|
|
| Fig. 1: Temperature dependence of (BH)max for established permanent magnet (PM) materials. (Image source: J. Ye, after Rom et al. [2]) |
As the world demand for electric vehicles (EVs) continues to grow, it has become increasingly important to refine the production of the core of EVs the electric motor that drives the vehicle. Considering how electric motors work, incorporating powerful yet compact permanent magnets becomes one of the most prioritized goals. Neodymium iron boride (Nd-Fe-B) permanent magnets are therefore determined to be the most optimal choice by the vast majority of EV manufacturers thanks to their high magnetic energy density (BH)max. [1] However, the production of Nd-Fe-B PMs requires the addition of rare earth element neodymium. Since rare earth resources are highly concentrated geographically, it is no surprise that Nd-Fe-B permanent magnets are only mass produced in selected countries that have abundant rare earth reserves and then exported to the rest of the world. In particular, China has a share of 92% in global magnet production in 2020. [1] Considering the complexity of world economics and politics, it is worth exploring the trade-offs of using non-rare-earth permanent magnets in terms of performance and price.
To compare the performance of rare-earth permanent magnets and typical non-rare-earth permanent magnets, we start with their respective magnetic energy density (BH) max. It can be loosely defined as the strength of the magnet per unit volume, and is inversely proportional to the volume of the magnet given the total magnetic strength. It has a unit of MGOe, which stands for mega-gauss-oersted. It is a widely used measure of the performance of a magnet. The detailed definition of (BH)max and its unit will not be discussed here, as it is not needed to understand the comparisons that follow.
From Fig. 1. it can be seen that the (BH)max of Nd-Fe-B PM is around 10-47 MGOe, depending on temperature, while the (BH)max of Sr-Fe-O (a widely used ferrite-based non-rare-earth magnet) is only around 3-5 MGOe. [2] Note that samarium-cobalt (Sm-Co) permanent magnets in the figure also comprise of rare earth, and is more expensive than Nd-Fe-B permanent magnets, thus making them only preferable in niche high-temperature applications. Since (BH)max is inversely proportional to the magnetic strength, it can be concluded that to produce a non-rare-earth Sr-Fe-O permanent magnets that has comparable performance to a Nd-Fe-B PM, the non-rare-earth permanent magnet needs to have roughly 10 times the volume of a rare-earth permanent magnet. Considering the densities of Nd-Fe-B and Sr-Fe-O PMs, ρ Nd-Fe-B = ~7.57 kg/m3 and ρSr-Fe-O = ~5.1 kg/m3, we can calculate that the mass of the non-rare-earth permanent magnet is ~6.74 times the mass of the corresponding rare-earth permanent magnet, assuming that they have the same magnetic performance. [3,4]
We now look at the cost comparison of producing an Nd-Fe-B permanent magnet for an EV motor and a non-rare-earth Sr-Fe-O permanent magnet of equivalent performance. The European Raw Materials Alliance (ERMA) estimates that at least 1.5 kg of Nd-Fe-B PM is needed per EV motor. [5] According to a Department of Energy report, Nd-Fe-B permanent magnets contain 30% rare earth by weight. [1] Therefore, we can calculate that there are 1.5 kg × 0.3 = 0.45 kg of neodymium in a regular EV motor. In the 2025 USGS Mineral Commodity Summaries, neodymium oxide has a price of $56/kg in the year 2024, making the basic rare earth feedstock price of an Nd-Fe-B permanent magnet to be 0.45 kg × $56/kg = $25.2. [6] Note that this is the floor price of neodymium in an EV motor excluding any processing, transportation, and labor fee. Also note that ERMA also estimates the per kilogram price of Nd-Fe-B permanent magnet to be €50 ($57.46) by the year 2030. [5] However, this number is largely a rough estimation, and with the lack of other reliable government-reported estimated/actual price of Sr-Fe-O permanent magnet, this number serves little comparison use.
The production of Sr-Fe-O involves iron oxide pigments and celestite. According to the USGS report, iron oxide pigments cost $2 per kilogram and the import price at port of exportation for celestite is $0.39 per kilogram. [6] Considering that the price of neodymium that was used in the previous calculation did not include tariff either, this is a fair comparison. Assuming the required weight ratio of iron oxide pigment and celestite is 87:13, the basic feedstock price per kilogram of Sr-Fe-O is roughly $1.8. Now, we multiply 1.5 kg by 6.74 to get the necessary magnet weight (10.11 kg), and then use this to estimate the floor price without any additional fee, which equals roughly $18.2 for one EV motor. This basic feedstock price is much lower than the Nd-Fe-B permanent magnet even when the calculated price of Nd-Fe-B permanent magnet has not included the 69% iron and the 1% boron.
It is both interesting and necessary to look at the cost breakdown of an electric motor to determine the effects of fluctuating magnet prices. According to a report published by Oak Ridge National Laboratory for the U.S. Department of Energy, the magnet cost of an electric motor roughly equals 19% - 23% of the total manufacturing cost. [7] The 2004 report estimated the magnet cost to be $32/lb ($70.55/kg) using "the most expensive class of magnets" then, which was sintered neodymium magnets. The report also noted that even when using the same raw materials, the sintered neodymium magnets were still 30-70% more expensive than the most commonly used compression-molded neodymium magnets. Therefore, if compression-molded neodymium magnets were used in this calculation, the fraction of the cost due to magnets can be as low as ~6.6%.
This answers why the cost estimation done in the previous section seems low. These estimations only take into account raw materials cost. However, from the Oak Ridge report, it can be easily inferred that the vast majority of the cost is spent on manufacturing processes, both for the magnet itself and for the other parts of an electric motor, such as the rotor and the stator. Thus, even though the raw material cost of neodymium is estimated to be $25.2 for a modern electric motor that has 1.5kg of Nd-Fe-B permanent magnet, the actual manufacturing cost may be much higher depending on the manufacturing process and other important parts of an electric motor.
Rare-earth magnets such as Nd-Fe-B enable compact, high-efficiency electric-vehicle motors because of their high magnetic energy density and strong resistance to demagnetization. Replacing them with ferrite magnets reduces material cost but requires much larger magnet volume and weight to achieve similar performance, leading to heavier and less efficient motors. Although ferrites eliminate supply-chain risks tied to rare-earth elements, they cannot yet match the power density demanded in modern EV drives. Future progress will rely more on recycling, material optimization, and diversification of rare-earth supply rather than full substitution by non-rare-earth magnets.
© Jason Ye. The author warrants that the work is the author's own and that Stanford University provided no input other than typesetting and referencing guidelines. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.
[1] B. J. Smith et al., "Rare Earth Permanent Magnets," U.S. Department of Energy, February 2022.
[2] C. L. Rom et al., "Emerging Magnetic Materials For Electric Vehicle Drive Motors," MRS Bull. 49, 738 (2024).
[3] A. Ikram et al., "Particle Size Dependent Sinterability and Magnetic Properties of Recycled HDDR Nd-Fe-B Powders Consolidated With Spark Plasma Sintering," J. Rare Earths 18, 90 (2020).
[4] J. C. Guzmán-Mínguez et al., "Greener Processing of SrFe12O19 Ceramic Permanent Magnets by Two-Step Sintering," Ceram. Int. 42, 33765 (2021).
[5] R. Gauß et al., "Rare Earth Magnets and Motors: A European Call for Action," European Raw Materials Alliance, 2021.
[6] "Mineral Commodity Summaries 2025," U.S. Geological Survey, March 2025.
[7] R. H. Staunton et al., "PM Motor Parametric Design Analyses for a Hybrid Electric Vehicle Traction Drive Application - Interim Report," Oak Ridge National Laboratory, ORNL/TM-2004/120, July 2004.