Abstract
In the production of metallurgical silicon, one of the byproducts is silicon dust. It has been reported that each ton of melted silicon causes the emission of 900 kg of dust. The significant amount of dust produced has a negative effect on the health of the factory personnel and the environment and leads to the depreciation of the factory equipment. On the other hand, the consumption of fossil fuels has greatly increased due to the rapid growth of the world population and industrialization. During combustion, the nitrogen and sulfur compounds in the fuel cause the emission of NOx and SOx, and these emissions cause acid rain, fog and global warming. Removing sulfur and nitrogen compounds from fossil fuels is important because of their harmful effects on the environment and human health. The aim of the present research is to use the silicon dust coming out of the metallurgical silicon production plant as an adsorbent for the adsorptive removal of indole and quinoline from the n-heptane model fuel. Modeling of equilibrium surface adsorption data was done with the help of Langmuir and Freundlich isothermal models. Also, the kinetics of adsorptive denitrogenation process was evaluated. From the results of adsorption, it was found that silicon dust removed 350 ppm of indole and quinoline in 120 min time, temperature of 20 ºC, with 0.1 g adsorbent per 10 mL of n- heptane fuel containing quinoline and indole, to the extent of 22.65, and 17.26 mg.g-1, respectively. The adsorption of nitrogen compounds in the fuel model showed the best fit with the Langmuir isotherm for quinoline and indole with the maximum adsorption capacity of 24.33 and 18.21 mg.g-1. Based on the experimental data, the pseudo-second order model showed the best fit for quinoline and indole with minimum square error of 0.9876 and 0.9999. From the isothermal and adsorption kinetic studies, it was found that the exhaust dust from metallurgical silicon production factories shows great potential in removing nitrogenous compounds.
Introduction
Crystalline silicon and its alloys are used in various industries such as aluminum, chemical, aviation, and automotive (Sizyakov, 2016). Silica fume is composed of very fine amorphous SiO2 particles and is produced as a byproduct in the production processes of silicon metal and ferrosilicon. During the carbothermal reduction of silica to silicon in an electric furnace, silicon monoxide gas is also produced at a temperature usually over 1800°C. Silicon monoxide comes out of the furnace with other gases and mixes with air and oxidizes and forms fine silica particles. Condensed silica, which is usually non-crystalline, is then collected with other dust and dense particles in the gas cleaning system (Sikarwar, 2023). It has been reported that each ton of melted silicon may emit 900 kg of dust (Leonova, 2019; Nemchinova, 2019). The significant amount of dust produced has a negative effect on the health and work ability of the factory personnel and leads to rapid depreciation of the factory equipment (Leonova, 2020). The consumption of fossil fuels has greatly increased due to the rapid growth of the world population as well as industrialization. These fuels contain the most harmful pollutants, i.e. sulfur and nitrogen compounds (Yang, 2004; Qu, 2016). Removing sulfur and nitrogen compounds from fossil fuels has attracted considerable attention due to its harmful effects on the environment and human health (Bauserman, 2008; Gosu, 2022). Hydrodenitrogenation (HDN) is currently used in refineries to reduce the nitrogen content of liquid fuels (Sikarwar, 2019). Hydrodenitrogenation is a costly and energy-consuming process. In addition, HDN is a slow kinetic process and requires more energy (Ahmadi, 2017). Therefore, an alternative technology to reduce the nitrogen content of liquid fuels should be explored. Adsorption is one of the alternative methods and plays an important role in removing undesirable compounds (Jawad, 2022; Reghioua, 2021; Jawad, 2020; Gosu, 2020). The purpose of this research is to use the silicon dust from the metallurgical silicon production plant as an adsorbent for the adsorptive removal of indole and quinoline from the n-heptane and to determine the characteristics of the adsorbent with different methods of determining characteristics such as FTIR, XRD, XRF, and BET. Modeling of equilibrium adsorption data was done with the help of different isothermal models. Also, the kinetics of adsorptive denitrogenation process was also evaluated.
Methodology
The n-heptane (n-C7H16), as model fuel (purity 99%), indole (purity 99% by weight) and quinoline (purity 99% by weight) were purchased from Sigma Aldrich. Silicon dust was obtained from metallurgical silicon production plant of Araz Shahr-Khoi Silicon Company located in West Azarbaijan province. The elements mass fraction of adsorbent was analyzed by Thermo Scientific XRF. For silicon dust phase analysis, X-ray diffraction (XRD) and variable diffraction angle 2θ was used in the range of 5 to 80 degrees. FTIR infrared spectrum was used in order to determine the surface functional groups. Physical adsorption of nitrogen at 77 K temperature was performed by BET analysis (Bel, Sorp, Japan) to calculate surface characteristics including BET surface area, pore volume, average pore size and porosity. The mother solution of the model fuel was prepared by dissolving the nitrogen compound of indole and quinoline in n-heptane. Model fuel with a certain concentration of indole or quinoline nitrogen compound was made by diluting the mother solution step by step with n-heptane. For each experiment, model fuel with a certain concentration of indole or quinoline was added to the required amount of adsorbent and the resulting mixture was stirred at different temperatures and times. Then, the solution was filtered from the solid phase using a syringe filter and its concentration was determined using a UV spectrometer. UV absorption at 313 nm and 287 nm was used to determine quinoline and indole concentrations, respectively. Adsorption isothermal models such as Langmuir and Freundlich were used in order to investigate the equilibrium data for the removal of indole and quinoline around silicon dust. The present research analyzes the experimental data with two well-known kinetic models; pseudo-first-order and pseudo-second-order models. The first and second order pseudo model cofficients are calculated by performing non-linear regression with experimental data (Sikarwar, 2023).
Conclusion
In this study, the silicon dust coming out of the metallurgical silicon production unit was used as an adsorbent in the denitrogenation of indole and quinoline from n-heptane. Factors such as process time, temperature, adsorbent dosage, and initial concentration of nitrogen compounds are respectively in the range of 0 to 120 min; 20 to 50 ºC, a 0.05 to 0.3 g and 100 to 350 ppm were investigated. From the results of adsorption, it was found that silicon dust removes 350 ppm of indole and quinoline in 120 min, 20 ºC, with 0.1 g per 10 mL of n- heptane fuel containing quinoline and indole, with adsorption capacities of 22.65 and 17.26 mg.g-1. The square values of the errors in the adsorption tests for Langmuir and Freundlich isotherms were 0.999 and 0.9666 for indole and 0.9928 and 0.914 for quinoline, respectively. Therefore, the data have a better fit with the Langmuir isotherm, and the silicon dust surface can be assumed to be homogeneous and the adsorption to be one layer. Also, the maximum adsorption capacity based on the Langmuir isotherm was calculated as 33.24 and 21.21 mg.g-1 for indole and quinoline, respectively. The squared values of the errors for the kinetics of the adsorption based on the pseudo-first-order and the pseudo-second-order model were 0.9787 and 0.9998 for indole and 0.9255 and 0.9992 for quinoline, and the second-order kinetic model was a better fit. The adsorption of indole and quinoline around the silicon dust gradually increased with the increase of time and reached more than 95% and 93% for indole and quinoline, respectively in 30 min and after 120 min the adsorption was almost stable. With the increase of adsorptive denitrogenation temperature, the adsorption capacity for indole and quinoline decreased, because the adsorption of quinoline and indole is exothermic. By increasing the silicon dust dose from 0.05 to 0.1 g, the adsorption capacity of indole and quinoline increased, and then it decreased by increasing the adsorbent dose from 0.1 to 0.3 g. In addition, the interaction between indole and silicon dust was more effective compared to quinoline, and a lower amount of silicon dust was needed to remove indole compared to quinoline. For the indole in the range of concentrations less than 200 ppm, the rate of adsorption capacity was higher compared to concentrations greater than 200 ppm. Since, the increase in the concentration gradient at the beginning of the adsorption. Also, its reason can be considered due to the reduction of spatial barriers between the indole and the adsorbent due to the decresae of the pore space. For quinoline, the adsorption capacity increased continuously by increasing the concentration from 100 to 350 ppm.
Keywords
Silicon dust; Adsorptive denitrogenation; Indole; Quinoline; Fuel.