Iron and Copper Oxide Molecular Emissions Detection in Flash Smelting Processes

Additonal authors: Toro, C.A.. Book title: Proceedings of the 58th Conference of Metallurgists Hosting Copper 2019. Chapter: . Chapter title:

Proceedings, Vol. Proceedings of the 58th Conference of Metallurgists Hosting Copper 2019, 2019

Torres, S. N.

In this paper, we report on the detection of spectral molecular emission of iron oxide (FeO) and copper oxide (CuxO) emitted by reactions occurring in a laboratory scale flash smelting process of pyrite (FeS2) as well as copper concentrate. The novelty of the technique is that, the molecular spectral emission sensing is achieved by using a low spectral resolution and low sensitive radiometer, enhanced with signal processing and matrix factorization techniques. The method was validated using a highly sensitive spectrometer, with a high spectral resolution; achieving a similar FeO spectral pattern as the one obtained after signal processing techniques over low resolution spectral signals. INTRODUCTION Visible-near infrared (VIS-NIR) spectroscopy is an optical contactless sensing technique commonly used to capture information about the emitted, absorbed, transmitted or scattered radiation in a physical or chemical process, with the presence or absence of an excitation light source such as a laser or a broad spectral band source. These techniques coupled with signal processing and machine learning algorithms, sometimes referred as chemometrics (Pisapia et al., 2018), can be found in processes such as: pharmaceutical (Kandpal et al., 2016); food (Coelho et al., 2016); combustion (Ballester & García-Armingol, 2010); siderurgy (Strakowski et al., 2014) industry, among others. In the extractive copper industry there are several technological developments using optical sensing techniques aimed to retrieve information about copper ore and concentrate mineralogical composition such as X-Ray Fluorescence, Qemscan® (Quantitative Evaluation of Minerals by Scanning Electron Microscopy) and X Ray Diffraction (XRD) analysis, however, because of the manual procedures to analyze the samples with these laboratory instrumentation, only discrete information during a daily based operation can be retrieved, making impossible for the operators to take decisions over the involved processes in real time. The aim of our research work presented here is to contribute with sensing and data analysis procedures based on VIS-NIR spectroscopy for the flash smelting process characterization, particularly in this paper, with the detection of FeO and CuxO molecular emissions in a laboratory scale setup.
Mots Clés: Copper 2019, COM2019