TW202428513A - Low carbon emission process for the production of silicon - Google Patents

Abstract

A process is disclosed of making silicon by carbothermic reduction of silica having a low carbon footprint, low NO _xemission, and reduced resource usage in the form of carbon. Also disclosed is a carbon capture method for a silicon making process by carbothermic reduction of silica using a combination of thermal plasma and high-pressure disproportionation of CO(g).

Description

Low carbon emission processes for silicon production

The subject matter of the present invention relates to the production of elemental silicon and, more particularly, to the production of elemental silicon in an arc furnace.

Elemental silicon has many industrial applications, such as steel, automotive, microelectronics and the solar industry. Silicon in the form of silicates accounts for 90% of the Earth's crust, making it the second most abundant element in the Earth's crust. However, silicon in the form of silicates cannot be used directly for industrial purposes. Silicon in the form of, for example, quartz, _SiO2, must be reduced to its elemental form, i.e. Si.

Elemental silicon used in industrial applications can have a wide range of purities. For example, as an alloying element in steel, silicon with a purity between 95% and 99% is usually used. This is called metallurgical grade silicon. For advanced applications, such as microelectronics, high purity silicon is required, typically 99.9999999%.

The production of elemental silicon relies on the production of metallurgical grade silicon, which is produced by the carbothermal reduction of silicon dioxide, which is carried out by an arc melting process. In this process, quartz _SiO2 is reduced to silicon in the presence of a carbon source such as coal, petroleum coke or charcoal by heating _it to high temperatures of over 1800°C in an arc furnace.

In this process, carbon reacts with oxygen in silicon dioxide SiO ₂ to produce gaseous carbon monoxide and elemental silicon. The overall stoichiometric reaction to form silicon is as follows. SiO ₂ (s) + 2 C(s) = Si(l) + 2 CO(g)

This process is done with excess silicon dioxide to prevent the formation of silicon carbide SiC. 2 SiC + SiO ₂ + SiC = 2 Si + 2 CO

These reactions are highly endothermic and occur at temperatures exceeding 1800 degrees. Due to the endothermic nature and the high temperatures required, the process requires a large amount of electricity, 13 to 15 kWh per kilogram of silicon produced on an industrial scale.

The process is carried out in an electric arc furnace (EAF) where the energy is provided by a submerged electric arc, meaning the arc is generated and maintained in a reacting load of quartz and carbon.

The maximum temperature in the furnace is well over 2000°C, which can be observed where the arc is maintained deep inside the load. At this point in the furnace, the reduction reaction of silica produces CO(g), which moves upward and leaves the load at an elevated temperature of 1000 to 1300°C [1].

Due to the size and complexity of this type of furnace, it is built with an open top, which means that air can easily penetrate the furnace and the oxygen in the air reacts with CO(g) in a high temperature process to produce CO ₂ (g). By reacting with the oxygen, the nitrogen in the air is heated to its reactive state, thereby producing NO _x . Both CO ₂ (g) and NO _x are pollutants to the environment, with CO ₂ (g) contributing significantly to global warming as it is a potent greenhouse gas, and NO _x causing acid rain, eutrophication, photochemical air pollution, and depletion of the ozone layer, and being harmful to human health [Reference 2].

Direct emissions of CO ₂ (g) can be as high as five kilograms per kilogram of silicon produced in a commercial silicon plant. The global CO ₂ footprint resulting from silicon production is 11.3 kg _e per kilogram of silicon produced.

In addition to process emissions of carbon in the form of _CO2 (g), initial carbon is a non-renewable resource that is consumed by carbothermal reduction of silicon dioxide. Common carbon sources in the silicon manufacturing industry are: coal, coal coke, petcoke, wood chips and charcoal. The first three can be classified as fossil carbon materials, the rest are called biogenic materials. In silicon plants, it is common practice to use a mixture of the above carbon sources, of which the majority (>60wt.%) is of fossil type, namely coal, coal coke and petcoke. On average, in order to produce one kilogram of silicon in a furnace, about 1.5 kilograms of carbon are irreversibly consumed.

Therefore, there is a need to develop silicon production processes that reduce carbon dioxide ( _CO2 ) emissions, nitrogen oxide (NOx ₎ emissions and reduce the use of non-renewable carbon.

_{NOx emissions from industrial processes can be reduced by capturing NOx in} the exhaust gas stream through a selective catalytic reduction process (SCR) or a selective non-catalytic reduction process (SNCR). However, due to the highly dilute gases coming from the open furnace tops used in industrial processes, _NOx abatement towers become too large and expensive.

CO ₂ can be captured from silicon furnaces using, for example, an amine process. However, this process becomes economically unviable due to the high dilution of CO ₂ in air. The most advanced flue gas capture solvents currently used to absorb CO ₂ from industrial gas streams are aqueous solutions of amines, especially monoethanolamine. Even in the best case, the main energy required to remove CO ₂ from the amine-rich stream (115 to 140 kJ/mol CO ₂ ) dominates the energy requirements of the process [Reference 5]. The main limitations of this process are: high energy intensity, also known as energy losses; and losses of absorbent, since all the absorbent entering the stripper cannot be regenerated [Reference 6]. Ethanolamine also presents certain health risks, as it can cause damage to the liver and kidneys, and at high exposures, to the nervous system. Also, the captured _CO2 must be isolated and stored or used off-site, resulting in additional _CO2 emissions, leaks, and inefficiencies in the transportation process.

Several new industrial measures have been developed to reduce CO ₂ emissions from silicon production processes. For example, it has been proposed to replace non-renewable carbon with biochar produced by thermal cracking and to use closed arc furnaces, thereby limiting the oxidation of CO to CO ₂ . The CO can then be used for other industrial applications [Reference 7].

The use of biochar has many disadvantages, including: the high energy requirements required to produce biochar from biomass such as wood, the limited capacity of the earth to produce biomass, the displacement of useful land for agriculture, and the fact that carbon thermal reduction of biochar ultimately results in the release of _CO2 into the atmosphere.

Elkem, one of the world's major silicon manufacturers, is working with the National Taiwan Normal University (NTNU) and SINTEF to investigate waste gas recycling as a possible way to improve carbon capture processes by increasing the CO ₂ concentration in the waste gas [Reference 8]. In conventional silicon production processes, CO ₂ concentrations are relatively low, only a few percentage points, making CO ₂ capture processes economically unfeasible. By recycling the waste gas back into the furnace, the CO ₂ concentration can be increased by more than 20%, making conventional CO ₂ processes economically affordable. However, it is unclear how to avoid air infiltration in silicon smelting furnaces using open submerged arc furnaces. One proposal is to use a closed arc furnace, which is described below in PCT Publication No. WO 2020/243812 A8. However, even if _CO2 is captured efficiently and economically feasible, it cannot be used in the silicon production process and needs to be isolated and stored.

In U.S. Patent No. 4,860,096, a closed furnace equipped with a transferred arc plasma arc torch is proposed to produce silicon from traditional raw materials. The inventor claims that the use of a transferred arc torch instead of a graphite electrode is advantageous because theoretically less energy is required to carbon-thermally reduce _SiO2 to Si compared to a transferred plasma arc. In addition, the inventor claims that a higher quality silicon product is expected because the use of graphite electrodes is avoided. However, the actual yield of reducing _SiO2 to Si using this process is unknown, in contrast to the well-proven submerged arc process that produces a conversion rate of 85 to 90%. This yield parameter is very important for estimating the energy efficiency of the process. Another disadvantage of this furnace is the need for water cooling to operate the transferred plasma arc torch. Water cooling could result in water leaking from the torch onto the molten silicon bath, which could cause a catastrophic steam hammer and explosion. In addition, the potential for scaling up such furnaces is questionable, as commercially available transferred arc torches are rated at a few megawatts at best, much lower than the hundreds of megawatts required for full-scale silicon production.

Enclosed electric arc furnaces (EAFs) have been proposed, for example, in the previously mentioned PCT publication No. WO 2020/243812 A8, which describes a consumable electrode vacuum arc furnace, and more specifically, a direct current consumable electrode vacuum arc furnace, in which water cooling is generally not required to cool the electrodes or any other components of the furnace, including the furnace shell, flange ports and electrical connections.

This type of closed furnace minimizes the formation of CO ₂ (g) by minimizing the infiltration of air into the furnace so that the furnace gas is enriched in CO (g).

CO(g) in its raw form is considered to be a building block for many useful chemicals, including but not limited to methanol, ethanol and formic acid. Therefore, from an environmental point of view, the main advantage of such a closed arc furnace over an open arc furnace is that the closed arc furnace allows the recovery of CO(g) which is the main by-product of the carbothermal reduction of the silicon production process. However, from the point of view of global economy and market needs, this first step does not provide a viable solution for a low carbon emission silicon process, since the CO(g) produced by this method can only be used in other industrial processes or secondary processes. For example, in order to produce useful chemicals such as methanol by catalytic synthesis, strict requirements are imposed on the quality, availability and consumption rate of CO(g). Therefore, under this approach, the CO(g) generated by the production of silicon needs to be stored and further upgraded until it can be used in secondary processes anytime and anywhere according to market demand and CO(g) shortage. Therefore, it can also be expected that CO(g) needs to be transported to the secondary process site through traditional transportation methods that rely on burning fossil fuels, which in turn emit CO ₂ (g). In addition, CO(g) is lost in the secondary process of conversion into useful chemicals because the yield of such industrial processes does not reach 100%, resulting in a higher carbon footprint.

Due to the increasing demand for silicon and the unsustainable use of fossil carbon, a cyclic process for producing silicon by carbothermal reduction is needed, in which the carbon is captured and returned to the process.

Therefore, there is a need for an environmentally friendly silicon production process that has a low carbon footprint, reduces consumption of limited natural resources, and has low pollutant emissions, such as low _NOx emissions.

Therefore, a novel silicon production process is needed.

Embodiments described herein provide, in one aspect, a process for producing silicon via carbothermal reduction of silicon dioxide having a low carbon footprint, low NO _x emissions, and reduced resource usage in the form of carbon.

Furthermore, the embodiments described herein provide, in another aspect, a carbon capture method for a process for producing silicon by carbothermal reduction of silicon dioxide, the carbon capture method combining the use of hot plasma and high pressure disproportionation reaction of CO(g).

In addition, the embodiments described herein provide, in another aspect, a process for producing solid carbon by combining hot plasma decomposition-ultra-fast quenching-disproportionation reaction of CO(s), wherein the CO(s) is produced by carbon thermal reduction of silicon dioxide to silicon using a closed arc furnace.

Additionally, embodiments described herein provide, in another aspect, a process for producing a high concentration CO(g) stream by carbothermal reduction of silicon dioxide to silicon.

Additionally, embodiments described herein provide, in another aspect, use of a closed arc furnace for minimizing the presence of excess air and/or oxygen during the carbothermic reduction of silicon dioxide to silicon in order to capture carbon.

Furthermore, the embodiments described herein provide, in another aspect, a process for forming a plasma gas stream from a pure or concentrated CO(g) stream to form solid carbon via thermal decomposition.

In addition, the embodiments described herein provide, in another aspect, an ultra-fast quenching process of CO(g) plasma generated by a plasma torch, wherein the process uses a converging nozzle to convert the thermal energy of the plasma gas into kinetic energy to rapidly reduce its temperature, thereby maximizing the formation and recovery of carbon in solid form.

Additionally, embodiments described herein provide, in another aspect, the use of an inert and/or reducing quench gas to increase carbon recovery in a CO(s) plasma stream by avoiding the reverse reaction of carbon with oxygen after a converging nozzle.

Furthermore, the embodiments described herein provide, in another aspect, a high pressure/moderate temperature process for the disproportionation of CO(g) to solid carbon with or without a catalyst.

Additionally, embodiments described herein provide, in another aspect, a process for recovering carbon in silicon production by carbothermal reduction of silica, wherein the process is performed by pelletizing carbon captured from a CO(g) stream from a closed arc furnace and a silica-containing material.

The above disadvantages can be at least partially overcome by the subject matter of the present invention, which uses a two-step process to capture and reuse carbon in silicon production by reducing silicon dioxide to silicon by conventional carbothermal reduction. In the first step, a method is provided for producing clean CO(g) from a carbothermal reduction furnace. In the second step, a method is provided for recovering carbon from CO(g) and then returning the carbon to the carbothermal reduction step.

Referring to FIG. 1 , a schematic diagram of a process for producing silicon with carbon recovery is shown, wherein silicon dioxide 201 and carbon 202 are continuously fed to a closed electric arc furnace (CEAF) 203. The CEAF 203 may be operated in a vacuum state, in an atmospheric pressure state, or exposed to an exhaust draft to avoid fugitive emissions to the environment. Where very low NO _x and CO ₂ emissions are required, particularly at the beginning of a carbothermal reduction process, a vacuum rated CEAF is preferred in order to allow minimization of residual oxygen within the furnace prior to application of energy via the arc and during operation at a negative operating pressure (i.e., below atmospheric pressure).

CEAF 203 minimizes the formation of CO ₂ (g) by minimizing the gas infiltration into the furnace so that the furnace gas is enriched with CO (g), which means that CO (g) from the carbothermal reduction process in the furnace cannot be oxidized by oxygen in the ambient air.

The reaction environment in CEAF 203 is controlled by a closed structure that minimizes air infiltration, which means that the presence of nitrogen from the air in the furnace environment has been minimized to zero. This inhibits the nitrogen and oxygen from the air from reacting at high temperatures to form _NOx .

The _NOx -free CO(g) stream contains a large amount of particulate matter in the form of silica fume, which is a byproduct of the incomplete reduction reaction of _SiO2 to Si. In order to further process the CO(g), the particulate matter is filtered out in the dust removal step 204.

The now particle-free CO(g) stream enters a decarbonization process where the solid carbon is recovered and can be returned to the silicon plant to re-react with silica. The first step of CO(g) recovery combined with the second step of extracting carbon from the CO(g) allows for on-site recycling of carbon in the silicon production process.

To capture carbon from CO(g), two methods can be used.

The first method occurs in a plasma reactor 205, where CO(g) is fed into a plasma that can conduct an electric current and the plasma can raise the temperature of CO(g) to a point where C and O can coexist in their elemental forms. The overall reaction can be written as follows. CO(g) = C(g) + O(g)

The second process occurs in the disproportionation reactor 206, where CO(g) molecules react at elevated temperature and pressure to extract 1 mole of solid C from 2 moles of CO(g). The overall reaction can be written as follows. CO(g) + CO(g) = C(s) + CO ₂ (g)

Each method can be used alone (on its own) or in combination to maximize carbon recovery.

The carbon obtained from both processes (205 and 206) goes through a carbon collection and pelletizing step 207 before being mixed with silicon dioxide 201, which can then be fed back to the CEAF 203 to produce additional silicon.

Remaining waste gas 208 leaves the process through an exhaust pipe.

Referring to Figure 2, an exemplary schematic diagram of a novel process for producing silicon using carbon capture is shown. Silicon dioxide, for example in the form of quartz, and carbon, for example in the form of coal, coal tar, charcoal and sawdust and/or mixtures thereof, are transported via a conveyor belt 1 to an open feed hopper 2 for temporary bulk storage. The mixture of carbon and silicon dioxide can then be fed to a closed feed hopper 4, which is isolated from the open environment via an airtight valve 3, such as a gate valve.

The closed feed hopper 4 is connected to a vacuum pump 5 and isolated from the CEAF 9. The vacuum pump 5 is used to remove residual air in the closed feed hopper 4.

In order to minimize the formation of NO _x at the beginning of the process, once the closed feed hopper 4 and the CEAF 9 are degassed using the vacuum pump 5, argon is injected from the source 7 by opening the valve 8, and once the pressure inside the CEAF 9 reaches atmospheric pressure, the valve 8 is closed.

This degassing process needs to be repeated to ensure that residual air is removed from the system.

To load the CEAF 9, the airtight valve 3 is opened to unload the mixture of carbon and silica into the closed feed hopper 4 and then closed. The vacuum pump 5 is then operated to remove the air trapped in the closed feed hopper 4 by a degassing process by backfilling the closed feed hopper 4 with argon from a source 7 via the opening and closing of the valve 8.

Next, valve 6 is opened to unload the raw material mixture such as quartz and carbon into CEAF 9.

The carbothermal reduction of silicon dioxide, for example in the form of quartz, with a carbon source occurs in the CEAF 9 in the absence of air, which results in low NO _x emissions from the process and the formation of a hot CO(g)-rich stream.

The hot CO(g)-rich stream leaves the CEAF 9 carrying fine silica particles from the incomplete reduction reaction of silica with carbon. Most of the fine particles in the hot gas stream are removed by a hot cyclone 10, such as a refractory lined cyclone allowing the processing of hot gases.

The gas leaving the hot cyclone 10 is directed to a heat exchanger 11, where the gas temperature is reduced, for example, to below 150° C. The gas can then be filtered through a high-efficiency filter system 12, such as a bag filter that can be combined with a HEPA filter, to remove residual particles not captured by the hot cyclone 10 from the cooled gas to maximize particle removal efficiency.

The now particulate-free CO-rich gas stream is driven by an exhaust (ID) fan 13. The ID fan 13 ensures that the process pressure is kept slightly below atmospheric pressure to avoid any escaping emission of CO(g) into the surrounding environment and to ensure the safety of the operating personnel.

The gas compressor 14 increases the pressure of the CO(g) rich stream to a moderately high level sufficiently high, such as up to 10 atm, to enable the plasma torch 16 to operate properly.

During the operation of the plasma torch 16, the buffer tank 15 is required to maintain the pressure of CO(g). The pressurized CO(g) enters the plasma torch 16 where it decomposes into C(g) and O(g). The plasma containing C(g) enters the quench system 17 which is highly effective in preventing the back reaction of C(g) and O(g) to CO(g), thereby minimizing the reduction of CO(g). Another function of this quench system 17 is to reduce the temperature of the plasma gas to the point where the carbon gas C(g) condenses into its solid form C(s).

Once formed, C(s) remains stable in the quenched gas stream and enters the cold cyclone 18, through which most of the C(s) is separated from the gas stream by the cyclonic effect of the gas stream.

The gas leaving the cold cyclone 18 should be further cooled by the heat exchanger 19 and to be cold enough, e.g., below 150°C, to be filtered through commercially available filter materials that can be used in a filter system 20 such as a bag filter, which can be combined with a HEPA filter to maximize the efficiency of removing remaining C(s) from the gas stream.

The residual CO(g) stream purged from C(s) is driven by a blower 21 into a gas compressor 22 to increase its pressure to a selected value, such as the selected value from Table 1 (below), for further processing into a CO(g) disproportionation reactor 23, which can be operated at up to 100 atm and at a high temperature, such as 800°C, to convert a portion of the CO(g) to C(s). For example, if the conversion of CO(g) needs to be 75% so that the recovery of C(s) is 30%, then the reactor 23 needs to be operated at 25 atm and a temperature of 800°C. In order to enhance the reaction process and reduce the energy requirements of such a disproportionation process, a catalyst material such as iron can be used.

The hot gases leaving the disproportionator 23 are cooled by a heat exchanger 24 before being discharged to the atmosphere. To avoid harmful CO(g) emissions to the atmosphere, the remaining CO(g) can be oxidized to _CO2 (g) by a thermal oxidizer. At this point, the concentrated _CO2 (g) stream can be further concentrated and reused off-site for useful industrial or commercial purposes.

The C(s) collected from the disproportionation reactor 23 is then transferred to the briquetting/granulation unit 25 together with the C(s) collected from the cold cyclone 18 and the filtration system 20 to be mixed with quartz so that the carbon from a portion of the primary carbon thermal reduction of silica to silicon is returned to the process, thereby achieving circular carbon utilization in the process.

Hot plasma can be formed by several methods known to experts in the field, including but not limited to AC arc and DC arc, radio frequency inductively coupled source (RF-IC), and microwave. Of these methods, arc and RF-IC are of interest because they have been used in industrial hot plasma torches, they can handle high flow rates of gas, and more importantly, they can produce the very high temperature plasma required by the plasma torch 16 of this process. Ideally, the plasma torch 16 can elevate the temperature of the bulk CO(g) to 8000°C to maximize the dissociation rate of CO(g) to C(g), and the temperature is not less than 5000°C to maintain C(g). Combinations of two or more hot plasma generation methods are also possible. For example, the arc-generated plasma can be combined with an RF-IC source to increase the gas volume and/or CO(g) retention in the hot plasma region.

3, there is shown an exemplary vertical cross-sectional schematic diagram of a plasma torch 16 and a quenching system 17 for first-stage carbon capture. The plasma torch 101 is powered by a power source 102, and a CO(g) stream 103 enters the plasma torch 101 at one or more locations, such as through a gas distributor 104 in the case of a direct current (DC) arc torch, where the arc voltage can be increased by a higher flow of plasma gas to enhance the power level.

A shield gas, such as argon, is injected through a gas distributor 105. The ratio of shield gas to plasma forming gas can be as low as 10% or even lower. For arc plasma torches that employ a tubular electrode configuration or a coaxial electrode configuration, for example, made of copper, no shield gas is required. This is advantageous because the gas flow remains rich in CO(g). The front end of the plasma torch 106 is a water-cooled flange that can be directly connected to a quench module 107 having a convergent-divergent shape that allows the hot plasma gas to reach velocities exceeding Mach 1 in the divergent region 110. The plasma gas entering the quench module 107 is compressed into the gradient zone 108, causing the gas velocity to reach Mach 1 in the pass zone 109. Once it enters the gradient zone 110, the plasma gas expands when it reaches a velocity exceeding Mach 1, converting its thermal energy into kinetic energy. A cooling rate greater than 5×10 ⁷ °C/s [Reference 9] ensures that C(g) in the plasma is converted to C(s). Since the residence time of the plasma CO(g) in the quench is limited by its high velocity, partial conversion of CO(g) to C(s) is expected. To further enhance the quenching process, a quenching gas such as an inert gas (e.g., Ar, He, _N2 ), or a reducing gas such as CO(g), _H2 (g) or _CH4 (g), or a combination of gases, should be injected through the gas distributor 111. The cold quenching gas contacts the gas exhausted from the gradual expansion area 110, so that the total temperature of the gas is reduced, further avoiding the reverse reaction of C(s) and oxygen. The remaining CO(g) that is not converted to C(s) is further subjected to the disproportionation reaction as described above to maximize the total carbon capture rate.

Two examples are used to further illustrate how solid carbon is produced from gaseous CO in the plasma reactor 205 and the disproportionation reactor 23.

Method 1 : CO plasma for carbon capture . The plasma composition of carbon monoxide can be predicted using the Gibbs free energy minimization method, assuming that the hot plasma reaches local thermodynamic equilibrium. The CO thermodynamic equilibrium composition can then be calculated over a wide range of temperatures. The CO plasma composition was calculated using HSC commercial software v. 8 and the results are shown in Figures 4a to 6. In order to predict the plasma state temperature of CO, it is necessary to define a plasma. For a gas to be considered a plasma, there should be enough ionized species present. In general, if a gas has at least 1% ionization (ionization degree), it can be considered a plasma (taking into account the high conductivity). The ionization degree can be written as follows. Here, n _i and n _n are the number densities of ions and neutral species, respectively.

In order to calculate the ionization degree, the particle number density of CO gas at different temperatures must be calculated. Figure 5 shows the particle number density of CO plasma calculated at atmospheric pressure. With the particle number density, the ionization degree of plasma at different temperatures can be estimated according to the above equation. The calculation result is shown in Figure 6.

As shown in Figure 6, the ionization degree is very low (<0.05%) below 6000°C. At 7800°C, the ionization degree of CO plasma reaches 1%. Therefore, it can be said that CO plasma is formed at a temperature above 7800°C.

However, CO begins to dissociate into monatomic species (i.e., C and O) at much lower temperatures, around 3000°C, at which point their number density becomes greater than 1016 m ^-3 . As shown in Figure 6, at temperatures above 8000°C, CO atomizes into free O(g) and C(g) atoms. Above 10,000°C, the composition of the plasma is primarily atomic oxygen and carbon. If very high quenching rates can be achieved, carbon can be directly converted to the solid phase and trapped. The quenching rate at lower temperatures should be higher than the rate for the CO formation kinetics.

As shown in Figures 4a and 4b, the formation of carbon is thermodynamically favorable at lower temperatures (< 2100°C), and carbon may be formed through the reaction: CO + CO = C + CO ₂ . This means that in a reaction system consisting of CO, the formation of carbon and CO ₂ (g) is thermodynamically favorable, and CO(g) should decay into these species.

Method 2 : CO dissociation for carbon capture . The disproportionation reaction of 2 mol of CO to 1 mol of carbon and 1 mol of CO ₂ shows that the dissociation of CO to carbon is feasible. CO(g) + CO(g) = CO ₂ (g) + C(s)

If the reaction is feasible, it is expected that the reaction should be carried out at high pressure due to the reduction in the volume of the gaseous reactants (2 mol of CO(g) converted to 1 mol of CO ₂ (g)), and low temperature should be favorable in order to slow down the reverse reaction of C and CO _2. In fact, as shown in Figure 7, the results of the thermodynamic calculations are consistent with this expectation.

According to the results shown in Figure 7, from a thermodynamic point of view, CO tends to be converted to _CO2 and C at low temperatures. However, this reaction is expected to be very slow. In order to increase the reaction rate, if the pressure is increased, the conversion of CO to _CO2 can be carried out at a higher temperature. For example, at a pressure of more than 100 atm, if the reaction temperature is increased from 400°C to 800°C, almost the same carbon conversion rate can be obtained.

Table 1 summarizes the thermodynamic equilibrium calculation results of the CO system at 800°C and different operating pressures. In short, the higher the reaction pressure, the higher the carbon conversion rate. Table 1 - Thermodynamic equilibrium composition of the CO system at 800°C* Substance (mole) Pressure (atm) 1 5 10 25 50 100 CO 0.7955 0.5063 0.3834 0.2539 0.1825 0.1302 CO ₂ 0.1022 0.2468 0.3083 0.3730 0.4078 0.4349 C 0.1022 0.2468 0.3083 0.3730 0.4087 0.4349 * FACTSAGE WEB CODE v. 7.1

Therefore, it is theoretically demonstrated that solid carbon can be extracted from CO(g) using a non-transferring DC plasma torch and then the remaining CO(g) is subjected to extreme quenching and high pressure-moderate temperature catalytic treatment to capture more carbon.

Although the above description has provided examples of embodiments, it should be understood that certain features and/or functions of the embodiments may be modified without departing from the spirit and principles of operation of the embodiments. Therefore, what has been described above is intended to illustrate the embodiments rather than to limit them, and a person of ordinary skill in the art will understand that other changes and modifications may be made without departing from the scope of the embodiments as defined by the attached patent claims.

References [1] Nils Eivind Kamfjord, Mass and Energy Balances of the Silicon Process (2012), PhD Thesis, Norwegian University of Science and Technology. [2] Edin Henrik Myrhaug, Halvard Tveit, Nils Eivind Kamfjord, GeirJohan Andersen and Åslaug Grøvlen, NO _x Emissions from Silicon Production (2012), Silicon for the Chemical and Solar Industry XI Bergen -Ulvik, Norway. [3] T. Lindstad1,2, SEOlsen1,2, G. Tranell2, T. Færden3 and J. Lubetsky, Greenhouse Gas Emissions from Ferroalloy Production (2007), INFACON XI, New Delhi, India, 18-21. [4] Gudrun Saevarsdottir, Thordur Magnusson, and Halvor Kvande, Reducing the Carbon Footprint: Primary Production of Aluminum and Silicon with Changing Energy Systems, Journal of Sustainable Metallurgy (2021), 7:848–857. [5] KZ House, AC Baclig, M. Ranjan, EA van Nierop, J. Wilcox, and HJ Herzog, Economic and energetic analysis of capturing CO ₂ from ambient air (2011), PNAS, 108 (51) 20428-20433. [6] IECM Technical Documentation: Amine-based Post-Combustion CO ₂ Capture (2018). [7] Elkem, The road to climate neutral metal production [https://www.elkem.com/innovation/long-term-rd/the-road-to-climate-neutral-metal-production/]. [8] I. Solheim, V. Andersen, and R. Jensen, Recirculating off-gas contributes to carbon capture (2021) [https://www.sintef.no/en/latest-news/2021/recirculating-off- gas-contributes-to-carbon-capture/]. [9] Benny T. Kuan, and Peter J. Witt, Modeling supersonic quenching of magnesium vapor in a Laval nozzle (2013) Chemical Engineering Science Volume 87, Pages 23-39.

This application claims the benefit of priority to currently pending U.S. Provisional Application No. 63/408,442, filed on September 20, 2022, which is incorporated herein by reference.

1: Conveyor belt 2: Open feed hopper 3: Airtight valve 4: Closed feed hopper 5: Vacuum pump 6: Valve 7: Source 8: Valve 9: Closed electric arc furnace (CEAF) 10: Hot cyclone 11: Heat exchanger 12: High efficiency filter system 13: Exhaust fan (ID fan) 14: Gas compressor 15: Buffer tank 16: Plasma torch 17: Quenching system 18: Cold cyclone 19: Heat exchanger 20: Filter system 21: Blower 22: Gas compressor 23: Disproportionation reactor 24: Heat exchanger 25: Briquetting/granulation unit 101: Plasma torch 102: Power supply 103: CO(g) feed stream 104: Gas distributor 105: Gas distributor 106: Plasma torch 107: Quenching module 108: Converging zone 109: Passing zone 110: Expanding zone 111: Gas distributor 201: Silica 202: Carbon 203: Closed electric arc furnace (CEAF) 204: Dust removal step 205: Plasma reactor 206: Disproportionation reactor 207: Carbon collection and granulation step 208: Waste gas

In order to better understand the embodiments described herein and to more clearly show how the embodiments are implemented, reference will now be made to the accompanying drawings by way of example only, which show at least one exemplary embodiment, and in which: FIG. 1 is an exemplary schematic diagram of a process for producing silicon using carbon recovery according to an exemplary embodiment; FIG. 2 is an exemplary schematic diagram of a process for producing silicon using carbon capture according to an exemplary embodiment; FIG. 3 is an exemplary vertical cross-sectional schematic diagram of a plasma torch and quenching module for first-stage carbon capture according to an exemplary embodiment; Figures 4a and 4b are two graphs showing the thermodynamic equilibrium composition of CO(g) over a wide range of temperatures (100 to 10000°C), wherein the Y-axis of Figure 4a is kmol and the unit of the X-axis is temperature (°C), and the unit of the Y-axis of Figure 4b is Log (kmol) and the X-axis is temperature (°C); Figure 5 is a graph showing the composition of CO plasma at 1 atm; Figure 6 is a graph showing the ionization of CO plasma at 1 atm; Figures 7a, 7b, 7c and 7d are four graphs showing the thermodynamic equilibrium of the C-O reaction system at 400, 600, 800 and 1000°C and a pressure range of 1 to 200 bar, respectively.

201:Silicon dioxide

202: Carbon

203: Closed Arc Furnace (CEAF)

204: Dust removal step

205: Plasma reactor

206: Disproportionation Reactor

207: Carbon collection and granulation steps

208: Waste gas

Claims (10)

A process for producing silicon by carbothermal reduction of silicon dioxide having a low carbon footprint, low NO _x emissions, and reduced resource usage in the form of carbon. A carbon capture method is provided for use in a process for producing silicon by carbothermal reduction of silicon dioxide, wherein the carbon capture method combines the use of hot plasma and high pressure disproportionation reactions of CO(g). A process for producing solid carbon by combining hot plasma decomposition-ultra-rapid quenching-disproportionation reaction of CO(s) produced by carbothermic reduction of silicon dioxide to silicon using a closed arc furnace. A process for producing a high concentration CO(g) stream by carbothermal reduction of silicon dioxide to silicon. A closed arc furnace is used to minimize the presence of excess air and/or oxygen during the carbothermal reduction of silicon dioxide to silicon in order to capture carbon. A process for forming a plasma gas stream from a pure or concentrated CO(g) stream to form solid carbon via thermal decomposition. An ultra-rapid quenching process of CO(g) plasma generated by a plasma torch, wherein the process uses a converging nozzle to convert the thermal energy of the plasma gas into kinetic energy to rapidly reduce its temperature, thereby maximizing the formation and recovery of carbon in solid form. An inert and/or reducing quench gas is used to avoid the back reaction of carbon and oxygen after a converging nozzle to increase the carbon recovery rate in a CO(s) plasma stream. A high pressure/moderate temperature process for the disproportionation of CO(g) to solid carbon with or without a catalyst. A process for recovering carbon in silicon production by carbothermal reduction of silicon dioxide, wherein the process is performed by pelletizing carbon (the carbon is captured from a CO(g) stream from a closed arc furnace) and a material containing silicon dioxide.