Optimization of parameters in SiC production via Acheson Process – a hot model study

Abstract

Silicon carbide (SiC) is produced industrially in a heat resistance furnace using quartz (silica) and raw petroleum coke (RPC) as the raw material. In this process, commonly known as the Acheson Process, a large current is passed through a graphite resistor (heating element) located at the center of the furnace which is surrounded by the mixture of the raw material. When the resistor temperature (i.e. core temperature) reaches to the reaction temperature, which is 15150C or above, silica and carbon react to form SiC and carbon monoxide. The reaction is highly endothermic in nature. The major drawbacks of the process are (a) lower production efficiency as about 15% (by weight) of the raw material get converted into SiC (theoretically it can go up to 41.6%), and (b) inefficient in terms of energy uses. The key parameters which affect this process are (a) core temperature, (b) reaction time, (c) composition of the raw material, (d) quality of the raw material such as particle sizes and impurities, and (e) the mode of power supply. No systematic study is available in the open literature for this process. Therefore, the process has been investigated experimentally at laboratory scale using a hot model setup. In this work, the effect of the following parameters has been studied, (a) core temperature, (b) experimental duration (i.e. time), and (c) the composition of the raw material. All the experiments have been carried out in a small version of an industrial furnace which was fabricated in the laboratory. The cylindrical chamber of the furnace (housing the raw material with a heating element at the center) has 20 cm internal diameter and 47 cm length. The power supplied to the heating element was controlled using a variac transformer whose voltage and current ratings were continuously adjusted. Carbon (99% pure) was procured from Rhodium ferroalloys, Bangalore and silica (99% pure) from Arjun minerals and chemicals, Bangalore. An in-house (custom built) assembly was designed to measure the core temperature using a pyrometer (LumaSense technologies, UK). The accuracy of the measurement was ±0.7% of the reading as quoted by the manufacturer. Temperature away from the core were measured using B-type (Pt-6%Rh, Pt-30%Rh) thermocouples (Chanda electrothermy private limited, Bangalore). These measurements helped to understand the progress of the reactions. For safety purpose, CO concentration was also monitored continuously. The quality of the products obtained was analyzed using three different techniques. First, the phases present in the samples were qualitatively identified using X-Ray Diffraction (XRD) technique. Here, it was observed that the peak height of the unreacted materials increases in the product from core to the periphery of the furnace. Second, the surface morphology and particle size of the products were examined using Scanning Electron Microscope (SEM) in which it was observed that the particle size of the product decreases when moving from core to periphery of the furnace. Additionally, Energy Dispersive Spectroscopy (EDS) was done to estimate the atomic percentage of different elements present in the sample. Third, the amount of SiC and other phases were quantitatively determined using chemical analysis. The experiments were carried out in three parts. In PART A the core temperature of the furnace was varied between 1800 and 21000C while keeping the other parameters constant; thereby optimizing the core temperature. In PART B the operation time of the experiment was varied in the range of 3 to 10 hours. In PART C the composition of the raw material was changed. In this part, excess silica and carbon (above the stoichiometric quantity) were independently varied in the charge in the range of 0 to 20%. SiC was deliberately mixed with the raw material in some experiments. In the last phase of this study, an experiment was carried out using the optimized parameters (obtained from PART A, B, and C study along with optimized SiC content in the charge). Generally, it was observed that the percentage (by weight) of SiC is decreased gradually when one moves away from the surface of the heating element. In combination with the chemical analysis, the experimental results obtained in PART A predict 19000C to be the optimum core temperature. These experiments were conducted for 3 hours where the raw materials were mixed in the stoichiometric ratio. SiC percentage was nearly zero for samples collected at and after a radial distance of 5 cm from the heating element. At a fixed distance, SiC percentage is increased as the core temperature increases until 19000C, after which a decreasing trend was observed. These results prove that 19000C is the required core temperature at which maximum SiC is produced in the present experiments. Experiments in PART B were conducted at 19000C core temperature and stochiometric charge composition. The mass of SiC was found to increase gradually when experimental duration was changed from 3 to 6 hours. However, after this time duration, SiC obtained did not change even for those experiments which were lasted till 10 hours. These results were verified using the chemical analysis. These results indicate that 6 hours is the optimum experimental operation time. In PART C, the experiments were carried out at 19000C core temperature for three hours. When excess silica was added in the raw material, it was observed that SiC yield is reduced significantly (compared to the experiment with stoichiometric ratio composition) with increased energy consumption, probably due to the excessive formation of SiO(g) which escaped out of the furnace. On the other hand, experiments with excess carbon yielded higher SiC percentage. Similar results were obtained when SiC was added in the stoichiometric ratio of the raw material, probably because it enhanced the thermal conductivity of the charge. Finally, the experiment, where 20% excess carbon and 10% SiC in the raw material was used, produced the best result in terms of SiC percentage. The optimum charge composition is, therefore, ‘20% excess carbon + 10% SiC’. Finally, an experiment (called as an optimized experiment) was carried out using all optimum parameters (19000C core temperature, 6 hours experimental time duration, and 20% excess carbon & 10% SiC). In this experiment, SiC was obtained up to 6 cm radially from the heating element. The results of this experiment are compared with industrial results. The weight percentage of SiC obtained in the optimized experiment was more than two times higher and the energy consumed was lower than those obtained in the industries. Both in terms of weight percentage and energy consumed, the optimized experiment performed exceedingly well compared to the industrial practice.