Silicon deposit and process for the production thereof
Patent Information
- Application Number
- PCT/EP2026/051913
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-02-24
- Filing Date
- 2026-01-26
- Publication Date
- 2026-08-27
Smart Images

Figure 00000033_0000
Abstract
Description
[0001] SILICON DEPOSIT AND PROCESS FOR THE PRODUCTION THEREOF
[0002] FIELD OF THE INVENTION
[0003] The present invention relates to the electrochemical production of metals and metalloids from their corresponding oxides in a molten salt bath. More specifically, it concerns a process for electrowinning silicon from silicon dioxide and the resulting silicon deposit product. The invention provides an improved method for silicon production with high efficiency, reduced emissions, and enhanced product recovery.
[0004] BACKGROUND
[0005] Silicon is a critical material in numerous industrial applications, including but not limited to electronics, photovoltaics, chemical synthesis, and metallurgy.
[0006] One of the most common methods for producing high-purity silicon is the carbothermic reduction of silicon dioxide in an electric arc furnace. In this process, quartz (SiO2) is reacted with carbon at high temperatures, reaching 2000°C, to form metallurgical-grade silicon (MG-Si) with a typical purity of 98-99%. While this method is well established, it has several drawbacks. The process requires significant energy input and specially sourced feedstocks, contributing to high production costs and substantial carbon dioxide emissions due to the use of carbon-based reducing agents. Additionally, the final silicon product contains impurities such as boron, phosphorus, and transition metals, necessitating further purification steps.
[0007] As an alternative to carbothermic reduction, electrochemical extraction methods have been explored, including the Cambridge FFC process (Fray-Farthing-Chen process). In this process, silicon dioxide is directly reduced in a molten calcium chloride (CaCk) bath using an applied electric current, eliminating the need for carbon as a reducing agent. While this method reduces CO2 emissions and operates at lower temperatures than arc furnaces, it suffers from low current efficiency, low current density, contamination risks due to calcium incorporation, and electrode degradation in the molten salt environment. These limitations have hindered its commercial scalability.
[0008] Given these limitations, there is a need for an improved process that enables the efficient and scalable production of silicon while minimizing energy consumption, reducing emissions, and enhancing the purity and recoverability of the final product.SUMMARY OF THE INVENTION
[0009] In a first aspect, the present invention provides an improved electrochemical process for producing silicon deposit, as set out in claim 1. The process comprises immersing a cathode and an anode in a molten salt bath, dissolving a feedstock comprising silicon dioxide in the electrolyte, and applying an electric current at a high current density to induce the reduction of silicon dioxide at the cathode. A key aspect is the reduction reaction leads to the formation of a silicon deposit adhered to the cathode, which is subsequently extracted from the molten salt bath. In a second aspect, the present invention relates to the silicon deposit as set out in claim 10.
[0010] Compared to conventional carbothermic reduction processes, the process according to claim 1 operates at significantly lower temperatures and does not rely on carbonbased reducing agents, thereby substantially reducing CO2 emissions. Additionally, in contrast to alternative electrochemical processes, such as the Cambridge FFC process, the present method exhibits improved current efficiency, improved electrode current density, limits / reduces contamination, and enables better separation of the silicon product from the electrolyte matrix.
[0011] A particular advantage of the present invention lies in the formation of a self-supporting silicon deposit, as defined in claim 10. This structure is believed to contribute to improved current efficiency by preventing reoxidation of the reduced silicon during the electrochemical process. In electrochemical systems, reoxidation can occur if the reduced product is exposed to the anodic environment, where oxidation reactions take place. By forming a coherent, self-adhering deposit at the cathode, the invention physically separates the reduced silicon from the anode region, minimizing exposure to oxidizing conditions and thereby enhancing the overall silicon yield.
[0012] Furthermore, the self-supporting nature of the silicon deposit facilitates its extraction while leaving behind unreacted feedstock and electrolyte, as detailed in claims 6 and 7. This enables an efficient recycling loop, allowing the electrolyte to be reclaimed and reused in the molten salt bath, further improving process sustainability. Additionally, the process as defined in claim 9 achieves high current efficiency, thereby lowering energy consumption per unit of silicon produced.Some scientific literature has explored the direct electrochemical reduction of silica to form pure silicon particles and dense silicon deposits, often aiming to provide purity superior to the metallurgical grade silicon suitable for direct use for photovoltaic wafer fabrication or eliminate the some or all subsequent purification in processes such as Siemen silicon purification or utilizing direct solidification techniques. However, such approaches face a fundamental scalability limitation. The formation of high-purity silicon particles requires precise crystal growth conditions, which are inherently slow. As a result, when process throughput increases, the rate of silicon deposition exceeds the crystal growth rate, leading to low yield, poor product quality or the inability to form dense deposits. The present invention overcomes this limitation by forming a growing silicon deposit layer, which allows for continuous silicon accumulation without the constraints of slow crystal growth kinetics. This structural feature enables industrial scalability, making the process viable for high-throughput silicon production.
[0013] By employing a fully electrified process, the present invention reduces reliance on fossil fuels by creating the possibility to employ renewable electric energy, and allowed for mor efficient use or eliminates the need for carbon-based reducing agents. Both aspects allow for significant reduction of direct and indirect CO2 emissions for the production of MG-Si. Reaching high purity Mg-Si grades allows to limit the reliance of silicon purification processes which employ hazardous gas-phase precursors for further downstream use of the product. The ability to operate within a controlled molten salt environment further ensures greater process stability and scalability for industrial applications.
[0014] DETAILED DESCRIPTION OF THE INVENTION
[0015] The present invention concerns a process for the production of a silicon deposit, as well as said silicon deposit.
[0016] Unless otherwise defined, all terms used in disclosing the invention, including technical and scientific terms, have the meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. By means of further guidance, term definitions are included to better appreciate the teaching of the present invention.As used herein, the term "silicon deposit" refers to a solid material comprising metallic silicon crystals embedded in an electrolyte matrix, which is obtained as a product of the electrochemical reduction of silicon dioxide in a molten salt bath.
[0017] As used herein, the term "metal- or metalloid-deposit" refers to a solid material comprising metallic metal or metalloid crystals embedded in an electrolyte matrix, which is obtained as a product of the electrochemical reduction of said metal- or metalloid's oxide in a molten salt bath.
[0018] The term "metallic silicon crystals" refers to crystalline silicon structures that are substantially composed of silicon atoms and have a silicon content of at least 95 wt.% relative to the total weight of the metallic silicon crystals.
[0019] The term "electrolyte matrix" refers to a solidified or semi-solid structure comprising one or more salts, wherein said salts include at least one alkali metal halide, alkaline earth metal halide, or cryolite. The electrolyte matrix may serve as a carrier phase in which metallic silicon crystals are embedded.
[0020] The term "molten salt bath" refers to a liquid-phase electrolyte system maintained at an elevated temperature, wherein the electrolyte comprises one or more salts selected from alkali metal halides, alkaline earth metal halides, and cryolite, and in which silicon dioxide can be at least partially dissolved.
[0021] The term "cryolite" refers to a sodium aluminum fluoride compound, generally represented by the formula NasAIFs, but with possibly warring ration of Na to Al content following the extended formula of NaxAlyF(x+3y), which can function as a component of the electrolyte in the molten salt bath.
[0022] The term "feedstock" refers to a material introduced into the molten salt bath, said material comprising silicon dioxide in a form suitable for electrochemical reduction to silicon.
[0023] The term "cathode" refers to an electrically conductive solid component that is immersed in the molten salt bath and serves as the site of electrochemical reduction of silicon dioxide, resulting in the formation of silicon deposit.The term "anode" refers to an electrode immersed in the molten salt bath, which participates in the electrochemical process by undergoing oxidation or providing charge balance for the reduction reaction occurring at the cathode.
[0024] The term "sacrificial anode" refers to an anode that undergoes oxidation and is at least partially consumed during the electrochemical reduction process.
[0025] The term "electrically conductive solid material stable under process conditions" refers to a material capable of conducting electrical current while maintaining structural integrity and chemical stability within the molten salt bath at the temperatures and conditions employed during the process.
[0026] The term "process conditions" refers to the operational parameters under which the electrochemical reduction takes place, including but not limited to the composition of the molten salt bath, temperature, current density, and electrode configurations.
[0027] The term "current density" refers to the electrical current applied to the cathode, expressed as amperes per square centimeter (A / cm2), which governs the rate of electrochemical reduction of silicon dioxide.
[0028] The term "silicon content" refers to the proportion of silicon within a given material, expressed as weight percentage (wt.%), either relative to the total weight of the silicon deposit or relative to the weight of the metallic silicon crystals.
[0029] The term "self-supporting" refers to the ability of the silicon deposit to maintain its structural integrity without substantial deformation or collapse under its own weight within the specified temperature range.
[0030] The term "feature size" refers to the characteristic dimensions of the metallic silicon crystals within the silicon deposit, measured in micrometers (pm).
[0031] The term "continuous metallic silicon network" refers to an interconnected structure of metallic silicon crystals within the silicon deposit, providing a pathway for electrical conductivity and / or mechanical integrity.
[0032] The term "electrical conductivity" refers to the ability of the silicon deposit to conduct electrical current, typically expressed in units of Siemens per meter (S / m).The term "chloride content" refers to the proportion of chloride ions present in the silicon deposit, expressed as weight percentage (wt.%), which may arise from the composition of the electrolyte matrix.
[0033] The term "reclaiming the electrolyte" refers to the process of recovering and reusing the electrolyte trapped in the silicon deposit by allowing it to separate or be removed from the solidified material.
[0034] The term "drip" in the context of the electrolyte matrix refers to the controlled release or drainage of excess molten electrolyte from the silicon deposit during or after extraction from the molten salt bath.
[0035] The term "current efficiency" refers to the proportion of electrical current effectively utilized for the reduction of silicon dioxide to silicon, expressed as a percentage, relative to the theoretical maximum efficiency.
[0036] The term "distance between the silicon deposit and the anode" refers to the minimum spacing maintained between the cathodically deposited silicon deposit and the anode during the electrochemical reduction process, ensuring controlled deposition and minimizing interference between the electrodes.
[0037] The term "metallic silicon" and "silicon metal" refers to silicon in its elemental form at zero oxidation state, by extension elements which have metallic features such as metalloids and metals are used interchangeably in the context of silicon in this disclosure.
[0038] The term "metalloids" is used to refer to chemical elements that exhibit properties intermediate between metals and nonmetals, including semiconductivity, variable electrical and thermal conductivity, and amphoteric chemical behavior. The recognized metalloids include boron (B), silicon (Si), germanium (Ge), arsenic (As), antimony (Sb), and tellurium (Te).
[0039] "A", "an", and "the" as used herein refers to both singular and plural referents unless the context clearly dictates otherwise. By way of example, "a compartment" refers to one or more than one compartment.
[0040] "Comprise", "comprising", and "comprises" and "comprised of" as used herein are synonymous with "include", "including", "includes" or "contain", "containing","contains" and are inclusive or open-ended terms that specifies the presence of what follows e.g. component and do not exclude or preclude the presence of additional, non-recited components, features, element, members, steps, known in the art or disclosed therein.
[0041] The recitation of numerical ranges by endpoints includes all numbers and fractions subsumed within that range, as well as the recited endpoints.
[0042] The expression "% by weight", "weight percent", "%wt" or "wt%", here and throughout the description unless otherwise defined, refers to the relative weight of the respective component based on the overall weight of the formulation.
[0043] Whereas the terms "one or more" or "at least one", such as one or more or at least one member(s) of a group of members, is clear per se, by means of further exemplification, the term encompasses inter alia a reference to any one of said members, or to any two or more of said members, such as, e.g., any >3, >4, >5, >6 or >7 etc. of said members, and up to all said members.
[0044] Unless otherwise defined, all terms used in disclosing the invention, including technical and scientific terms, have the meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. By means of further guidance, definitions for the terms used in the description are included to better appreciate the teaching of the present invention. The terms or definitions used herein are provided solely to aid in the understanding of the invention.
[0045] Reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment, but may. Furthermore, the particular features, structures or characteristics may be combined in any suitable manner, as would be apparent to a person skilled in the art from this disclosure, in one or more embodiments. Furthermore, while some embodiments described herein include some but not other features included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the invention, and form different embodiments, as would be understood by those in the art. For example, in the following claims, any of the claimed embodiments can be used in any combination.In a first aspect, the present invention provides a process for electrowinning a metal or metalloid from a metal oxide or metalloid oxide, wherein the metal oxide or metalloid oxide is dissolved in a molten salt bath and subjected to electrochemical reduction at a cathode, forming a metal or metalloid-containing deposit that adheres to the cathode.
[0046] Preferably, the process comprises:
[0047] • providing a molten salt bath comprising an electrolyte, preferably the electrolyte is selected from: cryolite, an alkali metal halide, or an alkaline earth metal halide;
[0048] • dissolving a feedstock comprising a metal oxide or metalloid oxide in the molten salt bath;
[0049] • immersing an anode and a cathode in the molten salt bath, wherein the cathode comprises an electrically conductive solid material stable under process conditions;
[0050] • applying an electric current, preferably said current has a a current density of at least 0.5 A / cm2; to induce electrochemical reduction of the metal oxide or metalloid oxide at the cathode;
[0051] • forming a metal or metalloid-containing deposit that adheres to the cathode, and
[0052] • extracting the cathode with the metal or metalloid-containing deposit adhered to it from the molten salt bath.
[0053] The scale of the electrochemical reduction and the resulting volume of the silicon deposit is defined by the total charge passed through the cell, expressed as the cumulative electric current over the duration of the process. This total charge ensures the formation of a self-supporting metallic network that can be extracted while remaining adhered to the cathode. In a preferred embodiment, the total charge passed through the cell is at least 2 kAh, more preferably at least 4 kAh, more preferably at least 6 kAh, more preferably at least 8 kAh, more preferably at least 10 kAh, more preferably at least 12 kAh, and most preferably at least 15 kAh. In a further preferred embodiment, the total charge passed through the cell is at most 20 000 kAh, more preferably at most 15 000 kAh, more preferably at most 10 000 kAh, more preferably at most 7 500 kAh, more preferably at most 5 000 kAh, more preferably at most 3 000 kAh, and more preferably at most 1 000 kAh, more preferably at most 500 kAh, more preferably at most 100 kAh, more preferably at most 75 kAh, more preferably at most 50 kAh, more preferably at most 40 kAh,more preferably at most 35 kAh, more preferably at most 30 kAh, and most preferably at most 25 kAh. In a preferred embodiment, the total charge passed through the cell is between 2 and 20 000 kAh, more preferably between 4 and 15 000 kAh, more preferably between 6 and 10 000 kAh, more preferably between 8 and 7 500 kAh, more preferably between 10 and 5 000 kAh, more preferably between 12 and 3 000 kAh, and most preferably between 15 and 2 000 kAh.
[0054] In a preferred embodiment, the metalloid oxide is silicon dioxide, and the process is utilized for the electrowinning of silicon. In a preferred embodiment, the electrolyte of the molten salt bath is selected from the group consisting of cryolite, an alkali metal halide, an alkaline earth metal halide, or mixtures thereof. More preferably, the electrolyte of the molten salt bath is selected from the group consisting of cryolite, an alkali metal fluoride, an alkaline metal fluoride mixed with aluminum fluoride, an alkali metal chloride, or an alkaline earth metal fluoride. Even more preferably, the electrolyte of the molten salt bath is selected from the group consisting of cryolite, an alkali metal fluoride, an alkali metal fluoride mixed with, or an alkaline earth metal fluoride. Still more preferably, the electrolyte of the molten salt bath is selected from the group consisting of cryolite, an alkali metal fluoride, or an alkali metal fluoride mixed with aluminium fluoride. Yet more preferably, the electrolyte of the molten salt bath is selected from the group consisting of cryolite or an alkali metal fluoride mixed with aluminium fluoride. Most preferably, the electrolyte of the molten salt bath is cryolite.
[0055] Suitable alkali metal halides for use in the electrolyte include sodium chloride, potassium chloride, lithium chloride, sodium fluoride, potassium fluoride, and lithium fluoride. Suitable alkaline earth metal halides include calcium chloride, magnesium chloride, barium chloride, calcium fluoride, magnesium fluoride, and barium fluoride. Among these, the fluoride salts and cryolite are preferred as they provide improved electrolyte conductivity and stability at high temperatures6. Specifically, the term "cryolite" as used herein refers to sodium aluminum fluoride compounds generally represented by the formula NaxAlyF(x+3y).
[0056] The molten salt bath is preferably maintained at between 900°C and 1100°C, more preferably at least 920°C, more preferably at least 930°C, more preferably at least 940°C. The molten salt bath is preferably maintained at a temperature of at most 1075°C, more preferably at most 1050°C, more preferably at most 1025°C, most preferably at most 1000°C. This ensuring that silicon dioxide and cryolite are fully dissolved, yet the formed silicon precipitates. These conditions are essential fordeposit formation. These conditions lead to high dissolution of silica in the electrolyte while minimizing volatilization and decomposition of electrolyte components.
[0057] Joule heating was observed to play a role in maintaining the bath temperature, particularly as the current density increases. In scaled-up industrial applications, it is expected that Joule heating will be a primary heat source, reducing or eliminating the need for external furnace heating. This can contribute to lower energy consumption and improved thermal efficiency of the process.
[0058] The feedstock comprises a metal oxide or metalloid oxide, which may be introduced as a solid ora pre-dissolved compound in the electrolyte. The preferred embodiment involves silicon dioxide, which can be provided in the form of quartz, fumed silica, precipitated silica, or silica-containing industrial by-products. In a preferred embodiment, the particle size of the feedstock ranges from 1 pm to 3 000 pm, more preferably from 10 pm to 1 000 pm. The preferred particle size of the feedstock may range from 1 pm to 500 pm, preferably 10 pm to 200 pm, to facilitate dissolution in the molten salt bath.
[0059] In a preferred embodiment, the bath contains at least 5 wt.% SiO?, more preferably at least 8 wt.% SiC>2, more preferably at least 10 wt.% SiC>2, more preferably at least 12 wt.% SiC>2, more preferably at least 15 wt.% SiO?. In a preferred embodiment, the amount of SiO? in the molten salt bath is maintained between 10 and 20 wt.% throughout the process, particularly throughout the reduction occurring when current is applied. In a preferred embodiment, the metal oxide or metalloid oxide-electrolyte solution is saturated with metal oxide or metalloid oxide throughout the process. More preferably, the metal oxide or metalloid oxide-electrolyte solution is oversaturated with metal oxide or metalloid oxide throughout the process. In a particular preferred embodiment, the bath contains at least 10 wt.% SiC>2, more preferably at least 12 wt.% SiC>2, more preferably at least 14 wt.% SiC>2, more preferably at least 15 wt.% SiC>2, more preferably at least 15 wt.% SiC>2, more preferably at least 16 wt.% SiC>2, more preferably at least 17 wt.% SiC>2. In a preferred embodiment, the bath contains at most 20 wt.% SiC>2, more preferably at most 19 wt.% SiC>2, more preferably at most 18 wt.% SiC>2, most preferably at most 17 wt.% SiC>2. These amounts are particularly preferred in cryolite as electrolyte. Saturation of the cryolite bath is observed around 10 wt.%. By working at around 15 wt.% SiC>2, a two-phase system was observed, with a denser, SiO2-rich sludge phase accumulating at the bottom of the crucible. To ensure continuous operation, it was found beneficial to pre-feed the bath with an excess of SiC>2, allowing for abuffer layer of quartz that ensures constant dissolution during electrolysis. As added benefit, this keeps the concentration of reactant in the electrolyte constant and resulted in improved process stability. This also allows for bulk-feeding reactant, rather than the need to continuously feed reactant, simplifying reactor and bath setup. In a further preferred embodiment, the electrodes and silicon deposit do not touch the denser SiO2 phase at the bottom. As the electrodes are typically hanging, suspended in the molten salt bath ensuring they hang sufficiently above the bottom of the bath is easily achieved. More preferably, the distance between the denser, glass-like SiO? rich phase and the electrode is at least 5 cm, more preferably at least 10 cm, more preferably at least 15 cm, most preferably at least 20 cm.
[0060] The cathode comprises an electrically conductive solid material stable under process conditions. In a preferred embodiment, the cathode comprises graphite, due to its high electrical conductivity, chemical stability, and mechanical durability. In alternative embodiments, the cathode comprises a refractory metal, such as tungsten, molybdenum, or tantalum, or a silicon-based material, such as doped silicon or silicon carbide. As deposit formation adhering to the electrolyte is essential to the process, the cathode needs to support the weight of the formed deposit. In a particularly preferred embodiment, the cathode comprises graphite or silicon-based material, more preferably said silicon-based material is chosen from doped silicon or silicon carbide.
[0061] In a preferred embodiment, the anodes are setup symmetrically with respect to the cathode. More preferably, electrode plates are utilized and each cathode plate is provided with two anode plates. As an adherent metal or metalloid deposit is envisioned, symmetric growth of the deposit with respect to the cathode is desirable to appropriately distribute weight and maximize deposit growth kinetics.
[0062] The anode is immersed in the molten salt bath. The process can function with both sacrificial anodes as well as inert anodes. In a preferred embodiment, the anode is a sacrificial anode, more preferably made of graphite.
[0063] In alternative embodiments, the anode comprises a inert or semi-inert material, preferably evolving oxygen at the anode under process conditions. More preferably, the anodes' surface comprises, preferably consists essentially of an inert material allowing for oxygen evolution, under process conditions. This minimizes degradation and contamination of the molten salt bath. Suitable materials are known to be metals forming conductive passivation layer and their alloys as well as conductive oxides stable under process conditions and their sintered mixes known as cermets. Morespecifically candidates can be found among: tin oxide, ruthenium oxide, or iridium oxide, iridium, chromium, nikel, copper, tungsten and their alloys for example Ni-AI, Cu-AI, Cr-Fe or Ir-W.
[0064] In a preferred embodiment, the anode is graphite and the ratio by weight of graphite anode to silicon produced is at most 1 / 1, more preferably at most 9 / 10, more preferably at most 8 / 10, more preferably at most 7 / 10, more preferably at most 6 / 10. In a preferred embodiment, the anode is graphite and weight of graphite anode relative to the total is at least 150 g / kAh, more preferably at least 160 g / kAh, more preferably at least 170 g / kAh, more preferably at least 180 g / kAh, more preferably at least 190 g / kAh, most preferably at least 200 g / kAh. When graphite as sacrificial anode the anode consumption is about 140-145 g / kAh. This corresponds to about 550 kg of graphite per ton of silicon produced. However, it is preferred to provide an small excess of the sacrificial anode to improve process stability and efficiency.
[0065] An electric current is applied at a current density of at least 0.25 A / cm2, more preferably at least 0.4 A / cm2, more preferably at least 0.5 A / cm2, more preferably at least 0.6 A / cm2, more preferably at least 0.7 A / cm2, more preferably at least 0.8 A / cm2, more preferably at least 0.9 A / cm2, more preferably at least 1.0 A / cm2. In another embodiment, the current density may be at most 5.0 A / cm2, preferably at most 3.0 A / cm2, more preferably at most 2.0 A / cm2, more preferably at most 1.5 A / cm2. In a more preferred embodiment, the electric current is reduced at the start of electrolysis and ramped up after the first 1 hour, more preferably 2 hours, more preferably 4 hours of the electrolysis reaction. The electrolysis reaction starts when an electric current is applied to the electrodes.
[0066] In a preferred embodiment, the electrolysis voltage (cathode voltage) is at least 3.0V, more preferably at least 3.5V. In another preferred embodiment, the electrolysis voltage does not exceed 4.5V, more preferably remains below 4.2V. In a further preferred embodiment, the electrolysis voltage is maintained below 4.0 V, more preferably below 3.5V, most preferably below 3.0V at the start of the electrolysis, more preferably throughout the first 30 minutes, more preferably throughout the first hour, more preferably throughout the 2 hours, more preferably throughout the first four hours of electrolysis. In a further preferred embodiment, the electrolysis voltage is ramped up after the electrode is activated. In another further preferred embodiment, the electrolysis voltage is ramped up after the throughout the first 30 minutes, more preferably throughout the first hour, morepreferably throughout the 2 hours, more preferably throughout the first four hours of electrolysis; more preferably in a gradual or step-wise manner.
[0067] In a preferred embodiment, the electrolysis is operated under current controlled conditions, wherein the applied electric current or current density is imposed and the resulting electrolysis voltage is monitored as a response of the electrochemical system. The voltage values described herein therefore reflect the measured cell voltage corresponding to the imposed current during different stages of the electrolysis, and serve to characterize the electrochemical behavior of the system during electrode activation and subsequent bulk deposition.
[0068] In a preferred embodiment, the electrolysis starts with a first step where the voltage is maintained below 2.0 V, more preferably below 1.8 V, more preferably below 1.6 V, more preferably below 1.4 V, and most preferably below 1.2 V. This first step is maintained for a charge of at least 0.01 kAh, more preferably at least 0.025 kAh, more preferably at least 0.05 kAh, more preferably at least 0.075 kAh, more preferably at least 0.1 kAh, more preferably at least 0.125 kAh, more preferably at least 0.15 kAh, more preferably at least 0.175 kAh, more preferably at least 0.2 kAh, and most preferably at least 0.25 kAh.
[0069] Subsequently, a second step is employed where the voltage is maintained below 2.5 V, more preferably below 2.4 V, more preferably below 2.3 V, more preferably below 2.2 V, and most preferably below 2.1 V. This second step is maintained for a charge of at least 0.01 kAh, more preferably at least 0.025 kAh, more preferably at least 0.05 kAh, more preferably at least 0.075 kAh, more preferably at least 0.1 kAh, more preferably at least 0.125 kAh, more preferably at least 0.15 kAh, more preferably at least 0.175 kAh, more preferably at least 0.2 kAh, and most preferably at least 0.25 kAh.
[0070] Following this activation, the electrolysis voltage is ramped up to an operating potential of at least 3.0 V, more preferably at least 3.1 V, more preferably at least 3.2 V, more preferably at least 3.3 V, more preferably at least 3.4 V, and most preferably at least 3.5 V. Throughout the bulk of the process, the electrolysis voltage is maintained at at most 4.5 V, more preferably at most 4.4 V, more preferably at most 4.3 V, more preferably at most 4.2 V, more preferably at most 4.1 V, and most preferably at most 4.0 V. This operating potential is maintained for a charge of at least 1800 kAh, more preferably at least 2000 kAh, more preferably at least 2500kAh, more preferably at least 3000 kAh, more preferably at least 4000 kAh, and most preferably at least 5000 kAh.
[0071] In a preferred embodiment, the electrolysis is initiated in a current density ramp step in which the current density is maintained below below 0.30 A per square centimeter, more preferably below 0.20 A per square centimeter, more preferably below 0.10 A per square centimeter and most preferably below 0.05 A per square centimeter. In a further preferred embodiment, during this ramp step the current density is increased to at least 0.30 A per square centimeter, more preferably at least 0.40 A per square centimeter and most preferably at least 0.50 A per square centimeter. Further more preferably, the increase is being carried out in discrete steps each having a magnitude between 0.01 and 0.15 A per square centimeter, more preferably 0.01 and 0.10 A / cm2, more preferably 0.02 and 0.05 A / cm2. Following completion of the current density ramp step, the electrolysis is preferably continued at a current density above 0.50 A per square centimeter, more preferably above 0.60 A per square centimeter, more preferably above 0.65 A per square centimeter, and most preferably in the range up to about 0.70 A per square centimeter.
[0072] For silica reduction to silicon in cryolite, these high current densities maintained throughout the bulk of operation, in combination with high current efficiencies naturally result in deposit formation as the growth rates exceed those that allow for metallic silicon production. The result is a silicon deposit comprising a porous, semi-continuous metallic silicon network in a matrix of cryolite. The inventors found the deposit primarily grows outwardly, without wishing to be bound by theory it is presumed to be the result of mass transfer limitations of the reactant (silica) within the porous structure as well as preferred crystal growth directions evidenced by the dendritic Si growth structures in the deposit.
[0073] However, applying high voltages and high current densities to clean electrodes results in over polarizing of said electrodes; leading to parasitic reactions and the formation of undesired byproducts, degradation of the electrodes and I or intercalation of ions into the electrodes and I or silicon deposit. These effects can be largely avoided by activation of the electrodes, particularly cathode, through a gradual ramp up. The gradual ramp up promotes the formation of dendritic Si, resulting in an activated electrode surface. The dendritic Si continues to grow to the adherent silicon deposit on the cathode, which beneficially allows for accelerated growth rates and thus silicon production rates due to the added cathodically activeSi surface, thus enabling high current densities at high current efficiencies. In this stage at the aforementioned conditions, parasitic reactions did not appear to be an issue and the current efficiency of silicon formation remained high, thus greatly benefiting upscaling of the silicon production.
[0074] Following the initial activation of the electrode through the voltage ramp-up period, it is critical to maintain the electrical parameters within the specified operating ranges to ensure the continuous growth of the porous dendritic silicon network. It has been found that significant reductions in the electrolysis voltage or current density during the bulk deposition phase can lead to the filling of the existing porous structure with dense silicon deposits. While such filling may result in a more dense product, it fundamentally alters the morphology of the deposit and significantly reduces the active surface area of the cathode. Consequently, the process preferably avoids drops in voltage or current density after the ramp-up period has been completed. If the voltage or current density drops significantly below the established operating set points, the high surface area structure must be re-developed through a subsequent activation phase or a new voltage ramp, which results in process inefficiencies and increased energy consumption.
[0075] To ensure process stability and the structural integrity of the growing silicon deposit, the current density is maintained such that it does not drop below a specific lower bound during the bulk electrolysis period. In a preferred embodiment, the current density after the initial ramp-up period is maintained at at least 0.25 A / cm2, more preferably at least 0.3 A / cm2, more preferably at least 0.35 A / cm2, more preferably at least 0.4 A / cm2, more preferably at least 0.45 A / cm2, and most preferably at least 0.5 A / cm2. Furthermore, to prevent the undesired filling of the dendritic structure, the electrolysis voltage after the initial ramp-up period is maintained at at least 2.6 V, more preferably at least 2.7 V, more preferably at least 2.8 V, more preferably at least 2.9 V, more preferably at least 3.0 V, and most preferably at least 3.5 V. By maintaining these parameters above these lower bounds, the process ensures that the rate of silicon accumulation continues to favor the expansion of the dendritic network rather than the densification of the existing matrix, thereby allowing for continuous high-throughput operation and consistent product quality.
[0076] During electrochemical reduction, a metal- or metalloid-containing deposit adheres to the cathode, which is subsequently extracted from the molten salt bath. In a preferred embodiment, where silicon is being electrowon, the resulting silicon deposit comprises metallic silicon crystals dispersed within an electrolyte matrix.More preferably, the deposit is self-supporting, forming a continuous network of metallic silicon crystals within the electrolyte matrix. Most preferably, the metallic silicon crystals adopt a dendritic growth pattern, resulting in a porous or sponge-like metallic silicon structure. The pores may be voids or (partially) filled with electrolyte and other impurities. The dendritic structure promotes the formation of a selfadhering deposit, thereby preventing reoxidation of the reduced metal or metalloid, thereby ensuring high current efficiency and minimizing losses due to reverse electrochemical reactions.
[0077] In a preferred embodiment, the silicon deposit and the anode are separated by a distance of at least 10 mm, preferably at least 15 mm, more preferably at least 20 mm, to prevent unwanted oxidation reactions at the anode. In another preferred embodiment, the electrode spacing is at least 50 mm, more preferably at least 65 mm, and most preferably at least 75 mm.
[0078] Larger electrode spacing prevents electrical shorts due to the deposit bridging the interelectrode gap and improves charge efficiency. These conditions provide an optimum between ensuring sufficient spacing to prevent electrical shorts due to the growing deposit, while avoiding excess distance between the electrodes and avoiding excess bath size which is detrimental for energy efficiency.
[0079] In another embodiment, the electrode spacing is at least 5 cm + 0.5 cm per kAh, more preferably at least 7.5 cm + 0.6 cm per kAh, more preferably at least 10 cm + 0.8 cm per kAh, more preferably at least 15 cm + 0.9 cm per kAh, more preferably at least 20 cm + 1.0 cm per kAh, and more preferably at least 30 cm + 1.5 cm per kAh. In a further preferred embodiment, the electrode spacing is at most 50 cm + 5 cm I kAh, more preferably at most 30 cm + 5 cm / kAh, more preferably at most 30 cm + 4 cm / kAh, more preferably at most 30 cm + 3 cm / kAh.
[0080] In an embodiment, the anode-cathode distance is defined by a predefined separation distance (X) and a growth allowance distance (Y), wherein X is in the range of 2 cm to 10 cm, more preferably 2 cm to 7 cm, more preferably 2 cm to 5 cm, most preferably 3 cm to 5 cm; while Y is in the range of 2.0 mm to 10.0 mm per Ah / cm2, more preferably 2.0 mm to 7.0 mm per Ah / cm2, more preferably 2.0 mm to 5.0 mm per Ah / cm2, and most preferably 3.0 mm to 5.0 mm per Ah / cm2. This distance ensures adequate spacing for electrochemical deposition while maintaining efficient current distribution and minimizing operational constraints. These values are optimized for electrowinning of silicon.The growth rate of the silicon deposit is typically around 0.8 to 1.2 cm / kAh observed as a total thickness change of a two facial graphite cathode with immersed area of 20 x 20 cm2. This growth rate is expected to increase with increasing current density and current efficiency. By maintaining electrode spacing in the aforementioned ranges, an optimum is found between ensuring sufficient spacing between electrodes and the growing deposit, while avoiding excessive spacing and bath size.
[0081] The thickness of the silicon deposit, measured as the greatest perpendicular distance from the cathode surface to the outermost edge of the deposit, is a critical parameter for industrial scalability. The formation of a self-supporting, coherent silicon network allows the deposit to maintain its structural integrity as it grows. In a preferred embodiment, the process is carried out until the silicon deposit has a thickness of at least 0.5 cm, more preferably at least 1.0 cm, more preferably at least 1.5 cm, more preferably at least 2.0 cm, more preferably at least 2.5 cm, and most preferably at least 3.0 cm. In another embodiment, the process is continued until the silicon deposit has a thickness of at most 50 cm, more preferably at most 40 cm, more preferably at most 30 cm, more preferably at most 25 cm, and more preferably at most 20 cm, more preferably at most 15 cm, more preferably at most 12 cm, more preferably at most 10 cm, more preferably at most 8 cm, more preferably at most 7 cm, and most preferably at most 6 cm. In a preferred embodiment, the process is continued until the deposit thickness is between 0.5 to 15 cm, more preferably between 1 to 10 cm, more preferably between 2 to 8 cm, more preferably between and 3 to 6 cm. In another preferred embodiment, the process is continued until the deposit thickness is between 5 and 50 cm, more preferably between 10 and 40 cm, more preferably between 15 and 30 cm, and more preferably between 20 and 25 cm. This substantial thickness, combined with the continuous metallic network, enables the deposit to act as a highly developed cathode surface, facilitating high current densities while remaining firmly adhered to the graphite substrate. In a particular preferred embodiment, the silicon deposit is self-supporting up to a maximum thickness of 20 cm, more preferably a maximum thickness of 30 cm, most preferably a maximum thickness of 50 cm, at a temperature of 1100°C.
[0082] The process may further comprise reclaiming the electrolyte trapped in the metal-or metalloid-containing deposit, enabling its reuse in the molten salt bath. In a preferred embodiment, the metal- or metalloid-containing deposit is extracted in such a way that the electrolyte matrix is at least partially allowed to drip off before further processing. This reduces material losses and improves process efficiency by enabling direct reintegration of the molten salt into the electrolyte bath.The process achieves a current efficiency of at least 90%, preferably at least 92%, more preferably at least 95%, most preferably at least 97%. The formation of a self-supporting metal- or metalloid-containing deposit enhances current efficiency by preventing back-reaction losses that would otherwise occur if the reduced metal or metalloid were re-exposed to the anodic region.
[0083] By employing a fully electrified process, the present invention eliminates the need for carbon-based reducing agents, significantly reduces CO2 emissions, and operates at lower temperatures than conventional carbothermic reduction methods. Furthermore, the ability to reclaim and recycle unreacted feedstock and electrolyte results in a highly efficient, scalable, and sustainable electrowinning process.
[0084] In a particular preferred embodiment, the process further comprises the step of reclaiming the electrolyte trapped in the silicon deposit. More preferably, said reclaimed electrolyte is reused in the molten salt bath.
[0085] In a preferred embodiment, the electrolyte matrix is at least partially allowed to drip from the silicon deposit during or after the extraction of the silicon deposit from the molten bath. Practically, the cathode and adherent deposit can be partially or fully raised above the molten bath, or alternatively moved to a reclaiming oven. By maintaining the temperature above the melting point of the electrolyte, the liquid electrolyte matrix is allowed to drip from the silicon deposit.
[0086] In a preferred embodiment, after electrolysis, the cathode with the deposit adhered to it is held in a controlled environment where molten electrolyte can drip off, thereby enriching the metallic silicon content of the deposit. This step ensures that the silicon purity is maximized before further processing or refining, while allowing recycling of the electrolyte, purity of the electrolyte allowing.
[0087] As the dendritic silicon structure is solid and self-supporting at these conditions, this provides an easy method to separate at least a part of the electrolyte from the metal or metalloid, preferably silicon. The resulting deposit thus has a higher concentration of silicon, and the reclaimed electrolyte can be reused. It is particularly advantageous that present process aims to avoid byproducts and parasitic reactions, allowing high reuse rates of the electrolyte by minimizing contamination. This reduces the overall electrolyte for a given amount of throughput significantly. Experiments revealed that the liquid electrolyte matrix can be removed from the deposit by allowing controlleddripping, leading to a higher silicon content in the final product. This effect was particularly noticeable in samples from spongy deposit regions, which showed significantly higher Si purity compared to more dense areas. This suggests that controlled electrolyte removal could be used as an additional purification step.
[0088] Although silicon dioxide is a preferred metalloid oxide, the process is applicable to the electrowinning of other metals and metalloids from their oxides, provided that a suitable molten salt bath and electrode configuration are employed. In alternative embodiments, the metal oxide may be selected from titanium dioxide, zirconium dioxide, aluminum oxide, or iron oxide. In a preferred embodiment, the metal or metalloid and the molten salt bath are selected such that, upon reduction, the metal or metalloid precipitates in solid form while its oxide remains in a liquid state, either as a molten phase or dissolved within the molten salt bath. The ability to form an adhering deposit on the cathode allows for enhanced metal recovery efficiency and improved process scalability.
[0089] In a second aspect, the present invention provides metallic deposit. Said metallic deposit is obtainable by a process in accordance with the first aspect. More preferably, said metallic deposit is produced by a process in accordance with the first aspect.
[0090] More preferably, said metallic deposit is a silicon deposit, comprising metallic silicon crystals embedded in an electrolyte matrix. The silicon deposit is obtainable as the product of the electrochemical reduction of silicon dioxide in a molten salt bath. The deposit exhibits a characteristic structure in which metallic silicon forms an effectively continuous phase or a dispersed crystalline phase, while the electrolyte matrix serves as a carrier phase, encapsulating the silicon.
[0091] It is noted that the metallic silicon preferably has a dendritic structure and is solid at both 20°C and 1000°C. In fact, silicon remains solid up to the typical Si melting point of 1415°C. Comparatively, the electrolyte is solid at 20°C but liquid at 1000°C. Consequently, the silicon deposit is effectively a dendritic, metallic silicon structure with porous cavities filled with liquid electrolyte at 1000°C in the molten salt bath. At elevated temperatures, the liquid electrolyte can be at least partially removed, leaving a more porous silicon deposit upon cooling. However, full removal of the electrolyte matrix requires further purification.The silicon deposit comprises at least 15 wt.% silicon, relative to the total weight of the deposit. In a preferred embodiment, the silicon content is at least 20 wt.%, more preferably at least 25 wt.%, even more preferably at least 30 wt.%, and most preferably at least 35 wt.%. In an alternative embodiment, the silicon content may be at least 40 wt.%, more preferably at least 50 wt.%, more preferably at least 60 wt.%, more preferably at least 70 wt.%, more preferably at least 75 wt.%, more preferably at least 80 wt.%, more preferably at least 85 wt.%, more preferably at least 90 wt.%, or higher, depending on the process conditions, particularly the degree of electrolyte removal during or after deposit extraction from the electrolytic cell. The silicon content is measured in the silicon deposit is preferably measured by elemental analysis.
[0092] The metallic silicon crystals or dendrites within the deposit have a silicon purity of at least 75 wt.%, relative to the total weight of the metallic silicon phase. In a preferred embodiment, the purity is at least 80 wt.%, more preferably at least 85 wt.%, even more preferably at least 90 wt.%, and most preferably at least 95 wt.%. In an alternative embodiment, the purity may be at least 96 wt.%, 97 wt.%, 98 wt.%, or 99 wt.%, ensuring minimal contamination with other elements such as aluminum, sodium, or other metal impurities from the electrolyte. While the silicon content in the deposit as a whole is limited, it is particularly advantageous that a phase of relatively pure silicon crystals or dendrites are formed. This is advantageous for its purification both in later and earlier stages; such as evidenced by the ability to remove electrolyte from the deposit by dripping. It is also provides many of the properties that are particularly advantageous for its production process.
[0093] The metallic silicon crystals preferably exhibit a feature size ranging from 1 pm to 10000 pm, depending on the electrochemical growth conditions. In a preferred embodiment, the feature size ranges from 10 pm to 5000 pm, more preferably from 50 pm to 2000 pm, and most preferably from 100 pm to 1000 pm. The crystal size distribution is influenced by current density, electrolyte composition, and temperature stability during the electrolysis process.
[0094] The silicon deposit is preferably self-supporting, meaning it maintains its structural integrity without substantial deformation or collapse over a wide temperature range. In a preferred embodiment, the deposit is self-supporting over a thickness of at least 5 cm, more preferably at least 10 cm, more preferably at least 15 cm, more preferably at least 20 cm, more preferably at least 25 cm, more preferably at least 30 cm, more preferably at least 35 cm, more preferably at least 40 cm, at atemperature of 1100°C, more preferably at a temperature of 1100°C and 20°C, most preferably over a temperature range of 20 to 1100°C. In another embodiment, the deposit is self-supporting for a thickness of at most 1 meter, more preferably at most 70 cm, more preferably at most 50 cm, over a temperature range of 20 to 1100°C. In this context, "self-supporting" refers to a deposit adhered to a cathode that, despite its substantial thickness, is able to maintain its structural integrity and remain adhered to the cathode without collapsing, cracking, or detaching due to its own weight. Furthermore, the deposit must retain these properties within the expected temperature range of the deposition process and subsequent handling or application. This includes resistance to thermal expansion, contraction, and potential softening or embrittlement at elevated or fluctuating temperatures, ensuring that it remains structurally sound and adherent across relevant thermal conditions. The term "thickness" or "maximum thickness" refers to the greatest perpendicular distance measured from the cathode surface to the outermost surface of the deposit at any point within the deposited area. It represents the thickest region of the deposit, which may occur due to variations in deposition conditions, material buildup, or process dynamics.
[0095] This property is attributed to the formation of a coherent silicon network within the electrolyte matrix, which provides mechanical stability. This property is highly desirable as it enables the adherence of the deposit to the cathode, enabling many of the benefits of present process. For example, improved current efficiency by reducing reverse reaction at the anode. Enabling high current densities and deposition rates because the coherent silicon network is believed to be sufficiently conductive, acting as the cathode surface. Enabling the cathode and silicon deposit to be raised at elevated temperature, thereby allowing removal of at least some of the electrolyte.
[0096] The silicon deposit contains an electrolyte matrix comprising at least one alkali metal halide, alkaline earth metal halide, or cryolite. In a preferred embodiment, the electrolyte matrix comprises cryolite, sodium chloride, potassium chloride, lithium chloride, calcium chloride, magnesium chloride, barium chloride, sodium fluoride, potassium fluoride, lithium fluoride, calcium fluoride, magnesium fluoride, or barium fluoride, or mixtures thereof. Further more preferably, the electrolyte matrix comprises cryolite or fluoride salts. Most preferably, the electrolyte matrix comprises cryolite. The composition of the electrolyte matrix influences the melting point, viscosity, and adhesion properties of the deposit, affecting both its separation from the cathode and its post-processing behavior.A significant advantage of the silicon deposit is its continuous metallic silicon network, which improves electrical conductivity and facilitates further purification processes. The resistivity of the silicon deposit is defined as its electrical resistivity at room temperature (RT), which provides a practical and reproducible measurement. In a preferred embodiment, the silicon deposit has an electrical resistivity of below 10 Ohm-cm, more preferably below 5 Ohm-cm, more preferably below 1 Ohm-cm, and most preferably below 0.5 Ohm-cm.
[0097] The silicon deposit preferably comprises chloride in an amount of less than 1 wt.%, preferably less than 0.5 wt.%, more preferably less than 0.1 wt.%, and most preferably less than 0.05 wt.%. Low chloride content is desirable as it avoids particular issues related to purification. Utilizing electrolytes which do not result in chloride contamination is therefore desirable. Advantageously, present process functions well with non-chloride electrolytes.
[0098] The silicon deposit preferably comprises silica (SiO2) in an amount of less than 10.0 wt.%, more preferably less than 5.0 wt.%, more preferably less than 1 wt.%, preferably less than 0.5 wt.%, more preferably less than 0.1 wt.%, and most preferably less than 0.05 wt.%. Low silica content is desirable as it avoids particular issues related to purification and improves downstream processing. Maintaining process conditions that limit residual SiO2 in the deposit is therefore desirable. Advantageously, the present process allows the formation of a silicon rich deposit while leaving unreacted silica in the molten salt bath and or in a denser SiO2 rich phase, which can be separated and recycled.
[0099] The silicon deposit of the present invention is characterized by a high surface-to-volume ratio of the metallic silicon crystals, which is a direct result of the dendritic growth promoted by the high current density and total charge passed during the process. This ratio is measured for the metallic silicon phase alone, effectively excluding the volume and surface area of the surrounding electrolyte matrix. In a preferred embodiment, the metallic silicon crystals within the silicon deposit exhibit a surface-to-volume ratio of at least 25 000 m2 / m3, more preferably at least 30 000 m2 / m3, more preferably at least 32 500 m2 / m3, more preferably at least 35000 m2 / m3, more preferably at least 37 500 m2 / m3, more preferably at least 40 000 m2 / m3, and more preferably at least 45000 m2 / m3, more preferably at least 50000 m2 / m3, more preferably at least 55 000 m2 / m3, more preferably at least 60 000m2 / m3, and most preferably at least 65 000 m2 / m3, preferably as measured by micro computed tomography (MicroCT).
[0100] The surface to volume ratio; as well as the porosity and determination of the structure of the metallic silicon within the silicon deposit are preferably determined by micro computed tomography, after removal of the electrolyte matrix.
[0101] Sample preparation and microCT method:
[0102] For this purpose the samples are prepared in accordance with the following steps: The sample is placed in an induction oven under Argon atmosphere at a temperature in the range from 1050 °C to 1150 °C for a duration of 30 hours, until the evaporation of the electrolyte matrix is complete and no visible fumes are present. The temperature is maintained strictly below 1414 °C, which corresponds to the melting point of silicon, in order to prevent melting of silicon and to preserve the original three dimensional architecture of the samples.
[0103] After removal of the electrolyte matrix, three dimensional imaging is performed by micro computed tomography. The samples are scanned using an EasyTom system available from RX Solutions. Image reconstruction and data processing are carried out using Avizo software. The micro computed tomography is performed using a small focal spot LaB6 filament operated at an accelerating voltage of 70 kV and a current of 140 pA, with a voxel size of 3.80 pm.
[0104] The method provides volumetric image data representative of the internal three dimensional structure of the samples while maintaining the structural integrity of the silicon phase.
[0105] In a preferred embodiment, the silicon deposit has a porosity of at least 30 percent, more preferably at least 40 percent, more preferably at least 45 percent, more preferably at least 50 percent, more preferably at least 55 percent, more preferably at least 57.5 percent, more preferably at least 60 percent, more preferably at least 62.5 percent, and most preferably at least 65 percent. In another embodiment, the porosity of the silicon deposit is at most 85 percent, more preferably at most 80 percent, more preferably at most 77.5 percent, more preferably at most 75 percent, more preferably at most 72.5 percent, more preferably at most 70 percent, and more preferably at most 67.5 percent. The porosity of the silicon deposit is preferably 30 to 85 percent, more preferably 50 to 80 percent, more preferably 55 to 75 percent, and most preferably 60 to 70 percent. This high level of porosity is maintained by the self-supporting nature of the dendritic silicon network which prevents the collapse of the structure even at the elevated temperatures of the molten salt bath; and a result of the process which desires a high degree of surfaceto volume ratio. This high level of porosity is maintained by the self-supporting nature of the dendritic silicon network, which prevents the collapse of the structure even at the elevated temperatures of the molten salt bath. The porous nature of the network is particularly advantageous as it provides an cathodically active surface for further growth of the network.
[0106] The formation of a self-supporting silicon deposit provides multiple advantages in the electrochemical process. First, it prevents reoxidation of the reduced silicon by maintaining a stable deposit at the cathode, thereby improving current efficiency and product yield. Second, it facilitates easy separation of the product from unreacted feedstock and electrolyte, enabling high material recovery rates.
[0107] The self-supporting, porous nature of the deposit also allows for controlled dripping of residual electrolyte, which improves silicon purity and reduces the need for additional refining steps. In experimental trials, regions of the deposit that underwent controlled electrolyte removal exhibited higher silicon content, suggesting that this technique can be employed to optimize product quality.
[0108] In summary, the present invention provides a silicon deposit that is structurally stable, compositionally optimized for high silicon yield, and capable of self-separating from the electrolyte phase, making it particularly advantageous for industrial silicon production. The deposit's electrical and mechanical properties further contribute to its suitability for downstream processing, including purification and metallurgical applications.
[0109] EXAMPLES
[0110] The experiments were conducted to optimize conditions for silicon deposit production using the Molten Salt Electrolysis (MSE) process. A standardized setup was employed across all experiments, involving a crucible containing a cryolite-quartz sand bath. The initial SiC>2 content in the bath was either saturated or oversaturated to ensure a consistent supply of reactants during electrolysis. Electrodes were graphite plates, suspended in the molten salt bath, in each experiment having one central cathode and two external anodes. In this setup, the current density was calculated considering both faces of the cathode. Electrodes were arranged with anode-cathode spacing ranging from 30 to 75 mm, chosen to balance efficient product deposition and minimize electrical shorts.The furnace temperature was maintained at 1000°C throughout the electrolysis experiments. In all experiments, temperature was ramped up at a rate of 100°C I hour until the bath was stable at a temperature of 1000°C prior to the start of electrolysis. The furnace was kept in nitrogen atmosphere to protect graphite elements from oxidizing in atmospheric air content of oxygen.
[0111] In all experiments, electrolysis was initiated by a gradual ramp up of the current density, in stepwise increases every 30 minutes, over a period of 150 minutes from zero to the set point for the experiment, unless stated otherwise.
[0112] The voltage reported for all experiments is the rectifier output voltage. Electrode voltage could not be measured in these experiments. Consequently, these values include voltage losses between rectifier and electrodes.
[0113] Quartz sand was fed stoichiometrically to compensate for its consumption during the process. Current densities were incrementally increased to evaluate their effect on deposit growth, product quality, and process efficiency. The results provide valuable insights into the scalability of the MSE process.
[0114] Example 1
[0115] In this experiment, the voltage was initially ramped up from 2V to 3.5V over a period of 150 minutes. Subsequently, the rectifier voltage was maintained at 3.5V, resulting in an average current of 150A. This corresponds to a current density of 0.25 A / cm2across the cathode area of approximately 20 x 15 x 2 cm2(two faces). The total charge passed was 15.3 kAh. The bath consisted of 21.7 kg of cryolite, comprising 18.3 kg of fresh cryolite and 3.4 kg of reused cryolite bath, along with 2.0 kg of quartz sand initially. The reused cryolite bath contained dissolved quartz from the previous process.
[0116] During operation, quartz sand was fed at a rate of 50 g per 45 Ah to maintain a SiC>2 content of approximately 10%.
[0117] Results: The process exhibited thermal stability throughout, with no significant irregularities. The current gradually decreased from 170A after the initial ramp-up, to 140A by the end of the experiment. Without wishing to be bound by theory, this is likely due to the resistance of the formed deposit layer.
[0118] The deposit growth rate was approximately 8 mm per 1 kAh, with the final depositthickness reaching about 15 cm (total, including both sides of the cathode). The total deposit mass produced was 11.8 kg, with an average silicon content of 36.5% by weight as determined by elemental analysis. The graphite anode consumption rate was calculated to be 145 g / kAh, while the silicon yield current efficiency was determined to be close to 100%.
[0119] The surface of the deposit was smooth, indicating no pockets of anodic gas stuck in the deposit. Upon autopsy of the deposit, no balls or droplets indicative of aluminium formation were noted. Aluminium formation would result in balls or droplets as it would be molten at the process temperature; and the same applies to the Al-Si 12.6%-87.4% eutectic alloy. However, slight asymmetry in anode consumption was observed, attributed to differences in electrical contact quality.
[0120] The bath revealed two phases. The top phase was softer, easier to remove and had a lighter bulk color. The bottom phase was hard, glassier and about 2 cm thick. The top phase was analysed in double, once at the interface with furnace atmosphere at the top of the bath and once at the interface with the bottom phase. The top phase near the air interface corresponded to 86 wt.% cryolite, 14 wt.% quartz. The top phase near the bottom phase interface corresponded to 82 wt.% cryolite, 18 wt.% quartz. The bottom phase was analysed and corresponded to 57 wt.% cryolite, 42 wt.% quartz. This suggests the formation of a quartz-glass phase. It also indicates an increase in the SiO? concentration with increasing depth in the cryolite bath.
[0121] Example 2
[0122] The second experiment aimed to increase the current density to 0.4-0.5 A / cm2, corresponding to a current of 300-400A. The current was not constant due to DC cable temperature limitations.
[0123] The electrodes were adjusted to a higher crucible position, resulting in electrode size of 20.0 cm x 21.0 cm x 2 (two faces). The total charge passed was 15.3 kAh. The bath comprised 27.46 kg of material, including 11.77 kg of reused cryolite, 12.69 kg of fresh cryolite, and 3.0 kg of quartz sand initially. The initial SiC>2 content was thus approximately 17% (oversaturation). Quartz sand was fed in larger batch sizes (275 g per 500 Ah, stoichiometric rate) to prevent feeding system clogging triggered by frequent small batch dispensing. The electrolysis voltage was maintained below 4.0V. The deposit was allowed to drip-out for a period of time upon completion of the electrolysis.Results: Current ramping at the start of the experiment showed a downward step in the voltage evolution seen in the first current steps for which current limit was artificially imposed by the rectifier setting. This indicates that developing the first layers of deposit is essential for high current operation as it contributes to the activation of the cathode with a highly developed surface of the rough silicon deposit.
[0124] Deposit growth was more rapid than in the first experiment, with a rate of approximately 1.1 cm per 1 kAh, resulting in a final deposit thickness of 15.5 cm. The deposit mass produced was 12.6 kg, with an average silicon content of 30% by weight. The graphite anode consumption rate was consistent with the first experiment at 141 g / kAh. The silicon yield efficiency was slightly reduced to 93%. While thermal stability of the process itself was maintained, challenges with cable overheating and reduced deposit quality highlighted the need for further optimization of the electrical setup.
[0125] The deposit showed denser and porous sponge regions. The deposit was broken and pieces of both denser and porous sponge regions were investigated. Elemental analysis of the dense regions showed a Si concentration of almost 30%. Elemental analysis of the porous sponge regions indicated a Si concentration between 45% and 50%. This is seen as an indication that allowing electrolyte to drip from the deposit can be a major improvement in the Si concentration of said deposit. At this time, the liquid salt matrix can be removed from limited areas of the deposit but these areas show a higher mass content of Si.
[0126] Example 3
[0127] This experiment was designed to achieve a current density of 1 A / cm2, corresponding to a current of 600A. However, due to cathode failure, the experiment operated primarily at 400-500A, with a total charge of 9.3 kAh. The bath composition was similar to the second example, with a total weight of material of 27.42 kg and an initial SiC>2 content of 17%. Cryolite was reused from previous experiments, with quartz and fresh cryolite added to obtain the desired oversaturation. Quartz sand feeding was aimed to maintain a stoichiometric rate, adjusted to 550 g per 1000 Ah to account for the higher current density. The bath and furnace setup included modifications such as doubling electrical connections to reduce resistance and accommodate higher currents. The deposit was allowed to drip-out for a period of time upon completion of the electrolysis.Polarization (IV) experiments were conducted at 0 kAh, 2.2 kAh and 9.0 kAh in the current range from 50A to 600A.
[0128] Results
[0129] The cathode broke throughout the experiment, interrupting the experiment. The cathode is believed to have broken at 17 hours; as there was an unusual dip in current followed by later recovery at this time in the current measurements. The break of the cathode plate was noticed at a planned inspection at 32h, which lead to the termination of the experiment.
[0130] Therefor this experiment did not result in stable operation. Prior to cathode failure, transient operation of 400-500A and peaks above 600A (during the recording of polarization curves) were achieved without issues.
[0131] Despite the interruption, the deposit growth rate was comparable to the second experiment, with a rate of approximately 1.1 cm per 1 kAh. The total deposit mass produced was 8.16 kg, with an average silicon content of 25% by weight. The graphite anode consumption rate was calculated to be 146 g / kAh, and the silicon yield efficiency was reduced to 84%, likely due to the cathode failure.
[0132] Despite the deposit development differences and cathode failure between IV experiments conducted at 0 kAh, 2.2 kAh and 9.0 kAh, the voltage range of all curves is very similar. Voltage gain was shown to be in linear proportion to the current which indicates no specific limit for the electrolysis kinetics up to 600A. Cell resistance was persistent at about 7-8 mOhm; with no clear impact from deposit deposition. The 3 IV curves showed a different OCV (read as the intercept point of the ohmic curve at zero current), with a 0.4V difference, the highest value was recorded for the 2.2 kAh experiment. No indication of current limits or changes in the reaction regime were observed or suggested in this data, both in terms of applied current and in terms of the growth of deposit changing the cathode.
[0133] Similar to example 2, dense and porous regions were observed. The porous regions had a similar and high silicon content compared to those in example 2. The dense regions had a lower silicon content compared to those in example 2, with a silicon content between 20 and 25%. This may suggest that higher deposit growth rates due to higher current density increases the ratio of electrolyte to metallic silicon in the deposit. In other words, the porosity of the dendritic silicon network is increased.This also suggests increasing benefits to electrolyte dripping from the silicon deposit with increasing current density.
[0134] At these higher currents, joule heating became increasingly influential. The furnace was operating at a significantly lower power setting to maintain the set temperature of 1000°C, once electrolysis was ongoing.
[0135] Across all experiments, the MSE process demonstrated potential for scalable silicon deposit production with consistent deposit growth and graphite consumption rates. The data indicate that higher current densities improve deposit growth rates but may reduce silicon content in the deposit, necessitating a balance between production speed and product quality.
[0136] Example 4
[0137] In this experiment, the structural properties of silicon deposits produced at high current densities were analyzed using high resolution micro computed tomography to evaluate the morphology of the metallic silicon network. The molten salt bath had a total weight of 27.5 kg, consisting of 14.5 percent initial silicon dioxide content and 85.5 percent reused cryolite. Quartz sand was fed at a stoichiometric rate of 275 g per 500 Ah to maintain reactant levels. The process utilized an initial current ramp starting at 25 A and increasing by 25 A increments over 10 hours to reach 225 A. Following this, the current was maintained between 200 and 225 A for 10 hours before setting the rectifier voltage at 7.0 V, which resulted in an operating current increasing from 300 to 400 A, corresponding to a current density of 0.5 to 0.67 A / cm2. A total charge of 9.6 kAh was passed during the process.
[0138] Polarization experiments were conducted at total charge intervals of 0 kAh, 3.4 kAh and 8.3 kAh. The results are shown in figure 1. The data indicated that at a fixed voltage, the current increased over time, which is attributed to the development of the cathode surface area. This shows that the formation of a highly developed cathode surface is beneficial to the energy efficiency of the process.
[0139] To analyze the deposit structure, 1 cubic centimeter samples were collected from the high current density zone of the cathodic deposit following cool down. The electrolyte matrix was removed by evaporation in an induction oven under an argon atmosphere at a temperature of at least 1150°C for 30 hours. The temperature was strictly maintained below the melting point of silicon (1414°C) to preserve the 3D dendritic structure.Micro CT scans were obtained using a voxel size of 3.80 micrometers, an accelerating voltage of 70 kV, and a 140 microampere current. Processing of the volumetric data revealed a highly interconnected dendritic network with a surface to volume ratio of 45600 to 49100 m2 / m3and a porosity of 61.72 to 68.59 percent.
[0140] Comparative example 5
[0141] A comparative experiment was conducted to evaluate the structural properties of silicon deposits produced at low current densities. The bath for this experiment had a total weight of 27 kg, with an initial silicon dioxide content of 14.8 percent and 85.2 percent reused cryolite. Quartz sand feeding was adjusted to 550 g per 1000 Ah. The current was maintained at 25 A for 1 hour before being increased to a constant 50 A, corresponding to a low current density of 0.08 A / cm2, for the remainder of the experiment. A total charge of 3.7 kAh was passed. Polarization experiments were avoided to ensure that silicon was not deposited at higher current densities even for short durations.
[0142] Following the same preparation and micro CT scanning protocol as described in Example 4, samples were collected and the electrolyte matrix was evaporated to resolve the 3D structure. Comparison of the micro CT data from Example 4 and Example 5 showed that the high current density deposit (Example 4) exhibited a significantly more developed dendritic network with higher porosity and a higher surface to volume ratio than the low current density deposit (Example 5).
[0143] Specifically, the low current density samples displayed a more compact metallic phase with a surface to volume ratio significantly below 34000 m2 / m3for all samples, as well as lower porosity between respective samples. This comparison confirms that operating at high current densities directly affects the morphology and properties of the formed silicon deposit.
[0144] The present invention is in no way limited to the embodiments described in the examples and / or shown in the figures. On the contrary, methods according to the present invention may be realized in many different ways without departing from the scope of the invention.
Claims
CLAIMS1. A process for producing silicon deposit, comprising the steps of:a. providing a molten salt bath comprising cryolite, an alkali metal halide or alkaline earth metal halide as an electrolyte;b. dissolving a feedstock comprising silicon dioxide in the molten salt bath; c. immersing an anode and a cathode in the molten salt bath, wherein the cathode comprises an electrically conductive solid material stable under process conditions;d. applying an electric current at a current density of at least 0.5 A / cm2to induce electrochemical reduction of silicon dioxide at the cathode, thereby forming a silicon deposit, wherein the electric current is applied for a duration such that a total current passed through the cell is at least 2 kAh ; ande. extracting the cathode with the silicon deposit adhered to it from the molten salt bath.
2. Process according to claim 1, wherein the molten salt bath is maintained at a temperature of at least 900°C and at most 1100°C during said process, preferably during steps b. to e..
3. Process according to any one of claims 1 or 2, wherein the cathode comprises graphite or a silicon-based material, more preferably said silicon-based material is chosen from doped silicon or silicon carbide.
4. Process according to any one of claims 1 to 3, wherein the current density is at least 1.0 A / cm2, preferably at least 1.4 A / cm2.
5. Process according to any one of claims 1 to 4, further comprising reclaiming the electrolyte trapped in the silicon deposit for reuse in the molten salt bath.
6. Process according to any one of claims 1 to 5, wherein the electrolyte matrix is at least partially allowed to drip from the silicon deposit during or after extraction of the silicon deposit from the molten salt bath.
7. Process according to any one of claims 1 to 6, wherein the silicon deposit and the anode are separated by a distance of at least 10 mm during step d.
8. Process according to any one of claims 1 to 7, wherein the process has a current efficiency of at least 90%.
9. A silicon deposit comprising metallic silicon crystals embedded in an electrolyte matrix; wherein the silicon deposit has a silicon content of at least 20 wt.% relative to the weight of the silicon deposit; and wherein the metallic silicon crystals have a silicon content of at least 95 wt.% relative to the weight of said metallic silicon crystals.
10. Silicon deposit according to claim 9, wherein said metallic silicon crystals are characterised by a surface-to-volume ratio of at least 40 000 m2 / m3as measured by the Sample preparation and MicroCT method.
11. Silicon deposit according to any one of claims 9 to 10, wherein said metallic silicon crystals are characterised by a porosity between 55% and 75% as measured by the Sample preparation and MicroCT method.
12. Silicon deposit according to any one of claims 9 to 11, wherein said silicon deposit is self-supporting up to a maximum thickness of 50 cm at a temperature of 1100°C.
13. Silicon deposit according to any one of claims 9 to 12, wherein said metallic silicon crystals form a continuous metallic silicon network, preferably said metallic silicon crystals form a dendritic structure.
14. Silicon deposit according to any one of claims 9 to 13, wherein said electrolyte matrix comprises an alkali metal halide, an alkaline earth metal halide or cryolite, preferably cryolite.
15. Silicon deposit according to any one of claims 9 to 14, wherein said metallic silicon crystals have a feature size of 1 to 10000 pm, preferably 10 to 1000 pm.