Industrial process for recycling permanent magnets
An industrial-scale process for recycling permanent magnets includes thermal treatment, mechanical cleaning, and hydrometallurgical recovery to efficiently separate and recover rare earth elements, addressing the challenges of handling large magnets and achieving high purity and sustainability.
Patent Information
- Application Number
- PCT/EP2025/069134
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2024-07-05
- Filing Date
- 2025-07-04
- Publication Date
- 2026-01-08
AI Technical Summary
Existing methods for recycling large permanent magnets, such as those from wind turbines, lack an efficient, automated, and continuous process for handling, separation, and coating removal, particularly due to their high coercivity and diverse assembly methods, which are not adequately addressed in current literature.
A process involving thermal treatment, vibratory separation, mechanical cleaning, and ultrasonic cleaning to demagnetize and separate magnet units, followed by reducing them to micronized powder, and a hydrometallurgical recovery method to extract rare earth elements efficiently.
The process achieves high separation efficiency, with over 95% of magnets being sorted into units, and recovers rare earth elements with a recovery rate exceeding 95% purity, while being environmentally sustainable and applicable at an industrial scale.
Smart Images

Figure IMGF000009_0001 
Figure IMGF000010_0001 
Figure IMGF000016_0001
Abstract
Description
[0001] INDUSTRIAL PROCESS FOR RECYCLING PERMANENT MAGNETS
[0002] The present application relates to the field of recycling of permanent magnets from different sources at an industrial scale, in an integrated, automated and continuous manner.
[0003] Background of the invention
[0004] With the global trend towards decarbonization, there is a growing shift towards the production of green energy, replacing fossil fuels. This has led to the increasing integration of permanent magnets in various industries, such as wind turbines and electric motors for vehicles or bicycles. However, these components have a limited lifetime, which raises concerns about their correct disposal and the preference for recycling. This is particularly crucial because these magnets contain critical raw materials such as Neodymium, Praseodymium, Dysprosium and Samarium, which may be in limited supply or face restrictions due to geopolitical factors. Recovering these materials has significant economic value.
[0005] Some efforts have already been made to manage End-of-Life Permanent Magnets (EoLPM). However, the existing literature lacks an exhaustive description of an industrial-scale process to recover Rare Earths from EoLPM.
[0006] For example, the recovery of rare earths from Neodymium-lron-Boron (NdFeB) magnets was performed at laboratory scale (Kumari et al., Hydrometallurgy, 2021, 105581) using a conventional method of oxidation roasting and acid leaching. This method was then adapted to a roasting process using water leaching and chlorination at a lower temperature, reducing the roasting temperature to 300°C. This method produced rare earth oxides with a purity of 99.2%. In addition, a leaching residue containing 96.4% FejOa was obtained as a by-product of the process. These procedures were applied to magnet blocks of rectangular dimensions (5.5x3x1.5 cm). These blocks are demagnetized in laboratory furnaces and then milled using small-scale equipment like ball mills, processes that present several problems and prove to be useless on larger scales. The treatment of large Permanent Magnet like Blocks sourced from wind turbines, often with metal encapsulation materials and epoxy coatings, remains conspicuously absent from the existing literature. The handling, separation and coating removal of large blocks of magnets, such as those used in wind turbines, are absent from published studies. The high coercivity of these blocks and the impact of different assembly methods and coatings used, pose difficulties that were not yet properly considered to ensure an efficient, automated and continuous process. In particular, the document CN115141942 discloses a method for recovering rare earth from neodymium-iron-boron waste and separating main element iron and application of the neodymium- iron-boron waste as a raw material in preparation of soft magnetic ferrite.
[0007] The document EP3791001 discloses a hydrometallurgical process which has the main aim of recovering the precious elements contained in waste permanent magnets. The method described is applicable to waste neodymium (Nd) magnets.
[0008] The document CN116479267 discloses a method for recovering rare earth elements from a waste neodymium-iron-boron magnet. The method comprises the following various steps.
[0009] Van Loy Steff et al. in Hydrometallurgy vol 191, 4 November 2019 describe a method for the recovery of rare-earth elements and Co from permanent NdFeB magnets by mechanochemically assisted ferric sulfate leaching was developed.
[0010] The document CN106498169 relates to the field of recycle of neodymium iron boron products via numerous process steps including calcination, reduction, extraction with organic solvent etc.
[0011] The document CN118103533 relates to a method for extracting at least one lanthanide element from a solid material to a method for implementing a composition comprising water, at least one organic aprotic solvent and at least one charged water-soluble auxiliary agent, and to the use of said composition for recovering lanthanide elements.
[0012] The document US2020 / 308670 relates to the recovery of rare earth elements (REEs) from a plurality of secondary sources like magnets. The process encompasses crushing and demagnetized by heating and roasting at high temperature, followed by acid leaching, followed by iron removal by precipitation and separation by filtration. The individual REE is then separated by liquid-liquid extraction. The extracted REE is then stripped out by acid. The individual rare earth element is then precipitated and dried.
[0013] The applicant therefore considers it necessary to propose a method for managing EoLPMs incorporating an efficient pre-treatment phase encompassing demagnetization, precise disassembly of Permanent Magnets and coating removal. This strategic step significantly conserves energy while effectively priming EoLPMs for the subsequent Rare Earths recovery process.
[0014] There is also a need to provide an improved process for the recovery of rare earth element.
[0015] Solution of the invention
[0016] To this purpose, the present invention relates to a process for treating permanent magnet blocks comprising at least a permanent magnet unit covered by a coating and bonding products that glue magnetic units together to form a block , comprising the steps of: Applying a thermal treatment to the permanent magnet block to eliminate magnetism and remove at least part of the organic components; vibratory separation to the units after said thermal treatment of the permanent magnet block, preferably on a vibratory conveyors, to recover said units; mechanical cleaning of said units, via for example shot blasting, preferably using a drum or tunnel shot blasting machines, or pickling or abrasive blasting, preferably in a tunnel sand blasting, to remove coating products, binding products, impurities or organic residues; ultrasonic cleaning of said units before the reduction to micronized powder;
[0017] Reducing said units to micronized powder.
[0018] The permanent magnets are understood as single unit or multiple units. A permanent magnet block preferably comprises at least two permanent magnet units bound by a binding product (i.e. a component or mix of components able to join, such as glue) to form said block. The magnet blocks are preferably magnets of at least 100 gr, preferentially equal or over 200 gr, presented as a single unit or together, forming blocks.
[0019] The thermal treatment applied to the permanent magnet block is preferably a pyrolysis treatment. The term « pyrolysis » has the meaning generally used in chemistry and it is well known in the art. Pyrolysis refers to the chemical decomposition of organic (carbon-based) materials through the application of heat in the absence or near absence of oxygen. In the inventive process, the optional pyrolysis is used to at least partially degrade the organic components, including the glue, present on the surface of the magnet units, into organic residues that can easily be removed.
[0020] The term «mechanical cleaning» refers to any method or combination of method that can be used to physically or mechanically remove solid impurities from the magnet unit. The preferred methods include drum or tunnel shot blasting or pickling or abrasive blasting, preferably in a tunnel sand blasting.
[0021] The thermal treatment is preferably run at a temperature of at least 200 °C, preferably at least 250°C, still preferably at least 300 °C or 310 °C, preferably over 400 °C, still preferably over 425°C.
[0022] In a preferred embodiment said thermal treatment is performed only once at a temperature of at least 400°C for at least 45min.
[0023] In order to make it suitable for industrial scale, the thermal treatment should not be too long. It preferably has a duration of 5 h or less, preferably 4h or less, still preferably 3h or less and ideally of 2h or less. It however has to be sufficiently long for complete elimination of magnetism and organic material and lasts preferably at least 0.5h, still preferably at least lh. Following the thermal treatment of the permanent magnet block, a vibratory separation step can be implemented to separate metallic components from the magnet units. This can be achieved using equipment such as a vibratory separator mat and, ideally, an automated sorting system employing robotic arms, image screening, and Al algorithms. The vibratory separation step is preferably performed so that 95%, more preferably 98% even more preferably 99% of the magnets blocks are separated into magnet units.
[0024] This step facilitates the separation of various materials within the permanent magnet assembly, enabling efficient sorting, preferably through fully automated processes. The sorted materials can be deposited into different containers, and the units can be automatically fed into the cleansing phase where the remaining coating layer will be removed.
[0025] Before proceeding with the reduction to powder, it is possible to apply a step of shot blasting to the units. This process ensures the complete removal of any impurities or organic residues, such as coating materials, from the surface of the units. Shot blasting can be performed using drum or tunnel shot blasting machines, or any other suitable device capable of removing this superficial layer through attrition or chemical processes, that is preferably equipped with a fire protection system to prevent any risk of fire caused by sparks from collision between the different parts of the magnet.
[0026] Additionally or alternatively, an ultrasonic cleansing step, using a ultrasonic cleaning machine can be employed to thoroughly cleanse the units of any remaining dust or impurities, before entering the grinding phase.
[0027] The permanent magnet micronized powder has an average particle size in the micrometer range, and the average particles size, as measured by direct imaging sieve analysis, but preferably by a laser diffraction analysis system. Is preferably comprised between 50pm and 500pm, preferably between 80pm and 250 pm, still preferably between 100 to 150 pm.
[0028] The particles should ideally maintain a spherical shape to the greatest extent possible. This ensures optimal performance and efficiency in their intended application.
[0029] The process of reducing the permanent magnet to powder can involve crushing the permanent magnet into smaller pieces, preferably in the cm range of, equal to or smaller than one centimetre, for example using a jaw crusher. Subsequently, grinding is performed using equipment such as attrition mills, ball mills, rod mills, or preferably a vertical roller mill. These two steps can also be conveniently combined into a single device, such as a vertical roller mill. Permanent magnets commonly contain rare earth elements, notably Neodymium (Nd), Praseodymium (Pr), Dysprosium (Dy), Samarium (Sm), and others, with Neodymium being of primary significance. The process of reducing these magnets to powder enables efficient isolation and recovery of these rare earth elements.
[0030] The micronized powder obtained by the process of the invention typically comprises between 10% and 50% of its weight, preferably between 20% and 40%, of at least one rare earth element preferably in the form of an oxide and between 50% and 90%, preferably between 60% and 80% of an iron oxide. The invention further relates to the combination of the preparation of the micronized powder detailed above with a process of selective recovery of elements that compose permanent magnets.
[0031] The invention also relates to the process itself of selective recovery of elements forming the permanent magnets.
[0032] Once the magnets have been reduced to powder, selective recovery of the REE elements can be carried out, either as a REE mixture or by isolating each element individually. Various recovery processes can be used.
[0033] The micronized powder of the invention or any other powder containing a rare earth element in combination with other elements, like for example iron, can then serve as a substrate for a bio- or hydrometallurgical recovery process of rare earths, in particular Neodymium. In this process, Iron (Fe) can also be obtained as a by-product.
[0034] These processes are detailed below.
[0035] Rare earth elements recovery, including optional recovery of iron species and acids recycling, from magnets reduced to powders.
[0036] The hydrometallurgical process for REE recovery is a comprehensive process that not only extracts REE but can also focuses on the efficient recovery of iron and the recycling of the acids used in the process, for example the oxalic acid and hydrochloric acid.
[0037] The hydrometallurgical process for rare earth element (RRE) recovery according to the invention comprises the following step of: submitting a powder comprising at least one rare earth element and iron, for example in an oxide form, to a strong acid leaching to dissolve at least the rare earth, and selectively precipitating at least one rare earth element of the list consisting in Neodymium (Nd), Praseodymium (Pr), Dysprosium (Dy) or Samarium (Sm) and wherein the concentration of said strong acid is comprised between 0.5 and 4 mol / L. In a preferred embodiment, the hydrometallurgical process for REE recovery does not encompass any step using a molten salt or a ionic liquid or any mixture thereof.
[0038] In a preferred embodiment, said strong acid is dissolved in water, preferably in water only.
[0039] In a preferred embodiment, said step of acid leaching to dissolve at least the rare earth is performed in the absence of any organic solvent.
[0040] In a preferred embodiment, the hydrometallurgical process for REE recovery does not encompass any further step of liquid-liquid extraction, preferably it does not encompass any further step of liquidliquid extraction where one of the liquid is said strong acid and the other liquid is an organic solvent.
[0041] The powder can be the powder obtained from the first aspect described above or any other powder obtained from magnets.
[0042] The selective precipitation of at least one rare earth element allows to separate this or these rare earth elements from the iron species and potentially other species comprised in the powder. Selectivity may be relative to one specific REE, i.e. one REE is precipitated, or generic to REE versus other chemical species. It is not necessarily economically relevant to isolate the various REEs from each other, as to be reused in magnet production, these will be combined again. It is then a possibility to isolate the REEs as mixtures. Usually, REEs have almost identical chemical properties, therefore difficult to separate from each other.
[0043] Preferably, the process further comprises a step of recovering the acid. One of the main concerns of the process is its environmental impact. Recovering the acid makes it possible to reduce the amount of raw material needed for the process, as well as reducing waste management costs if the acid were to be disposed of.
[0044] The hydrometallurgical process for rare earth element (RRE) recovery is preferably performed on a powder having an average particle size of 500 pm or less, preferably 450 pm of less, preferably 400 pm or less, 350 pm or less or 300 pm or less.
[0045] The leaching step of the hydrometallurgical process for rare earth element (RRE) recovery is preferably performed at a temperature comprised between 25 °C and 80 °C, preferably between 30 °C and 75 °C, preferably between 35 °C and 70 °C, preferably between 35 °C and 65 °C, still preferably between 40 °C and 60 °C.
[0046] The leaching step of the hydrometallurgical process for rare earth element (RRE) recovery preferably has a duration of below 6h, preferably below 4h, still preferably below 3h. The acid leaching is preferably performed in the presence of a strong acid, like for example hydrochloric acid, sulfuric acid and / or nitric acid. Preferably hydrochloric acid is used. The concentration of strong acid can range between 0.5 and 2 mol / L, preferably between 1 and 2 mol / L.
[0047] The term « strong acid » has the meaning generally used in chemistry and is well known in the art. A strong acid is an acid that completely dissociates into its ions when dissolved in water. Hence a strong acid releases protons (H+) in solution. Common examples include hydrochloric acid (HCI), sulfuric acid (H2SO4), and nitric acid (HNO3). The strong acid of the disclosure are preferably dissolved in water only.
[0048] The leaching step can be performed using a solid-liquid ratio between 1 and 80 g / L of powder in acid solution, preferably between 5 and 75 g / L even more preferably between 5 and 65g / L, the more preferred between 5 and 45g / L. Low concentrations allow effective leaching but represent a higher process costs per weight of powder treated, while to high concentrations result in lower leaching yields.
[0049] Optionally, the hydrometallurgical recovery process of the invention can further comprise a step of oxidizing dissolved Iron (II) if any to Iron (III), by adding an oxidizing agent like for example hydrogen peroxide and potassium permanganate, in order to improve the purity of REE precipitates. As iron(ll) can lead to the formation of FeC2O4precipitates during the precipitation step of REE, this oxidation to iron(lll) avoids such byproduct precipitation and allows a higher purity of the recovered REE.
[0050] The iron optionally, hydrometallurgical recovery process of the invention can further comprise a step of reducing dissolved Iron (III) to Iron (II), by adding metallic iron into the solution, in order to precipitate and recover the Iron. The iron is preferably recovered as a salt, for example ferrous oxalate (iron(ll) oxalate).
[0051] Preferably, the powder comprising at least a rare earth element and iron, originates from permanent magnets, preferably from large scale permanent magnets.
[0052] Said powder preferably has an average particle size comprised between 50pm and 500pm as measured by laser diffraction analysis.
[0053] Said powder preferably comprises between 10 and 50 weight% of at least one rare earth element, preferably in the form of an oxide, and between 50 and 90 weight% of iron oxide. Iron oxide here designates iron at any oxidation degree.
[0054] In the initial stage of the process, a powder comprising at least one rare earth and iron, like the micronized powder obtained from the treatment of permanent magnets of the invention, is submitted to leaching, for example by placing the powder in a solution of hydrochloric acid or nitric acid in concentrations ranging from 0.5 to 4 M (pH<l), preferably in presence of water as the only solvent. This leaching process preferably occurs at temperatures ranging between 25°C to 80°C, more preferably 30 to 80 °C, even more preferably from 25°C to 65°C, the most preferred from 25°C to 60°C. It is usually performed under agitation for a duration of up to 5 hours, preferably up to 2 to 3 hours, ideally 0,5 an hour.
[0055] The leaching process can be followed by a REE precipitation step. In this step, the dissolved REE is placed in presence of an organic acid to precipitate an organic salt of REE. This salt is for example oxalic acid. Typically, the oxalate / REE molar ratio is ranging from 1.5:1 to 20:1, preferably from 2.0:1 to 5.0:1.
[0056] The precipitation step can be conducted at temperatures ranging from 25 to 85gC, while maintaining the pH of the solution between 0.5 and 4.5. Agitation is applied throughout the precipitation process, which lasts from 0.5 to 6 hours, preferably from 1 to 6 hours.
[0057] After reprecipitation, the resulting solid can be filtered.
[0058] It can be further purified, by any method known to a person skilled in the art, like for example recrystallisation.
[0059] The oxalate REE salt can be further converted to an REE oxide by a thermal treatment, at temperatures comprised between 600 °C and 800 °C, preferably between 650 °C and 750 °C. The thermal treatment can for example be applied for 30 minutes, 1 h, 1.5h or longer. The oxides can then be reincorporated in new magnets. The recovery rate stands over 95%, with over 99% in purity.
[0060] In parallel, in an optional embodiment, iron can also be selectively recovered. The iron (III) species dissolved in the leaching solution can be placed in presence of metallic iron and an organic acid to precipitate iron (II) in the form of a salt of said organic acid. For example, when the organic acid is oxalic acid, a salt of iron (II) oxalate is precipitated according to the following reaction:
[0061] Global equation: Fe3+(aq) + Fe(s) + H2C2O4 (aq) 2Fe2+(aq)+ C2O42(aq) + H2 (g)
[0062] Intermediate equations:
[0063] The ferrioxalate complex reacts with solid iron and water to form iron(ll) oxalate dihydrate and free oxalate ions in solution: Then, metallic iron is oxidized to iron(ll) ions, while in parallel, protons (H+) can be reduced to molecular hydrogen (H2gas) if they are present in excess
[0064] Fe (s) + 2H+(aq)-> Fe2+(aq) + H2(g)
[0065] Finally, iron(ll) ions react with oxalate ions in the presence of water to form iron(ll) oxalate dihydrate, a solid precipitate:
[0066] In a preferred embodiment, in said hydrometallurgical recovery process the organic acid is oxalic acid, and a salt of iron (II) oxalate is precipitated according to one of the following equation:
[0067] Fe3++ Fe(0) + C2O4H2-> 2Fe2++ C2O42' + H2
[0068] 2[Fe(C2O4)3]3“ (aq) + Fe (s) + 6H2O (I) -> 3FeC2O4-2H2O (s) + 3C2O42- (aq)
[0069] Ultrasounds can be applied to assist the reduction step to extract the oxalate ions by forming the FeC2O4-2H2O precipitate.
[0070] The precipitate of iron (II) can be recovered by filtration.
[0071] To recover oxalic acid, the precipitate (FeC2O4.2H2O) can be dissolved in the presence of an acid, preferably hydrochloric acid. By crystallization, H2C2O4crystals can be obtained, along with a solution rich in iron as FeCI2.
[0072] After crystallization, hydrochloric acid can be separated and purified for reuse in subsequent cycles of the hydrometallurgical process.
[0073] There remains a need for an industrial-scale process to recover rare earth elements from EoLPMs that is applicable at industrial scale and fully respects the environment, requiring the development of an alternative and sustainable leaching technology. The applicant has therefore developed a new process for this purpose, which can be applied to the micronized powder obtained by the treatment process explained above or other powders obtained from EoLPMs.
[0074] The leaching of elements from EoLPM is a critical step in the recovery process of rare earth elements. Chemical and biotechnological steps can be combined, resulting in more efficient hybrid leaching strategies. Oxidation process where the recovered REE under the form of oxalates are oxidized by thermal treatment, with temperatures ranging from 600 to 800gC, forming REE oxides (REEO). The recovery rate stands over 95%, with over 99% in purity
[0075] The invention will be better understood using the following description of several implementations of the invention with illustrations on the following figures:
[0076] Figure 1: magnet (a) before and (b) after abrasive blasting;
[0077] Figure 2: Leaching results for Fe, Nd and Pr with HCI;
[0078] Figure 3: Mass degradation for the leaching using different particle sizes;
[0079] Figure 4: Leaching results for Fe, Nd and Pr with HCI;
[0080] Figure 5: Leaching results for Fe, Nd and Pr for different time points;
[0081] Figure 6: Precipitation results for Fe, Nd and Pr;
[0082] Figure 7: Precipitation results for Fe, Nd and Pr;
[0083] Figure 8: process of the invention;
[0084] Figure 9: Comparison of results obtained for different environmental indicators between the two scenarios.
[0085] PREPARATION OF PERMANENT MAGNETS FOR NEODYMIUM RECOVERY
[0086] Permanent Magnets (PM) are received in different forms, sizes, coatings, and encasement materials, and the following process is developed to handle them in a fully automated manner and on an industrial scale. The examples 1 to 4 and 6 are optional pre- and post-treatments.
[0087] In examples 1 to 4, NdFeB permanent magnets have been treated.
[0088] Example 1 - Demagnetizing / Coating Removal
[0089] 1.1. Magnets heating at 400°C for 45 minutes in a muffle furnace.
[0090] The magnet preparation process began by placing a piece of around 37.5 kg, containing multiple magnets, 60 individual magnets of approximately 45 cm x 20 cm x 12 cm, glued together and protected with stainless steel plates, at 400°C for 45 minutes in a muffle furnace (Barracha brand, model 9, power 27.5 kW), with the purpose of demagnetizing the magnets.
[0091] After the muffle furnace had cooled down, the demagnetized magnets were used in the remaining stages of the magnet preparation process.
[0092] 1.2. Separation of the material The piece containing the demagnetized magnets was then subjected to a separation process, in order to retrieve the magnet units. This was done manually for the purpose of the experiment but it is intended to use vibratory conveyors in order to ensure a fully automated process.
[0093] 1.3. Shot blasting or pickling for external coating removal.
[0094] The demagnetized magnet units underwent a sandblasting process to proceed with their pickling. After the process, the magnets were free of the outer coating and any adhesive residue.
[0095] Example 2 - Demagnetizing / Coating Removal
[0096] An alternative process was also performed.
[0097] 2.1. Magnets heating at 450°C for 4 hours in Pyrolysis furnace.
[0098] The magnet preparation process began by placing a piece of around 37.5 kg, containing multiple magnets, 60 individual magnets of approximately 45 cm x 20 cm x 12 cm, glued together and protected with stainless steel plates, at 450°C for 4 hours in a furnace Model 05163 manufactured by Termometalicas (custom made, devoid of magnetic materials, to allow the safe handling of the magnets. Motorized feeding cart with appropriate containers, that are fed automatically, and able to handle small magnets as well as large blocks and a maximum charge of 1500 kgs. Propane or natural gas burner with an output of 200,000 kcal / h; Maximum temp: 1100°C; Maximum consumption 40Nm3 / h (on all 3 burners) )
[0099] Following this treatment, the magnets have lost a significant part of the outer coating and any adhesive residue (epoxy and others).
[0100] 2.2. Separation of the material
[0101] The piece containing the demagnetized magnets was then subjected to a separation process, in order to retrieve the magnet units. This was done manually for the purpose of the experiment but it is intended to use vibratory conveyors in order to ensure a fully automated process.
[0102] 2.3. External coating full removal
[0103] The Magnets (in bulk) were subjected to abrasive blasting (abrasive: spherical steel, hardness 46 51 HRC, dimensions between S 230 and S 390, equipment DRACO Model SG2 RNP FAC can be used). The result is illustrated on figure 1.
[0104] An ultrasonic cleaning machine can also be used to remove the remaining superficial impurities. This is a fully automated process and there is no manual manipulation of the magnets.
[0105] A Tunnel Sand Blasting Machine can also be used in order to ensure a continuous flow of the material, without human intervention. Example 3 - Grinding of the magnet
[0106] 3.1. Load application to obtain small magnet pieces
[0107] Separated and demagnetized magnet units, as obtained in example 1 or example 2 underwent a load application process in order to break the magnet into small pieces for subsequent grinding.
[0108] Initially, the magnet units of about 200g or more were placed inside a mold, and then a load was applied until the magnet fractured, using a Universal Press (with a maximum capacity of 1600 kN). At the end of the load application, the magnets were reduced to small pieces.
[0109] 3.2. Grinding or milling to less than 210 pm.
[0110] The grinding of the magnet pieces was initiated using a mill, for about lh per kg. The powder obtained has a size less than 210 pm (±149 pm), as analyzed by laser diffraction.
[0111] Example 4 - Grinding of the magnet
[0112] Industrial equipment from Loesche was used to performed grinding. This equipment provides several units online, which the magnet units go through.
[0113] First, magnet units enter a crusher (units preferably have a size of below 100 mm), where they are crushed into pieces having a size of two cm or less, preferably one cm or less These pieces are conveyed to a mill where they are micronized to a powder having a average grain size of between 50 and 200 pm. The powder is then discharged into containers or big bags.
[0114] The resulting magnet powder was subject to roasting for 2 hours as 900gC (Nabertherm furnace - maximum temperature of 11009C), to oxidate and allow an easier treatment at later stages.
[0115] Example 5 - Rare earth elements recovery, including optional recovery of iron species and acids from magnets reduced to powders
[0116] 5.1. Materials
[0117] Powder Magnet (magnet III) obtained from example 3 was used in this experiment. Nitric Acid (70 %, Fisher) and hydrochloric acid (37 %, Fisher) were used for acid digestion and leaching experiments. The multi-element ICP quality control standard solution for REE, with a concentration of each element of 200 mg / L, was purchased from CPAchem. The multi-element ICP quality control standard solution for Fe, with a concentration of 1000 pg / mL, was purchased from PerkinElmer.
[0118] 5.2. Powder characterization The crushed magnet was characterized accordingly to the particle size and elemental analysis.
[0119] The sieving process was performed using 580 g of magnets, which were separated into different sizes by different sieves (450, 300, 180 and 100 pm).
[0120] Elemental analysis of the starting powder magnet was carried out by microwave assisted acid digestion (Speedwave Four, Berghof) and subsequent determination of element concentrations present in the sample solutions was conducted by inductively coupled plasma-optical emission spectroscopy - ICP-OES (Optima 8000, PerkinElmer).
[0121] 25 mg of magnet was digested using 10 mL of HCI. The acid digestion was performed according under the operation condition presented in Table 1. The solutions obtained in the leaching were filtered in filter paper and brought to a constant volume with Milli-Q® water. Prior to ICP analysis, each sample was filtered through a pore size of 0.22 pm.
[0122] Table 1: Microwave assisted acid digestion operation methodology
[0123] 5.3. Characterization of REE oxides
[0124] TGA analysis was performed at CeNTI - Centre for Nanotechnology and Advanced Materials (Portugal). The analysis was carried out in a synthetic air atmosphere from 30 to 1000 °C with a heating of 20 °C / min. The sample was set to a room temperature (= 21 °C) and relative humidity (= 41%).
[0125] 5.4. Analytical quantification
[0126] All liquid samples were analyzed at the Inductively Coupled Plasma - Optical Emission Spectrometry, ICP-OES, (Optima 8000, PerkinElmer). The operating conditions were, namely, RF power at 1350 W, argon plasma flow at 10 L / min, auxiliary gas flow at 0.2 L / min, and nebulizer gas flow at 0.60 L / min.
[0127] The main quantified elements were Fe, Nd and Pr with the wavelengths of 238.204, 401.225 and 390.844, respectively. Other elements were also quantified during the characterization process, the wavelengths used were Tb - 350.917, Dy - 353.17, B - 249.772, Al - 396.153, Ni - 221.648, Cu - 327.393, Mn - 257.61 and Cr - 267.716.
[0128] 5.5. Statistical analysis
[0129] The results obtained from the leaching quantification of each metal were analyzed using the Two-Way
[0130] ANOVA. The Bonferroni's multiple comparison test was used for both data sets. The ANOVA analyses were performed using the software Graph Pad Prism version 8.0.2 (Graph Pad Software, Inc, San Diego, CA, USA). The results were only considered significantly different when the probability (p-value) was lower than 0.05, assuming a 95% confidence interval.
[0131] 5.6. Magnet powder characterization
[0132] The sieving process was performed using 580 g of magnets, which were separated into different sizes by different sieves (450, 300, 180 and 100 pm).
[0133] The distribution of the magnet mass after the sieving are shown in Table .
[0134] Table 2: Magnet size distribution after the sieving process
[0135] The results show that most of the powder magnet has a particle size below 300 pm.
[0136] This can be important for the leaching process, as the acid have a higher surface area to attack, which lead to a better leaching. However, smaller particles can make the leaching process harder by forming sludges. In this work, the subsequent leaching assays were carried out using a particle size below 300 pm.
[0137] The results of the elemental analysis are shown in Table 3:
[0138] Table 3; Results of the elemental analysis of the magnet.
[0139] The main element present in the magnet is the Fe with 72.34 %, followed by Nd and Pr with 20.73 and 4.67 % respectively, as expected. Also, other REE might be found present in magnets, as Tb and Dy . however, only Dy was found as constituent of this magnet. As the main constituents of the magnet are Fe, Nd and Pr, the following analysis will be focused for their concentrations.
[0140] 5.7. Leaching experiments
[0141] The experimental section was caried out using a water bath (Julabo SW22) between 25 °C and 65 °C, and HCI or HNO3 concentration of between 0.5 M and 2 M and a solid to liquid ratio of 5 g / L to 65 g / L. Adding to those conditions, the assays were carried out for 6 h with an agitation of 150 rpm and 100 mL of a solution of hydrochloric (HCI) acid were used. Liquid samples were taken from the solution after 6 h. Each sample was filtrated by a paper filter and a syringe nylon filter 0.22 pm before being quantified in Inductively Coupled Plasma - Optical Emission Spectrometry, ICP-OES.
[0142] The leaching yield of elements was calculated from equation 1. Equation 1
[0143] Where mieachate (mg of element) is the element from the NdFeB mass found in the leachate, mtotai (mg of element) is the total mass of the element that is present in that sample.
[0144] The mtotai for a given element was calculated from equation 2. Equation 2
[0145] Where mmagnet (g) is the magnet mass used in the given assay, Cdigestion (mg of element / g of magnet) is the mass of the element per g of magnet that was quantified by microwave digestion.
[0146] 5.7.1. A first leaching assay was done using the S / L ratio of 65 g / L, a temperature of 65gC, an HCL concentration of 2 M and a particle size below 300 pm and the influence of the time were studied. The results are shown in Figure 2.
[0147] With these conditions it was only achieved a mass degradation of 64.63 ± 1.16 % with possible metal precipitation. The results for each metal leaching are very similar between each other for the different time, with only some statistical differences found for Pr. These results suggest that the leaching process occur mainly in the first 2 h of the assay. The possible precipitation is supported by the low mass degradation and with the observation of the filter paper that show some possible metal precipitates.
[0148] 5.7.2. A second leaching assay was done using the S / L ratio of 50 g / L, a temperature of 65 °C, an HCI concentration of 2 M. The particle size and the influence of the time were studied. The results are shown in Figures 3 and 4.
[0149] Overall, it can be seen that it was possible to achieve a near total magnet leaching from the mass degradation results, for all particle size ranges, apart from the particle size over 450 um. The only statistical difference was found when comparing the particle size over 450 um with the others. This result suggests that a particle size over 450 pm difficult the leaching process, as there is a smaller magnet surface for the HCI react. 5.7.3. A third leaching assay was done using the S / L ratio of 50 g / L, a temperature of 65gC, an HCI concentration of 2 M and a particle size between 180 and 300 pm, and the influence of the time were studied. The results are shown in Figure 5.
[0150] For this assay, it was obtained a mass degradation of 98.99 ± 0.05 %, supporting that it was possible to achieve a near total magnet leaching, supporting the results obtained from Figure 4.
[0151] The cost associated for the leaching of 1 kg of magnet using temperature of 65 C°, acid concentration of 2 M, S / L ratio of 50 g / L for 0.5h is 1.621 or 1.774 € with or without considering the heating, respectively. The leaching of 1 kg of Nd considering the leaching efficiency from Figure 24, would be 4.981 kg of magnet and the cost associated would be 8.075 or 8.838 € with or without considering the heating, respectively.
[0152] 5.8. Precipitation Experiments
[0153] The precipitation assays were carried out using a leaching solution (500 mL) with a concentration of 26915.23 ± 1772.93 mg / L, 8202.55 ± 319.84 mg / l and 1706.8 ± 45.07 mg / L for Fe, Nd and Pr respectively.
[0154] This solution was stirred with an agitation of 150 rpm at 44 °C for 4 h, in presence of oxalate at a molar oxalate / REE ratio of 3.9. The precipitation results are illustrated in Figure 6.
[0155] The precipitation yield of elements was calculated from equation 3.
[0156] Y=mi ni t i ~ms oi ut t c1„ „ Equation 3 ipreci pi t at i ± U U
[0157] "‘i ni t i
[0158] Where m solution is the element mass that is present in solution and m initial is the total mass of the element initially quantified in the solution.
[0159] A second precipitation experiment was carried out using a leaching solution (3 L) with a concentration of 30077 ± 306 mg / L, 12863 ± 57 mg / L, 3529 ± 39 mg / L, 583 ± 3 mg / L for Fe, Nd, Pr and Dy respectively. This solution was stirred with an agitation of 500 rpm at 25 °C for 30 min, in presence of oxalate at a molar oxalate / REE ratio of 17 The precipitation results are illustrated in Figure 7.
[0160] The second experiment demonstrates that it is possible to recover more than 98% of the rare earth elements without iron precipitation. The precipitates consist of a mixture of Nd, Pr, and Dy. The fact that the recovered materials are in mixed form is just as valuable as if they were pure, since the mixture can be directly reused in the same proportions for manufacturing new magnets. Example 6 - REE Oxide preparation
[0161] The rare earth element (REE) oxides were obtained by calcining the oxalate precipitates from Example 5 at 700 °C for 2.5 hours, using a heating rate of 7.5 °C / min and a cooling rate of 1.7 °C / min. The thermal behavior and conversion were confirmed by thermogravimetric analysis (TGA).
[0162] Example 7 - Life Cycle Assessment
[0163] On the basis of the experimental data obtained in the previous examples, a life cycle assessment (LCA) was conducted on the proposed solution for the recovery of rare earth elements from Neodymium magnets (FeNdB).
[0164] The objective of this life cycle assessment is to evaluate the environmental performance of recovering Rare Earth Elements (REEs), such as neodymium and praseodymium, through the reuse of NdFeB magnets from recycled materials. The methodology adopted is based on the stages of a Life Cycle Assessment (LCA) using the multicriteria decision method MARS-SC (Methodology for Relative Sustainability Assessment of Constructive Technologies) [E. R. Teixeira, R. Mateus, A. F. Camoes, L. Bragan^a, and F. G. Branco, "Comparative environmental life-cycle analysis of concretes using biomass and coal fly ashes as partial cement replacement material," J. Clean. Prod., vol. 112, pp. 2221-2230, 2016, doi: https: / / doi.org / 10.1016 / jjclepro.2015.09.124.], [R. Mateus, S. Neiva, L. Bragan^a, P. Mendonca, and M. Macieira, "Sustainability assessment of an innovative lightweight building technology for partition walls - Comparison with conventional technologies," Build. Environ., vol. 67, pp. 147-159, 2013, doi: https: / / doi.Org / 10.1016 / i.buildenv.2013.05.012.1.
[0165] A cradle-to-gate analysis was conducted to assess the production of REEs (a market product) using both the conventional method of mineral extraction and the method of the invention, which is intended for commercial application. For the mineral extraction scenario, the analysis included the impacts generated from the production of raw materials, their transportation to the REE production facility, and the REE production process itself. For method of the invention, the steps considered are detailed below. These boundaries were selected because REEs have the same commercial value, whether sourced from the traditional method (used as the baseline in this study) or recovered from other materials. Therefore, the functional unit for this study was the amount of REE recovered per kilogram of magnet, equivalent to 0.226 kg of Neodymium Oxide (NdjOa) and 0.047 kg of Praseodymium Oxide (PrjOa).
[0166] 7.1. Inventory analysis In a life cycle assessment, the inventory covers the quantification of input parameters such as energy consumed, materials, and chemicals used, as well as output parameters resulting from each process executed in all stages of the life cycle, including emissions and generated waste. For this analysis, SimaPro 9.5.0.3 platform was used to quantify impact categories, drawing on data from the Ecoinvent V3 database for processes and data where information was not available. This database provides comprehensive information on various materials and processes across different regional contexts. The MARS-SC methodology is based on three sustainability categories: environmental, functional, and economic. However, this study focuses solely on the environmental category.
[0167] Two scenarios were considered in the assessment:
[0168] A. The process of the invention was analyzed based on processes identified in the flowchart presented in Figure 8, along with detailed materials and quantities outlined in Table 4. The impacts of magnet manufacturing processes were not considered, as they are treated as waste. According to the MARS-SC methodology, impacts from material production, when treated as waste, are not accounted for [L. Bragan^a and R. Mateus, "Sustainability assessment of building refurbishing operations," Port. SB07 Sustain. Constr. Mater. Pract. Chall. Ind. New Millenn., p. 381, 2007.]. The stages involved include acid leaching, filtration, iron oxidation, REE precipitation, filtration, calcination, ultrasonic reduction, acid dissolution, and crystallization (phases one and two). The inputs and outputs of each phase were documented in a report based on laboratory work conducted, summarized in the flow diagram of fig 8, which was also provided to perform the respective calculations.
[0169] To summarize the flowchart of figure 8, 1kg of a NdFeB magnet is submitted to Acid Leaching at 65 _C in step A. to this purpose, 2.53 L HCI (37%) , 12.87 L water and 2 kWh were involved (to form a 2 mol / L HCI solution, for a solid to liquid ration of about 65 g / L.
[0170] The Fe and REE rich leachate was filtered in step B to removed 0.03 kg undigested magnet.
[0171] In step C, 1.26L hydrogen peroxide (30 v / v) was added to the leachate to convert iron(ll) to iron(lll).
[0172] In step D, 0.629 kg of oxalic acid is added to precipitate the RRE at room temperature. The REE oxalate precipitate (0.743 kg Ndj CjC h and 0.154 kg Prz CzCUh ) is filtered out in step E, along with 0.009 kg impurities and further subjected to a calcination step F at 600 °C for 4h (using 2 kWh) to provide 99% pure REE oxides (0.226 kg of NdjOa, 0.047 kgPrjOa), generating 0.563 kg of gaseous residues.
[0173] In step G, the iron rich filtrate is reduced in presence of ultrasound and addition of 144 g of metallic iron to precipitate iron oxalate (FeCjCU), which is redissolved in step H by addition of 2M HCI for a solid to liquid ratio of about 200 g / L, leading to a solution containing FeCL and oxalic acid. Step I is a crystallization step which allows to recover 80% of oxalic acid as crystals. The remaining solution comprising iron chloride can undergo a crystallization step J to isolate solid iron chloride which are separated from the HCI supernatant (25.18 kg).
[0174] This acid supernatant can be reused in step H, provided concentration adjustment in step K. The oxalic acid recovered from step I can be reused in step D.
[0175] * The gaseous residue was not considered in the impact analysis, as its composition was not identified, thus preventing an exact definition of the impacts generated. This could introduce variability in the impact results. ** Reuse of materials obtained after the initial recovery process of NdFeB magnets.
[0176] B. Mineral Extraction: Traditional Production of Neodymium and Praseodymium:
[0177] For the traditional production process of Nd and Pr, the processes listed in the Ecoinvent database were used, along with their respective location from the point of sale to Portugal. The nearest point of sale to the destination location was found in Germany, with the supplying company Labkontor GmbH [https: / / chemondis.com / search / ?query=neodimio].
[0178] 7.2. Impact Assessment Data from the life cycle inventory are converted into potential environmental impacts through the application of Life Cycle Impact Assessment (LCIA) methods. The CML-IA baseline method (version 3.09) was used to assess environmental indicators, expressed in specific categories. Simultaneously, the Cumulative Energy Demand (CED) method (version 1.11) was employed to evaluate energy- related inputs. Within the methodology, environmental performance is evaluated based on the following environmental impact indicators: Global Warming Potential, Ozone Depletion Potential, Acidification of Soil and Water, Eutrophication Potential, Formation Potential of Tropospheric Ozone, and Depletion of Abiotic Resources of Fossil Fuels (as presented in the following table 5).
[0179] Table s
[0180] These results are illustrated in figure 9.
Claims
Claims1. Hydrometallurgical process for REE recovery comprising the following steps of: submitting a powder comprising at least one rare earth element and iron to a strong acid leaching to dissolve at least the rare earth, and selectively precipitating at least one rare earth element of the list consisting in Neodymium (Nd), Praseodymium (Pr), Dysprosium (Dy) or Samarium (Sm) and wherein the concentration of said strong acid is comprised between 0.5 and 4 mol / L.
2. Hydrometallurgical recovery process according to claim 1, further comprising recovering the acid.
3. Hydrometallurgical recovery process according to claim 1 or 2, wherein the powder comprises more than one rare earth element, and the selective precipitation yields a mixture a rare earth elements.
4. Hydrometallurgical recovery process according to one of claim 1 to 3, wherein the powder has an average particle size of 500 pm or less, as measured by laser diffraction analysis.
5. Hydrometallurgical recovery process according to one of claim 1 to 4, wherein the leaching step is performed at a temperature comprised between 30 °C and 80 °C.
6. Hydrometallurgical recovery process according to one of claim 1 to 5, wherein the leaching step has a duration of up to 5 h.
7. Hydrometallurgical recovery process according to one of claim 1 to 6, wherein the strong acid leaching is hydrochloric acid, sulphuric acid or nitric acid or any mixture thereof.
8. Hydrometallurgical recovery process according to one of claim 1 to 7, wherein the leaching step is performed using between 1 and 80 g / L of powder in acid solution.
9. Hydrometallurgical recovery process according to claim 1 to 8, wherein the powder comprising at least a rare earth element and iron originates from permanent magnets, preferably from large scale permanent magnets.
10. Hydrometallurgical recovery process according to one of claim 1 to 9, wherein the powder comprises between 10 and 50 weight% of at least one rare earth element, and between 50 and 90 weight% of an iron oxide.
11. Hydrometallurgical recovery process according to claim 10, further comprising oxidizing dissolved iron (I) oxide and / or iron (II) oxide , if any, to iron (III) oxide.
12. Hydrometallurgical recovery process according to claim 11, wherein the iron oxidation step involves the addition of a peroxide.
13. Hydrometallurgical recovery process according to any of claim 1 to 12, further comprising a reduction step wherein the dissolved REE is placed in presence of an organic acid to precipitate an organic salt of REE.
14. Hydrometallurgical recovery process according to claim 13, wherein the organic acid is oxalic acid.
15. Hydrometallurgical recovery process according to one of claim 13 to 14, further comprising the step of placing iron (III) in presence of an organic acid and metallic iron to precipitate iron (II) in the form of a salt of said organic acid.
16. Hydrometallurgical recovery process according to claim 15, wherein the organic acid is oxalic acid, and a salt of iron (II) oxalate is precipitated according to the following reaction: 2[Fe(C2O4)3]3’ (aq) + Fe (s) + 6H2O (I) -> 3FeC2O4-2H2O (s) + 3C2O42’ (aq)17. Hydrometallurgical recovery process according to one of claim 13 to 16, wherein the reduction step is assisted by ultrasounds.
18. Hydrometallurgical recovery process according to claim 13 to 17, further comprising: Recovering the precipitate of organic salt of iron (II) and / or REE.
19. Hydrometallurgical recovery process according to claim 13 to 18, wherein the the oxalate / REE molar ratio is comprised between 1.5:1 and 20:1, preferably from 2.0:1 to 5.0:
1.
20. Hydrometallurgical recovery process according to claim 13 to 19, wherein the reduction step is conducted at temperatures ranging from 25 to 85gC, while maintaining the pH of the solution between 0.5 and 4.5.
21. Hydrometallurgical recovery process according to claim 18, wherein the recovered REE under the form of an organic salt are oxidized by thermal treatment, with temperatures ranging from600 to 8009C, forming REE oxides (REEO).
Citation Information
Patent Citations
Method for recovering rare earth from neodymium iron boron waste and separating main element iron and application of method in preparation of soft magnetic ferrite as raw material
CN115141942A
Hydrometallurgical method for the treatment of permanent magnets
EP3791001A1
Neodymium iron boron scrap recovery process
CN106498169A
Methods for recovering rare earth elements and separating the main element iron from neodymium iron boron waste and its application as a raw material in the preparation of soft magnetic ferrites.
CN115141942B
Method for recovering rare earth elements from waste neodymium-iron-boron magnet
CN116479267A