Automated Electron Microscopy for Mineralogical Characterization
- PDF / 811,165 Bytes
- 14 Pages / 612 x 792 pts (letter) Page_size
- 36 Downloads / 236 Views
Automated Electron Microscopy for Mineralogical Characterization Rolando Lastra © Department of Natural Resources Canada, (2011). All rights reserved
CANMET, 555 Booth St., Ottawa, Ontario, K1A 0G1, CANADA ABSTRACT Full automation of electron microscopy is now a mature technology that is being applied internationally mainly for mineralogical characterization. This technology has increased the speed and reliability of the characterization of ores and mineral processing products. It allows developing the most appropriate beneficiation technology for a new ore body. It helps to determine the potential, the optimization and the limitations of mineral concentrator plants. It can also be applied to the betterment of the environmental management of the metallurgical residues. This presentation will discuss the main approaches for fully automated electron microscopy. Additionally, an application case is presented, focusing on the characterization of complex ore of rare earth minerals. PRINCIPLES OF IMAGE ANALYSIS Image source Traditionally, the petrographic optical microscope has been used for mineralogical characterization of samples mounted on thin or polished sections. Therefore, the early image analysis systems to discriminate between minerals were interfaced to petrographic optical microscopes. However, the results are not satisfactory when a black & white (B&W) video camera is used to capture the image. This is because in reflected light, the grey levels of many minerals overlap and it is not possible to discriminate between the minerals. Figure 1 shows a typical image of a polished section with quartz, pyrite, chalcopyrite, sphalerite and secondary copper minerals (covellite, digenite, bornite, etc). Figure 1 includes a histogram of the grey levels of the minerals in the sample. The histogram clearly shows strong overlaps in the grey levels of sphalerite and the secondary copper minerals. Also, there is a strong overlap between pyrite and chalcopyrite. Therefore, there is not enough difference in the grey levels of the minerals to facilitate their discrimination. For the case of B&W images of transmitted light optical microscopy, there are even less ore minerals that can be discriminated. The development of colour video cameras of high quality has provided great advantages. With colour images there are in fact three signals (RGB). Thus overlapping of the intensity levels of the minerals in all the three RGB signals is less common. Several basic precautions must be taken when using colour cameras. These take into consideration the stability of the illumination source and the stability of the camera [1,2]. With the availability of high quality colour cameras, many researches have tried again to develop image analyzers interfaced to optical microscopes to discriminate minerals. Nevertheless, the results are still not completely satisfactory. The problem is that there are parts of some mineral that are discriminated and assigned to other minerals and vice versa [3]. Despite this problem, it is claimed [3] that for the d
Data Loading...
