Vis enkel innførsel

dc.contributor.advisorTangstad, Merete
dc.contributor.advisorSævarsdottir, Gudrun Anbjørg
dc.contributor.advisorEinarsrud, Kristian Etienne
dc.contributor.authorHoover, Haley
dc.date.accessioned2023-09-25T12:53:38Z
dc.date.available2023-09-25T12:53:38Z
dc.date.issued2023
dc.identifier.issn2703-8084
dc.identifier.urihttps://hdl.handle.net/11250/3091823
dc.description.abstractSilicon is becoming an increasingly important part of the modern world through its applications in electronics, solar cells, and silicones, to name a few. Norway is a significant producer of silicon, competing with much larger nations such as Russia and the United States. As of 2019, Norway was the third largest producer in the world. Production of silicon via carbothermic reduction of quartz (SiO2) is an energy intensive and carbon dioxide (CO2) emitting process. As the industry seeks to reduce both energy consumption and carbon emissions, the understanding of the energy distribution in the furnace is increasingly important. Most of the electrical energy in the furnace will be directed to an arc, which supplies the heat in the lower area of the furnace, where the silicon producing reaction occurs. Electrical current may also travel through the conductive silicon carbide (SiC) crust. Any remaining current must pass through the charge mix, consisting of quartz and carbon materials. As the carbon material enters the furnace, it is heated and exposed to silicon oxide (SiO) gas. Both the changing temperature and the chemical reactions with the SiO gas will transform the material to SiC, changing its electrical properties. There is scarce literature on the resistivity of the materials as they are transformed in the furnace. Thus, the resistivity of the partially transformed carbon materials (char, coal, and charcoal) in the furnace was investigated as the main part of this thesis. Additionally, the transformation of the carbon to SiC was also investigated, as well as the resistivity of charge mixtures that included quartz. Finally, an excavation of a 160 kW pilot scale silicon furnace was presented. The findings in this thesis aim to provide more knowledge to the conductivity of the materials in the furnace as they are transformed. This will ensure that the energy distribution in the furnace maintains the temperature and resistivity profile needed for good operations. Good operations will lead to lower energy consumption and CO2 emissions. Furthermore, the data may be used to improve existing models of the furnace through providing a temperature profile for the conductivity of the charge mix. The transformation of coal, char, and charcoal to SiC was studied through placing a carbon bed above an SiO generator and heating it at temperatures between 1750-1850 °C. The main mechanism of SiC formation followed the grain model for all three carbon materials, where SiC formed along the outer edges of the carbon particle and around the pores. Elemental silicon began to form in the pores in areas with high SiC content, but areas of carbon could still be found on the same particle. It was seen that since the material is transformed in a packed bed, there was unequal access to the SiO gas and a temperature gradient, causing some particles to be more highly transformed than others, even within the same area of the setup. Elemental silicon was produced in varying amounts, with charcoal producing the most and at a lower temperature than the char and coal. Mechanical strength of the materials from coal and char was studied using compression testing. It was seen that the mechanical strength decreased as the coal was converted to SiC, until silicon began to form. However, the opposite was seen in the char and in general, the char was weaker than the coal. Partly transformed charcoal was too weak to be measured. All materials were weaker than the literature reported values for commercial SiC. The resistivity of coal, char, and charcoal was measured between 25-1600 °C in a four-point method setup. The range for all three carbon materials was between 7-20 mΩm at 1600 °C. Charcoal had the highest range, containing both the lowest and highest resistivity at 1600 °C. Two industrial charcoals and one charcoal made from woodchips at 1600 °C were tested. The homemade charcoal was less sensitive to temperature and was conductive at low temperatures. However, at high temperatures it had a higher resistivity then the industrial charcoal. Three types of char were tested, and the range was between 10-15 mΩm at 1600 °C. Both untreated and heat-treated chars were tested, and the lower volatiles in the heat-treated chars lowered the resistivity and made it less sensitive to temperature changes. One type of coal was tested, and it had a range of between 5-10 mΩm at 1600 °C. Density explained the trends at 1000 °C, but at 1500 °C the density could not predict the resistivity. The resistivity of partially transformed carbon materials containing SiC ranging from 30-72 % and Si ranging from 0-36 % was measured between 25-1600 °C. The resistivity decreased with increasing temperature for all materials. The resistivity increased with increasing SiC content, and then decreased as elemental silicon formed. The presence of silicon lowered the resistivity, but this was not related to the quantity of silicon in the sample within the range produced. It was found that above 1300 °C, the SiC crust will be more conductive than the partially transformed SiC material. At lower temperatures, the partially transformed SiC may be as conductive or less conductive than the SiC crust. Resistivity of charge mixtures containing quartz, char, woodchips (charcoal), and silica-iron ore were also measured and were found to be dependent of the amount of char in the sample, as it was the main conductor in these experiments. An excavation of a 160 kW pilot scale Si furnace was conducted with the intention of studying the various zones of the furnace and comparing them with theory as well as results from other silicon furnace excavations. The results were similar to those of other excavations. However, a top cavity was seen high in the furnace and was held together by a layer of loose charge and condensate. The condensate contained silica, alumina and potassium and sodium oxides. No cavity was seen around the electrode tip, but it was likely small during operations and collapsed during cooling. No SiC crust was found, but SiC was seen in other forms throughout the furnace. A large silicon-slag layer was seen, with the slag consisting of silica, alumina, and calcia, usually mixed with SiC. Silicon was also found in droplets above the cavity and in more sizable quantities below the cavity. The resistivity profile of the various zones found in the furnace was also estimated based on the previous work. It predicts that the silicon-slag layer will have the lowest resistivity, followed by the partly reacted charge materials, which will have different resistivities based on the temperature and composition. The condensate layer and areas of inactive charge will likely be nonconductive based on the SiO2 content and the low temperature.en_US
dc.language.isoengen_US
dc.publisherNTNUen_US
dc.titleElectrical Resistivity of Materials in the Silicon Furnaceen_US
dc.typeDoctoral thesisen_US
dc.subject.nsiVDP::Technology: 500::Materials science and engineering: 520en_US


Tilhørende fil(er)

Thumbnail

Denne innførselen finnes i følgende samling(er)

Vis enkel innførsel