WO2025219160A1 - Method for treating silicon-containing solid bodies - Google Patents

Abstract

The invention relates to a method for treating silicon-containing solid bodies in a reactor (10). The method is characterized by the following steps: a) setting a treatment temperature in the reactor (10), said treatment temperature equaling or being above the melting point of silicon under the conditions in the reactor (10); b) introducing silicon-containing solid bodies, which form a first region (18), into the reactor (10), c) a second region (20) comprising a pretreatment slag being provided in such a way that the first region (18) is located above the second region (20); d) providing a third region (22), said third region (22) comprising a metal alloy; and e) providing a fourth region (24), in which an electrolyte is provided and in which silicon is produced electrolytically; f) said liquid silicon forming a fifth region (26) adjacent to the fourth region (24), and said first region (18), second region (20), third region (22), fourth region (24) and fifth region (26) following one another, in the stated order, in a reaction direction through the reactor (10).

Description

Process for treating silicon-containing solid bodies

The present invention relates to a method for treating silicon-containing solid bodies. In particular, the present invention relates to a method by which high-purity silicon can be produced from silicon-containing components with high efficiency.

Energy generation based on solar cells is becoming increasingly important. However, there is currently no sustainable solution for recycling silicon waste aimed at directly producing high-purity silicon (>99.9999%), which is primarily used for solar cell manufacturing. Conventional production methods require between 3.5 and 9.5 kg of silicon to produce 1 kg of high-purity wafers, with an average consumption of between 0.4 and 2.1 MWh/kg. Given the exponential increase in the use of solar cells in the fight against climate change, the world will face a huge influx of silicon waste from end-of-life solar cells in the coming years, which must be properly treated. Solar cell production primarily requires silicon with a purity of at least 99.9999%. Several elements such as Fe, Al, Cu, B, and P are also required to produce a high-quality panel. Conventional production of high-purity Si follows two main routes. The first is the chemical route, involving the preparation and purification of silanes, followed by reduction to silicon metal by chemical vapor deposition, e.g., the Siemens process. The second route is the metallurgical route, in which primary metallurgical Si is purified through a combination of slag treatment, vacuum refining, and fractional solidification.

CA2645161A1 describes a process consisting of two interconnected molten salt electrolysis cells. In the first cell, quartz (silicon oxide) is reduced to silicon metal, forming a metal alloy of silicon and another precious metal. The silicon from this alloy is then further electrolytically refined in an adjacent cell, forming a high-purity silicon metal layer in the cathode region.

According to GB 2477782 A, silicon kerf, a high-purity metallic powder produced during wafer sawing, is recycled through a slag treatment in which the fine kerf powder is melted together with a slag system consisting of SiOi-CaO-CaFi. This slag facilitates the coagulation of the molten silicon droplets, allowing the molten kerf to be completely coagulated and more efficiently recovered by purging with inert gas in the melt. Furthermore, this process can reduce harmful elements such as B and P to acceptable levels for solar material through slag treatment.

In CA 2634592 Al, metallurgical silicon metal is produced by directed fractional

Crystallization with a curved growth front, which is designed to be as large as possible crystallization interface. In this process, a portion of the treated silicon metal is recovered as solar-grade silicon.

FR 2559473 A1 describes a process for producing purified silicon in which a material forming a silicon source is subjected to an electrolytic reaction. The process uses a three-layer liquid system consisting of an anode layer, a cathode layer, and an intermediate electrolyte layer arranged one above the other. The anode layer consists of a silicon-containing liquid alloy, the cathode layer of liquid silicon, and the intermediate layer of a silicon dioxide-containing liquid electrolyte.

US 8303796 B2 describes a process for the cost-effective production of high-purity silicon, which can be suitable as a starting material for solar cells. Specifically, in the process, silicon dioxide is subjected to molten salt electrolysis in an electrolysis vessel, and steps are carried out in the following order: step (1) in which the silicon content in a silicon-containing alloy, which is in a liquid phase at the electrolysis temperature, is increased by using the alloy as a cathode and conducting electrolysis; step (2) in which the silicon-containing alloy serving as the cathode is removed from the electrolysis vessel before it reaches the concentration at which silicon begins to precipitate at the electrolysis temperature; step (3) in which silicon is solidified by cooling the removed silicon-containing alloy within the temperature range higher than the eutectic point but lower than the electrolysis temperature; and step (4) in which the solidified silicon is collected.

EP 2274 767 A4 describes methods, systems, and devices for recovering high-purity silicon, silicon carbide, and PEG from a slurry produced during a wafer dicing process. A silicon-containing material can be used to produce a silicon-rich composition. Silicon carbide and PEG recovered from the silicon-containing material can be used to produce a wafer sawing fluid. The silicon-rich composition can be reacted with iodine-containing compounds, which can be purified and/or used to form deposited silicon of high purity. The produced silicon can be used in the photovoltaic or semiconductor industries.

Several other research studies and inventions have been conducted to upgrade metallurgical silicon to solar-grade silicon. However, no known process addresses the purification of highly contaminated silicon residues, such as recycled silicon-containing material from decommissioned solar panels, into solar-grade silicon metal. Therefore, the above-mentioned applications are only suitable for a narrow range of feedstocks, while offering little or no flexibility to treat a much broader range of metallic silicon-containing sources, even those of lower purity.

The known state of the art therefore still offers potential for improvement.

It is therefore the object of the present invention to at least partially improve at least one disadvantage of the prior art. In particular, the object of the present invention is to provide a solution by which even highly contaminated silicon waste, such as silicon scrap, can be treated simply and efficiently to produce ultrapure silicon or its derivatives.

The object is achieved according to the invention by a method having the features of claim 1. Preferred embodiments of the invention are disclosed in the subclaims, in the description and in the figures, wherein further features described or shown in the subclaims or in the description or the figures can be individually or in any combination may constitute an object of the invention unless the context clearly indicates the opposite.

The present invention relates to a method for treating silicon-containing solid bodies in a reactor, the method comprising the following steps: a) setting a treatment temperature in the reactor, wherein the treatment temperature is equal to or above the melting point of silicon under the conditions in the reactor; b) introducing silicon-containing solid bodies forming a first region into the reactor, wherein c) a second region comprising a pretreatment slag is provided such that the first region is located above the second region; d) providing a third region, wherein the third region comprises a metal alloy; e) providing a fourth region, wherein the fourth region comprises an electrolyte; wherein f) the liquid silicon forms a fifth region which is adjacent to the fourth region, in particular above the fourth region, wherein the first region, the second region, the third region, the fourth region, and the fifth region follow one another in this order in a reaction direction through the reactor.

Such a process can offer significant advantages over state-of-the-art solutions. In particular, the process can provide ultrapure silicon with a purity of at least 99.9999 wt.% and can therefore be considered a recycling process. The process can thus be viewed as a process for refining silicon-containing solids, in particular metallic silicon scrap. However, it is also It is possible that the process includes further process steps based on the ultrapure silicon produced.

Fundamentally, the method according to the invention is a method for treating silicon-containing solid bodies in a reactor. The described method comprises the following steps, whereby the method is not intended to be limited to the described steps, but rather it is possible within the meaning of the present invention to add further steps. Furthermore, in particular five different regions are provided or created which differ in their composition, with the first region, the second region, the third region, the fourth region and the fifth region following one another in this order in a reaction direction through the reactor. The reaction direction is to be understood in particular as a direction which leads through the layers from the starting product, i.e. the silicon-containing solid bodies, to the fifth region, i.e. the purified, in particular liquid, silicon.

The present method initially comprises step a), namely setting a treatment temperature in the reactor, wherein the treatment temperature is equal to or above the melting point of silicon under the conditions in the reactor. Under standard conditions, i.e., in particular at an ambient pressure of 1 bar, the reactor interior is thus heated to a temperature of > 1414 °C, preferably > 1500 °C, approximately in a range from > 1414 °C to < 2000 °C, for example, in a range from > 1500 °C to < 1700 °C.

The present process further comprises step b), namely the introduction of silicon-containing solid bodies forming a first region into the reactor. For the purposes of the present invention, silicon-containing solid bodies are understood to mean any solids that comprise a certain proportion of silicon. Silicon-containing solids are, accordingly, in particular silicon-containing solids, preferably metallic silicon-containing solids, such as metallic silicon scrap. As described in more detail below, an advantage of the process described here is that the selection of silicon-containing solids is not limited to pretreated or pre-purified solids or to bodies with a high degree of silicon purity, but rather that the process allows for considerable flexibility in the selection of starting materials. The silicon-containing solids are introduced into the reactor and thereby form a mass, which is referred to as the first zone.

In the context of the invention, a region is to be understood as a volume in which a specific substance, in the first region the silicon-containing solid bodies as starting material, are present.

The first zone is located above a second zone, which contains a pretreatment slag. Thus, the silicon-containing solids are applied to the second zone in the reactor, thus forming the first zone.

Accordingly, the process according to process step b) comprises providing a second region comprising a pretreatment slag, wherein the first region is located above the second region. The pretreatment slag can also be referred to as LES (light-engineered slag) and serves to pretreat or precondition the silicon-containing solid bodies.

Due to the fact that the prescribed treatment temperature has been set inside the reactor, the silicon present in the silicon-containing solid bodies melts and flows into the second area, coming into contact with the pretreatment slag, which leads to pretreatment and, in particular, coagulation of the silicon. More precisely, when the silicon-containing solid bodies in solid state are mixed with the second When the silicon-containing solids come into contact with the first zone, they initially tend to float on the pretreatment slag. However, when the silicon-containing solids reach the melting temperature of silicon, the silicon melts and passes through the second zone, where it simply flows downwards. Upon passing through the second zone, the first silicon impurities are left behind. These can usually be thermally decomposed, are therefore particularly volatile components, and/or have a high affinity for oxygen. Non-metallic impurities can be absorbed into the pretreatment slag as gangue.

For this purpose, the pretreatment slag is preferably designed as follows. First, slag is generally understood to be a substance comprising a mixture of oxides and other components, in particular cations, such as aluminum (Al), sulfur (S), phosphorus (P), fluorine (F), boron (B), or the like. Regarding the arrangement of the various regions, in order to facilitate simple implementation, the preconditioning slag should have a density that is greater than that of the applied silicon-containing solid bodies and, above all, lower than that of liquid or molten silicon. In particular, the density can be at least 5% lower than that of molten silicon, where 5% refers to the density of liquid silicon.

The composition of the pretreatment slag also promotes the absorption of impurities and oxides from the silicon-containing solids and enables the conditioning of the metal before mixing with the underlying alloy of the third region or the penetration of the metal into the alloy.

The pretreatment slag may in particular comprise a system based on the components silicon oxide (SiOi), calcium oxide (CaO), and aluminum oxide (Al2O3), optionally with a lithium salt, such as in particular lithium oxide (LiO), being used for It can be added to reduce the density. Alternatively or additionally, calcium fluoride (CaFi) may be added, depending on the starting materials, i.e., the silicon-containing solids. For example, the pretreatment slag can contain the following components in the following proportions, expressed as wt.% based on the pretreatment slag used in its raw state:

Silicon oxide (SiCh) in a range of > 10 wt% to < 80 wt%;

Calcium oxide (CaO) in a range of > 10 wt% to < 60 wt%; aluminum oxide (Al2O3) in a range of > 5 wt% to < 30 wt%; lithium oxide (LiiO) in a range of > 5 wt% to < 20 wt%; and calcium fluoride (CaF) in a range of > 0 wt% to < 15 wt%.

In a preferred embodiment, the pretreatment slag may comprise the following components in the following proportions, indicated in wt.% based on the pretreatment slag used in the raw state, i.e. without taking into account any reactions during the process, although the values may equally refer to the pretreatment slag in the process:

Silicon oxide (SiOi) in a range of > 40 wt% to < 54 wt%;

Calcium oxide (CaO) in a range of > 15 wt% to < 31 wt%; aluminum oxide (Al2O3) in a range of > 5 wt% to < 21 wt%; lithium oxide (Li2O) in a range of > 5 wt% to < 16 wt%; and calcium fluoride (CaF2) in a range of > 0 wt% to < 7 wt%.

The exact amount of the individual components can be selected depending on the density of the silicon-containing solids and their composition. For example, many silicon-containing solids contain a certain amount of oxidized silicon, which manifests itself in the pretreatment slag by increasing the silicon content and also by increasing the viscosity. On the other hand, parts of solar cells, as exemplary components of silicon-containing solids, usually have a A conductive layer of aluminum also increases the AlCh content in the pretreatment slag during processing. Furthermore, without the use of lithium, maintaining the density gap over time when using only SiCl, CaO, and Al2O3 is often difficult. In the case of recycled solar cells, these can be preferentially added due to the presence of aluminum in the cell, along with a corrective amount of SiO2, CaO, and Li2O2 to prevent the loss of the density gap or density difference after the pretreatment slag has collected the impurities associated with the cell.

Furthermore, other elements, such as MgO, NaiO, can also be used in the pretreatment slag.

It should also be mentioned that the addition of lithium or fluorine in the pretreatment slag increases the volatilization of these elements in the flue gas, which can be continuously added together with the addition of silicon-containing solids.

The pretreatment slag can be in the form of a melt of the components. The second region can accordingly comprise the pretreatment slag in liquid or molten form.

Furthermore, the pretreatment slag, with the function of preconditioning the silicon-containing solids, in particular silicon scrap, and therefore adjusting their chemical composition, has no impact on the continuity of the process, but rather serves to optimize the required slag mixture or composition of the slag without compromising the cleaning in the third area. The LES serves to precondition the slag so that regulating its chemical composition does not affect the continuity of the process, but rather serves to optimize the required slag mixture without compromising purification in the second chamber. The slag (LES) specifically serves to precondition the introduced silicon-containing solid in such a way that elements with a higher oxygen affinity than silicon remain in the slag and do not influence the chemical composition of the electrolyte (the HES) within the second chamber, thus negatively impacting the electrolytic purification in the second chamber. The slag (LES), which comes into contact with the scrap, initially absorbs the impurities with high oxygen affinity, including any oxides already present. The liquid silicon from the incoming scrap passes through the LES and is assimilated by a Cu-Si alloy melt, whereby metallic impurities with lower oxygen affinity than silicon (Si), such as iron (Fe), copper (Cu), lead (Pb), silver (Ag), tin (Sn), etc., are also absorbed by the Cu-Si alloy or dissolved in it and remain in the alloy.

The method further comprises, according to step d), the provision of a third region, wherein the third region comprises a metal alloy, such as a silicon-copper alloy. This enables the pretreated and, in particular, coagulated silicon to mix with the metal alloy after it has passed through the pretreatment slag. In this process, the impurities with low oxygen affinity present in the starting substance, i.e., in the silicon-containing solids, such as iron (Fe), nickel (Ni), lead (Pb), tin (Sn), and silver (Ag), are also absorbed by the Cu-Si alloy, remaining in the alloy and thus being separated from the silicon. The metal alloy can be in liquid or molten form. The third area can accordingly include the metal alloy in liquid or molten form.

The presence of copper as a component of a metal alloy in the third zone, which can form a base alloy as described below, is related not only to its low cost but also to its high density and copper's ability to dissolve other metals, thus reducing the activity of lower-concentration elements that could contaminate the electrolyte in the second chamber. In principle, however, other alloys besides copper alloys are also possible in the third zone, such as alloys based on nickel, iron, or silver.

According to the further method step e), a fourth region is further provided, wherein the fourth region comprises an electrolyte, in particular one that is liquid at the processing temperature, and wherein silicon is electrolytically produced in the fourth region. In this region, in particular, pure silicon is continuously produced by electrolysis using the electrolyte. For the electrolysis, an electrical voltage is preferably applied in this region. The electrolyte preferably contains a high concentration of SiO2, and by applying a potential between corresponding electrodes, Si ions are selectively transferred from the lower melt to the upper melt.

The electrolyte preferably has a higher density than liquid silicon and furthermore advantageously a relatively high electrical conductivity. If the electrolyte has a higher density than liquid silicon, the latter floats on top. Thus, step f) is realized in a particularly advantageous manner in that the liquid silicon forms a fifth region, which is adjacent to, in particular above, the fourth region. Once a certain layer thickness of purified silicon has been reached, the silicon can be continuously removed from the reactor or further processed.

The electrolyte can preferably be a slag, also known as HES (heavy-engineered slag). It preferably exhibits specific thermophysical properties with respect to viscosity and electrical conductivity.

The electrolyte can be in the form of a liquid or molten slag. The fourth region can contain the electrolyte in liquid or molten form.

For electrolytes designed as HES, for example, the melting point, density, and conductivity are particularly important properties that should be considered. It is recommended to use silicon dioxide concentrations of more than 20 wt.% to achieve a good ion flow through the electrolyte. Regarding density, CaO and BaO are elements that can be advantageously used to increase the density without compromising the purity of the silicon during electrolysis. The use of fluorides, such as CaFi, increases electrical conductivity and thus improves process efficiency. However, their use is limited by the corrosion resistance of the furnace refractory materials, which could compromise the quality of the purified metal. Good purification results have been achieved, for example, with SiOi contents between 30-40 wt.%, BaO between 20-60 wt.%, preferably 20-40 wt.%, CaO between 20-40 wt.% or 0-10 wt.%, and CaFi <40 wt.%. The compositions 40 wt.% SiO2-30 wt.% CaO-30 wt.% BaO and 35 wt.% SiO2-30 wt.% CaO-30 wt.% BaO-5 wt.% CaF2 showed good metal quality and stability results, for example, when using AlOa-TiCl. Compositions 40 wt.% SiO2-5 wt.% CaO-55 wt.% BaO and 40 wt.% SiO2-55 wt.% BaO-5 wt.% CaF2 also showed good metal quality and high process stability. Electrolytes for the purification of silicon with melting points below 1400°C and densities above those of liquid silicon (2.61 g/cm ³ at 1500°C) can preferably be achieved using raw materials in the following composition ranges, where the components are given in wt.% based on the electrolyte used in the raw state, i.e. without taking into account any reactions during the process, but can equally refer to the electrolyte in the process:

Silicon oxide (SiCE) in a range of > 20 wt% to < 60 wt%;

Calcium oxide (CaO) in a range of > 0 wt% to < 50 wt%;

Barium oxide (BaO) in a range of > 0 wt% to < 60 wt%;

Aluminium oxide (AI2O3) in a range of > 0 wt% to < 20 wt%;

Magnesium oxide (MgO) in a range of > 0 wt% to < 10 wt%; and calcium fluoride (CaF ) in a range of > 0 wt% to < 60 wt%.

Accordingly, it may generally be advantageous for the electrolyte to contain silicon oxide and for further components to be added which are selected from the group consisting of calcium oxide, barium oxide, aluminum oxide, magnesium oxide and calcium fluoride, in particular in the proportions mentioned above.

In particular, electrolytes with melting points below 1400°C and densities between 2.8 and 3.2 g/ ^cm³ can be used. This is advantageously possible with the compositions described above. Furthermore, they have an electrical resistance, particularly specific electrical resistance, between 0.8 and 3 ohm-cm, which makes these compositions particularly suitable as electrolytes.

The process described above allows a very effective treatment of silicon-containing solid bodies in a simple manner, in which even relatively unclean or impure starting substances with regard to the silicon content can be used. Specifically, the pretreatment slag prevents the electrolyte from rapidly becoming excessively contaminated by impurities present in the scrap used as starting materials. This, in particular, enables continuous and robust purification or refining of silicon, regardless of the source of the silicon scrap used.

If the density of the pretreatment slag is lower than that of liquid silicon but slightly higher than that of solid silicon, five layers, or regions with different features and properties, exist in the process. This makes this invention a significant improvement over existing three-layer electrolyzer technology and suitable for the purification of silicon waste, for example, for wafer production (waste-to-wafer concept) or for the production of solar cells, technically known as photovoltaic cells. Therefore, the arrangement produced by this process can also be referred to as a five-layer electrolyzer.

By combining the properties of the respective areas, it was possible to easily purify silicon from a wide variety of starting materials, especially metallic silicon-containing starting materials, such as silicon waste, especially metallic silicon scrap, such as solar cells, cast silicon fractions (SSF, silicon small fraction), kerf, and metallurgical silicon, and to produce high-purity silicon. The process proved to be robust and long-term stable, thus achieving a continuous and flexible purification process. This enables the production of high-purity silicon for the production of semiconductors, especially wafers for solar panels.

It should also be noted that, in the context of the circular economy, for example, the third area and the pre-treatment slag contain impurities that may also be of great economic interest, which further highlights the economic advantages of the process. In principle, it can be advantageous to carry out the process under an inert atmosphere of protective gas. This can prevent unwanted side reactions and, for example, oxide formation, thus ensuring particularly high long-term stability and robustness. For example, the protective gas comprises nitrogen, argon, or an argon-hydrogen mixture, whereby the hydrogen content in the protective gas can be < 4 wt.%. In this embodiment, it can be provided, in particular, that the reactor is filled with protective gas in addition to the described areas.

It may further be preferred that the process be carried out in a reactor with at least two interconnected reaction chambers, wherein the first region and the second region are arranged in the first chamber, wherein the third region is arranged in at least one of the first and second chambers, and wherein the fourth region and optionally the fifth region are arranged in the second chamber. In particular, this embodiment allows for a particularly simple process control in which silicon-containing solid bodies are continuously applied to the first region and high-purity silicon can be removed in liquid form in the fifth region or downstream of the fifth region. For example, the third region can be located at the bottom of the reactor and connect the first reaction chamber to the second reaction chamber. In the first reaction chamber, the second region can then be located above the third region and the first region above the second region. In the second reaction chamber, the fourth region can be located above the third region and the fifth region above the fourth region. In principle, however, further embodiments of the reactor or the arrangement of the respective regions are possible without departing from the scope of the present invention.

It may be particularly preferred that the electrolysis in the second

Reaction chamber generated thermal energy to temper the first This embodiment highlights a further advantage of the process described here: by using two interconnected reaction chambers, the excess energy generated during electrolysis is transported to the first chamber. Accordingly, energy can be saved and energy costs reduced. Furthermore, the process can thus be particularly sustainable.

In principle, the silicon can be generated directly in the fifth region and extracted therefrom and used accordingly for the production of silicon-containing components, such as for the manufacture of solar cells. In one embodiment, however, the resulting ultrapure silicon can also be crystallized directly. In other words, liquid silicon extracted from the fifth region can be recrystallized. This makes it possible, for example, for the produced silicon to be further processed directly in an industrial process, for example using a controlled crystallization technique, to manufacture corresponding products, such as wafers. Such crystallization techniques are described, for example, in Czochralski, "Directional Solidification," and near-net-shape methods in WC O'Mara, RB Herring, and LP Hunt, "Handbook of Semiconductor Silicon Technology," 1. Park Ridge, NJ: William Andrew, 1990. For this purpose, an additional reaction chamber can be provided for the crystal growth method, which, for example, is present in addition to a first reaction chamber for pretreatment or preconditioning, and a second reaction chamber for electrorefining, as described above. In this integrated process, the purified silicon is fed into the crystallization chamber while still in a molten state. This process is also referred to as the waste-to-wafer concept, in which silicon-containing solids are generated as waste on one side of the reactor, and crystallized products, such as monocrystalline ingots, are produced on the other side. These products can then be sliced and processed into wafers. Here, in particular, silicon-containing solids such as metallic silicon scrap can be introduced into the first reactor chamber, with high-purity liquid metallic silicon is produced, from which silicon-containing products, such as monocrystalline ingots, can be produced directly from the liquid state in a third reactor chamber.

It may further be preferred that the process be carried out continuously. As already indicated above, it is thus possible for silicon-containing solid bodies to be continuously applied to the second region as a first layer, the silicon being purified by the process and appearing as the fifth phase in highly pure form. The process according to the invention makes it particularly advantageous for the electrolyte or the fourth region to not become contaminated, or at least not to such an extent that regeneration or replacement would be necessary after a short time, even with comparatively impure silicon bodies as the starting product. This makes it possible to operate a system with long-term stability, which provides the prerequisite for a continuous process. This is a significant advantage over prior art solutions.

The invention is explained below by way of example with reference to the attached drawings, wherein the features shown below can represent an aspect of the invention both individually and in combination, and wherein the invention is not limited to the following drawing, the following description and the following embodiment.

They show:

Fig. 1 is a schematic representation of an embodiment of a reactor for carrying out a process according to an embodiment of the present invention;

Fig. 2 is a schematic representation of the functioning of the method according to the invention; Fig. 3 is a schematic representation of an embodiment of a further reactor for carrying out a method according to an embodiment of the present invention; Fig. 4 is a schematic representation of an embodiment of a further reactor for carrying out a method according to an embodiment of the present invention; Fig. 5 is a schematic representation of an embodiment of a further reactor for carrying out a method according to an embodiment of the present invention; and Fig. 6 is a schematic representation of an embodiment of a further reactor for carrying out a method according to an embodiment of the present invention.

Figures 1 and 3 show a schematic representation of an embodiment of a reactor 10 for carrying out a method according to an embodiment of the present invention. The method that can be carried out by the reactor 10 is, in particular, a continuous process for purifying silicon-containing solid bodies, such as silicon scrap, in the liquid state using two interconnected reaction chambers, a first reaction chamber 12 and a second reaction chamber 14. When the method according to the invention is carried out in the reactor 10, five zones are formed, which is why the reactor 10 can be referred to as a five-layer electrolyzer when used. This represents a significant improvement over a known three-layer electrolyzer, enabling the use of silicon scrap of different qualities and also corresponding mixtures of products of different purity as starting products, without the need for constant replacement of the electrolyte. High-purity silicon (>99.9999%), for example for the production of wafers, can be produced through a five-stage electrolysis process in which impure metallic silicon-containing starting material is recycled and refined into high-purity silicon metal.

The reactor 10 comprises, as described, a first reaction chamber 12 and a second

Reaction chamber 14. An inlet 16 is also provided through which the The starting product, namely the silicon-containing solid bodies, can be introduced into the reactor 10 or into the first reaction chamber 12. This forms a first region 18, which comprises the silicon-containing solid bodies. The first region 18 is provided on a second region 20, which comprises a pretreatment slag. The second region 20 is located above a third region 22, which comprises a metal alloy, such as a copper alloy. The third region 22 forms the bottom region and connects the first reaction chamber 12 and the second

Reaction chamber 24. Above the third region 22, the second chamber contains the fourth region 24, which contains an electrolyte. Above this fourth region 24, the fifth region 26 is located, which contains highly pure silicon formed. The latter can be removed from the reactor 10 via an outlet 28.

In a reactor 10 equipped in this way, the treatment of silicon-containing solids can proceed as follows. Si scrap is introduced as the starting material in its solid state into the first reaction chamber 12. Through heat transfer from the molten pretreatment slag and external heat, the scrap begins to melt, changes its density, and passes through the pretreatment slag, leaving the gangue in the slag. The molten, preconditioned silicon is mixed with a molten Cu-Si alloy, for example. In the second reaction chamber 14, the metal alloy, for example, the Cu-Si alloy, is in contact with an electrolyte, referred to here as HES. Through an external potential introduced by a positive electrode 34 and a negative electrode 32, with the negative electrode 32 located in the fifth region 26 and the positive electrode in the third region 22, the ionized silicon or the Si ions move through the electrolyte of the fourth region 24 and create a layer of pure silicon as the fifth region, which, after reaching a certain height, continuously flows through the outlet 28 from the reactor 10 designed as a furnace. Accordingly, silicon is electrolytically purified or produced in the second reaction chamber 14. The entire The process is protected by a protective gas, such as an Ar atmosphere slightly enriched with H2, which can be introduced into the reactor 10 through the inlet 30. Furthermore, the reactor 10 will be heated, as indicated by the arrows, so that the reactor 10 or the reaction chambers 12, 14 can be heated to a treatment temperature.

The process taking place in the reactor is further illustrated in Figure 2. Figure 2 describes in more detail the mechanism that purifies the starting product, i.e., the silicon-containing solids, throughout the entire process. In the first reaction chamber 12, the pretreatment slag in contact with the silicon-containing solids begins to absorb impurities with high oxygen affinity, including the oxides already present. Molten or molten silicon flowing through the second region 20 is assimilated by the metal alloy in the third region 22, leaving behind metallic impurities such as Fe, Cu, Pb, Ag, and Sn, with lower oxygen affinity than silicon. The presence of copper in the metal alloy designed as a bottom alloy in the third region 22 is advantageous due to its high density and the ability of this metal to dissolve other metals and thus reduce the activity of smaller elements that could contaminate the electrolyte in the second reaction chamber 14. In the second reaction chamber 14, the formed, not yet completely pure silicon comes into contact with the electrolyte of the fourth region 24. This is advantageously designed such that it has a higher density than liquid silicon and possesses a relatively high electrical conductivity. The electrolyte contains, for example, a high concentration of SiO . By applying a potential between the electrodes 32, 34, which are connected to a voltage source 36, Si ions are selectively transferred from the lower melt to the upper, i.e., fourth, region 24. In particular, if the electrolyte in the fourth region 24 has a higher density than pure liquid silicon, this floats on top as the fifth region and can be removed from the outlet 28. Figure 3 further shows that, in addition to the first reaction chamber 12 and the second reaction chamber 14, the reactor 10 has a third chamber 15 for receiving the produced pure silicon. All chambers can be designed as crucibles. The empty spaces between the crucibles, made of aluminum oxide, for example, are filled with a monolithic refractory material, also based on Al2O3. Furthermore, it is shown that the reactor 10 can be arranged in a crucible 42, made of graphite, for example, and can be closed at its top by a ceramic lid 40, made of AlOa fluoride material, for example. Furthermore, a thermocouple 38 for controlling the temperature of the reactor 10 with an induction coil is shown. The structure is thus heated with an induction coil and installed in the crucible 42, which serves as a susceptor for the induction field and homogenizes the heat transfer to the system. This reactor 10 could be operated as follows. The scrap was fed in its solid state from the top of the first reaction chamber 12. The feed material was either granules (kerf or VNS) or chips (PV cells or metallurgical Si). The process began by heating the reactor 10 to 1350°C. Cu chips were then added, and after melting, polysilicon was added until a concentration of between 30 and 40 wt.% was reached. After the silicon was added, a 20-minute holding time was required for homogenization. The pretreatment slag and the electrolyte were added sequentially. Once the slags were in the reaction chambers 12, 14, the reactor 10 was sealed with a ceramic cover, the electrodes 32, 34, roughly configured as cylindrical graphite electrodes, were positioned and connected, and the reactor 10 was heated to 1500°C. Finally, the electrolysis cell, i.e., the second reaction chamber 14, is switched on simultaneously with the feeding of scrap into the first reaction chamber 12. Regardless of the scrap tested, the quality of the produced silicon was similar in terms of purity, demonstrating the robustness of the process. Figure 4 shows a further embodiment of a reactor for carrying out the described method, wherein the same reference numerals generally apply to identical or comparable components. The reactor 10 according to Figure 4 differs from the reactor 10 shown in Figure 1 in particular in that a heater is explicitly shown, which is formed by induction coils 44 for heating the reaction chambers 12, 14.

The configurations of Figures 5 and 6 also differ essentially in the heating system used. Figure 5 shows a combination of an AC submerged arc furnace heater 46, for example, comprising a three-phase AC submerged arc for the first reaction chamber 12, and a burner 48, for example, powered by hydrogen (H2), for the second reaction chamber 14.

In Figure 6, corresponding burners 48 are provided for the first reaction chamber 12 and the second reaction chamber 14. The burner 46 for the first reaction chamber 12 can be operated, for example, by a methane/air or methane/oxygen mixture, for example with a lambda value of <1, and the burner 48 for the second reaction chamber 14 can be operated, for example, by hydrogen peroxide. A suitable target temperature is generally approximately 1500-1600 °C and is in any case higher than the melting temperature of silicon under the conditions prevailing in the reactor 10.

10 reactor

12 first reaction chamber

14 second reaction chamber

15 Third Chamber

16 Entrance

18 first area

20 second area

22 third area

24 fourth area

26 fifth area

28 Outlet

30 Entrance

32 Electrode

34 Electrode

36 Voltage source

38 thermocouple

40 lids

42 crucibles

44 Induction coil

46 AC submerged arc furnace heater

48 burners

Claims

Claims 1. A method for treating silicon-containing solid bodies in a reactor (10), characterized in that the method comprises the following steps: a) setting a treatment temperature in the reactor (10), wherein the treatment temperature is equal to or above the melting point of silicon under the conditions in the reactor (10); b) introducing silicon-containing solid bodies forming a first region (18) into the reactor (10), wherein c) a second region (20) comprising a pretreatment slag is provided such that the first region (18) is located above the second region (20); d) providing a third region (22), wherein the third region (22) comprises a metal alloy; e) providing a fourth region (24), wherein the fourth region (24) comprises an electrolyte; and wherein silicon is produced electrolytically in the fourth region (24); wherein f) the liquid silicon forms a fifth (26) region adjacent to the fourth region (24), the first region (18), the second region (20), the third region (22), the fourth region (24) and the fifth region (26) following one another in this order in a reaction direction through the reactor (10). 2. The method according to claim 1, characterized in that the method is carried out in a reactor (10) having at least two interconnected reaction chambers (12, 14), wherein the first region (18) and the second region (20) are arranged in a first reaction chamber (12), wherein the third region (22) is arranged in at least one of the first reaction chamber (12) and the second reaction chamber (14), and wherein the fourth region (24) and optionally the fifth region (26) are arranged in the second reaction chamber (14). 3. Method according to claim 2, characterized in that the third region (22) is present as a bottom region of the first reaction chamber (12) and the second reaction chamber (14) and connects the first reaction chamber (12) and the second reaction chamber (14) to one another, and in that in the first reaction chamber (12) the first region (18) and the second region (20) are present above the third region (22) and in that in the second reaction chamber (14) the fourth region (24) and the fifth region (26) are present above the third region (22). 4. The method according to claim 2 or 3, characterized in that thermal energy generated during the electrolysis in the second reaction chamber (14) is used to temper the first reaction chamber (12). 5. A method according to any one of claims 1 to 4, characterized in that the pretreatment slag comprises at least one of silicon oxide, calcium oxide, and aluminum oxide. 6. A process according to claim 5, characterized in that the pretreatment slag comprises the following components in the following proportions, expressed in wt.% based on the pretreatment slag: Silicon oxide in a range of > 10 wt% to < 80 wt%; Calcium oxide in a range of > 10 wt% to < 60 wt%; aluminum oxide in a range of > 5 wt% to < 30 wt%; lithium oxide in a range of > 5 wt% to < 20 wt%; and calcium fluoride in a range of > 0 wt% to < 15 wt%. 7. Method according to one of claims 1 to 6, characterized in that the metal alloy comprises a copper-silicon alloy. 8. The method according to any one of claims 1 to 7, characterized in that the electrolyte comprises silicon oxide and optionally further components selected from the group consisting of calcium oxide, barium oxide, aluminum oxide, magnesium oxide and calcium fluoride. 9. A method according to claim 8, characterized in that the electrolyte comprises the following components, expressed in wt.% based on the electrolyte: Silicon oxide in a range of > 20 wt% to < 60 wt%; Calcium oxide in a range of > 0 wt% to < 50 wt%; Barium oxide in a range of > 0 wt% to < 60 wt%; Aluminium oxide in a range of > 0 wt% to < 20 wt%; Magnesium oxide in a range of > 0 wt% to < 10 wt%; and calcium fluoride in a range of > 0 wt% to < 60 wt%. 10. The method according to any one of claims 1 to 9, characterized in that the silicon-containing solid bodies comprise silicon waste, in particular from solar cells, a cast silicon fraction, kerf and metallurgical silicon. 11. Method according to one of claims 1 to 10, characterized in that the silicon formed is crystallized. 12. The method according to any one of claims 1 to 11, characterized in that the method is carried out continuously. 5 13. Method according to one of claims 1 to 12, characterized in that the Process is carried out under an inert atmosphere.