REPOSITORY > RESULTS

Abstract

Currently, sintered neodymium-iron-boron (Nd–Fe–B) permanent magnets (PMs) have a clear domination over other rare-earth (RE) PMs, such as samarium-cobalt (Sm–Co) by means of both value and weight shares. Sintered Nd–Fe–B PM consumption has been estimated to be driven by hybrid and electric vehicles and wind turbines in the next 20 years, though the use of these magnets in other applications will support further demand growth. Generally, the sintered Nd–Fe–B PMs incorporate 20–30 wt.% of rare-earth elements (REEs) that include mainly Nd with a small addition of Dy and/or Tb, and as such represent an important secondary REEs resource. Despite that, less than 1% of REEs are being recycled from End-of-Life (EoL) products, with REE-containing PMs representing the largest share of these products. This is a very small percentage, especially if we consider that the recycling of the sintered Nd–Fe–B magnets is an important strategy for reducing the environmental dangers associated with the RE mining and overcoming the well-documented supply risks associated with the REEs.
With the aim of recycling the sintered Nd–Fe–B magnets, various approaches, such as direct re-use, alloys recovery and REEs recovery are currently at different technology-readiness levels. A direct re-use of the Nd–Fe–B magnets is considered as the most economical and ecological way. However, extra addition of REE materials, e.g., Nd hydride, is required due to the loss in the liquid phase and the non-separation of REE oxide (REO) phases. As a result, sintered Nd–Fe–B magnets produced from the repeated recycling of materials by direct re-use methods tend to have poorer magnetic properties as the number of cycles increases. Alloys recovery, such as melt-spinning and liquid-metal extraction, is able to recover both RE-transition metal (RE–TM) alloys and TM-free RE metals that are ready for new magnets making. However, the processes generally operate at 750–950 °C with long durations that are energy-intensive. REEs recovery, such as gas-slag extraction and hydrometallurgy, is applicable to both slightly and heavily oxidized magnet material with varied compositions. Especially, hydrometallurgy is seen as the most promising, as it requires relatively simple equipment to extract the REEs in a continuous operation at mild temperatures. However, this multi-step process consumes large quantities of acids and alkalis as well as large quantities of wastewater. Still, further research is needed to make the recycling processes for Nd–Fe–B magnets more economical, sustainable and environment-friendly.
Sintered Nd–Fe–B PMs consist of REE-rich grain boundaries, representing about 10–12% of the magnet, and the Nd2Fe14B grains of the matrix phase, which is practically oxygen-free, accounting for 85–87% of the magnet. An electrochemical process was developed to recover the Nd2Fe14B grains from the sintered Nd–Fe–B magnets. The procedure is based on the anodic etching of the sintered Nd–Fe–B magnets in a non-aqueous dimethylformamide (DMF)-0.3 mol L–1 FeCl2 bath. Selective recovery of Nd2Fe14B grains was realized within the applied current density < 5 mA cm–2 based on the etching priority of phases: metallic Nd > intergranular NdFe4B4 > matrix Nd2Fe14B. The total energy consumption of the proposed recycling route is estimated to be ~2.99 kWh kg–1, which is comparable to the state-of-the-art direct re-use methods. A recycling route for the sintered Nd–Fe–B magnets is proposed based on the electrochemical etching that ends up with the Nd2Fe14B grains, REE-containing electrolyte and REE-based particles, and pure Fe metal as the final products with the only consumption of FeCl2 and electricity. The proposed recycling route is currently the only procedure that enables repeated recycling of sintered Nd–Fe–B magnets in a closed-loop system.
The REE-containing electrolyte with REE chlorides and FeCl2 dissolved can be further treated by the conventional hydrometallurgical process for high purity of the REEs recovery. Alternatively, these chlorides can also be used as metal precursors for RE–TM alloys synthesis by electrodeposition which is promising in producing thin film magnets.
Electrodeposition of Nd–Fe from the 1-ethyl-3-methylimidazolium dicyanamide ([EMIM][DCA]) ionic liquid (IL) using metal chlorides was investigated. We observed that Nd3+ cannot be electro-reduced independently, although it can be co-electrodeposited inductively on a Cu substrate with the presence of Fe2+. The transmission electron microscopy (TEM) analysis combined with electron-energy-loss spectroscopy (EELS) verified the formation of NdC2 as an indirect proof that Nd3+ is reduced to Nd0 during the electrodeposition process. The TEM/EELS was also able to confirm that the deposition of the Nd–Fe starts with the sole deposition of Fe, followed by the co-deposition of Nd–Fe. This is in agreement with the transition-state theory, which has the Fe initially reduced to an activated state (Fe*), where it is able to catalyse the reduction of the RE3+ to RE0. This new insight into the electrodeposition process brings us a very important step closer to being able to recycle REEs efficiently and even to realize electrodeposited REE-based PM thin films at a mild temperature, thus giving us a sustainable, green-chemistry approach that provides a genuine alternative to high-temperature molten salt electrolysis.
To reduce the environmental impact of REEs recovery, a facile closed-loop process to selectively separate and recover the REEs and TMs from the sintered Nd–Fe–B magnets was demonstrated. First, using a room-temperature electrolysis step, the sintered Nd–Fe–B magnet was completely leached on the anode, while the Fe metal was selectively deposited on the cathode. By balancing the dissolution and consumption of the Fe, selective leaching of the REEs was realized, which saved 84.4% in terms of acid consumption and did not require the use of any alkali for the Fe precipitation. The REEs were selectively precipitated as REE-sodium sulfate double salts, which meant we have been able to recycle the electrolyte in a closed-loop, avoiding any wastewater discharge that has to be done using conventional hydrometallurgy. This two-step method was able to recover 92.5% of the REEs with a purity of 99.4%.