Laserinduced fluorescence as sensor for mining : a guide for geologists and mining engineers

The suitability of laserinduced fluorescence as a mining sensor has been studied in this thesis. Laserinduced fluorescence is a long known phenomenon, which finds main applications in combustion analysis and medicine. Utilisation of LIF on minerals is based on a fingerprinting approach, meaning that calibration of the system to the minerals which have to be sorted is always necessary. Over a three year period, measurements on different raw materials have been performed as well as investigations on disturbance factors. Additionally, concepts for testing and application have been developed. Principally, two different type of tasks can be carried out with LIF. One is the separation of material groups, the other is a quantitative correlation between fluorescence characteristics and mineral contents. For the first type, usage of classification systems such as support vector machines is advised, if a large number of different materials has to be separated. In other cases, where only two or three groups are to be distinguished, choice of suited fluorescence wavelength can normally be carried out by hand. For the quantitative correlation of fluorescence to mineral contents, several samples with known mineralogical composition have to be measured and a correlation between certain wavelength ratios and the desired mineral has to be established. For both applications, though, detailed testing and technical layout is a must for successful implementation. The great advantages of LIF are on the one hand the high information content, based on the combination of wavelength and time resolved measurements. On the other hand, LIF shows only little sensitivity to disturbance factors such as dust and moisture compared to other optical sensors in the visible spectra. As disadvantages, LIF is not well established in the mining industry and has to be setup for each application. Although central components needed for a LIF system are available from different manufacturers, only few companies offer complete LIF systems. Also integration of a LIF system in the mining environment is technically complex and costly. Consequently, LIF should only be considered, if more established sensor systems can not solve the sorting task in question. Notwithstanding the discussed limitations and challenges, the results presented in this thesis clearly document the suitability of LIF for different sorting tasks in the mining industry. Especially for the case of lignite and associated waste rocks, sorting of a large number of materials with LIF resulted in high recognition results which have to be challenged by other sensor systems. Future developments and applicability of LIF systems in the mining industry strongly depend on advancement of the laser as well as optical systems.

Grantor dc:publisher

Publikationsserver der RWTH Aachen University

Year dc:date

2008

Author dc:creator

• Vinzelberg, Gero

Contributors dc:contributor

• Nienhaus, Karl

• info:eu-repo/classification/ddc/550
• Berührungsloser Sensor
• Laserinduzierte Fluoreszenz
• Mustererkennung
• Geowissenschaften
• Bergbausensor
• sensor
• laserinduced fluorescence
• pattern recognition

Statement dc:rights

• info:eu-repo/semantics/openAccess

Language dc:language

eng

Identifier

https://publications.rwth-aachen.de/search?p=id:%22RWTH-CONV-112611%22

OAI record oai:identifier

https://publications.rwth-aachen.de/oai2d?verb=GetRecord&metadataPrefix=oai_dc&identifier=oai%3Apublications.rwth-aachen.de%3A50047