheric dust becomes threatening to human health. A SIMS microprobe for ultra fine feature analysis is used to image the elemental composition at the surface of submicrometer urban dust particles collected at two measurement stations in the Grand Duchy of Luxembourg. The NanoSIMS 50 has been chosen because it creates one intensity image for each selected element in a high spatial resolution down to 50 nm. The atmospheric fine dust consists of a mixture of organic and inorganic compounds. The elemental composition at the surface of particles was studied using a global image segmentation technique to separate the signal from the background of the particles. The analysis of the binary intensity images was carried out using several shape and proximity measures. The patch shape complexity and distribution for industrial/urban particles were found to differ significantly from the solids collected from a forest site. We conclude that the methodology developed in the study is a reliable tool to differentiate between potential sources of airborne particulate matter.

Keywords: Atmospheric dust; Chemical mapping; Grand Duchy of Luxembourg; NanoSIMS 50; Proximity indices; Shape indices; Source ascription』

1. Introduction
2. Materials and methods
　2.1. The SIMS microprobe for ultra fine feature analysis - NanoSIMS 50
　2.2. Image segmentation of chemical patches and estimation of shape and proximity indices
3. Results
　3.1. Chemical mapping on the surface of dust particles
　3.2. Distribution of different chemical elements on the surface of particles
　3.3. Interpretation of shape and proximity indices to distinguish between material sources
4. Conclusions
Acknowledgements
References