What Magnetic Materials Are Commonly Used in New Energy Vehicle Motors? A Complete Analysis of Drive Motor Magnet Applications

The electric vehicle powertrain is converging on a handful of magnetic material families, yet the selection landscape is shifting faster than most design teams realize. Rare-earth supply volatility, rising power-density targets, and cost-down pressure from OEMs are reshaping which materials win in production programs versus which merely look good on a datasheet. This article maps the current state and near-term trajectory so that design engineers and sourcing teams can make informed, defensible decisions for the next platform cycle.

The Core Decision Facing EV Motor Engineers Is Not Simply "Rare Earth or Not"

The real question is a multi-axis trade-off: how much torque density can you sacrifice (or must you preserve) given your thermal envelope, duty cycle, cost ceiling, and supply-chain risk appetite? A city-car traction motor at 50 kW peak and a performance sedan at 250 kW peak may land on entirely different material strategies even within the same OEM group.

Today's production landscape includes:

• Sintered NdFeB (with varying heavy-rare-earth content) — still the default in high-performance traction motors. Dysprosium and terbium cost, plus geographic concentration of supply, continue to drive reformulation rather than wholesale abandonment.
• Grain-boundary-diffused (GBD) NdFeB — places heavy rare earth mainly at grain boundaries, so bulk Dy/Tb content can fall while high-temperature coercivity is largely retained. This is now a common specification path for new programs, not a specialty option.
• Ferrite (SrFe₁₂O₁₉ / BaFe₁₂O₁₉) — re-emerging in cost-sensitive A/B-segment vehicles and in auxiliary machines. Low remanence has to be offset by rotor topology (spoke-type, flux-concentrating, or a larger stack) rather than by grade selection alone.
• Hybrid rotors (NdFeB + ferrite layers or poles) — used to split the torque, cost, and demagnetization burden instead of committing the entire air-gap flux to one material family.
• Bonded NdFeB and compression-molded magnets — a smaller share of traction rotors, but useful where net-shape geometry, high pole count, or integrated housings reduce assembly cost enough to offset lower remanence.

Magnet Material Families Should Be Compared on System-Level Decision Axes, Not Datasheet Values Alone

The table below compares families on the axes that actually drive motor-level decisions: not only intrinsic magnetic properties, but also volume, thermal burden, supply risk, and landed magnet cost. Magnetic ranges are typical published grade windows at 20 °C. Volume and cost ratios are illustrative and topology-dependent; they are screening aids, not quotes.

What this means for your design. If the platform is in the roughly 80–150 kW class with moderate packaging constraints, GBD NdFeB or a hybrid rotor is usually the first pair of options worth closing on. A ferrite-dominant traction machine becomes realistic when diameter and stack length are allowed to grow and peak-power demand stays modest. Datasheet Br/Hcj ranking without a rotor topology and a thermal duty cycle will mis-rank all five columns.

Demagnetization Risk and Thermal Design Windows Determine Material Viability

Two coupled constraints decide whether a grade is viable in a traction rotor. They are screening checks, not a substitute for nonlinear electromagnetic-thermal analysis at temperature.

Constraint 1 — Operating-point coercivity margin (engineering screen):

A commonly used design screen is $K_{\mathrm{d}\mathrm{e}\mathrm{m}\mathrm{a}\mathrm{g}}\gtrsim 0.3$, where $T_{max}$ is the worst-case local magnet temperature (including fault-current events) and $H_{\mathrm{l}\mathrm{o}\mathrm{a}\mathrm{d},\mathrm{p}\mathrm{e}\mathrm{a}\mathrm{k}}$ is the peak demagnetizing field in the magnet from stator MMF under short-circuit or field-weakening overshoot. Both $H_{cj}$ and $H_{\mathrm{l}\mathrm{o}\mathrm{a}\mathrm{d},\mathrm{p}\mathrm{e}\mathrm{a}\mathrm{k}}$ are in A/m, so $K_{\mathrm{d}\mathrm{e}\mathrm{m}\mathrm{a}\mathrm{g}}$ is dimensionless.

This inequality is a margin heuristic, not a physical trip level. Irreversible loss occurs when the local operating point drops below the knee of the $J$–$H$ curve at that temperature. $H_{cj}$ is the $J=0$ intercept; the knee field $H_k$ can sit well above $H_{cj}$ if squareness is poor, and local field concentration at magnet corners is higher than a mean $H_{\mathrm{l}\mathrm{o}\mathrm{a}\mathrm{d}}$. A result below ~0.3 means the design is too close to the knee for manufacturing scatter, hot-spot uncertainty, and fault transients — it does not automatically mean the rotor will demagnetize, and a result above 0.3 does not prove it will not. Close the loop with temperature-dependent FEA and measured recoil data.

What this tells you practically. A standard N42-class sintered grade (typically an 80 °C thermal class, no GBD) cannot hold a useful knee margin at a 180 °C rotor hot spot. The practical responses are a higher thermal class (SH/UH/EH), GBD processing, a thicker magnet or flux-barrier layout that lowers $H_{\mathrm{l}\mathrm{o}\mathrm{a}\mathrm{d},\mathrm{p}\mathrm{e}\mathrm{a}\mathrm{k}}$, or a lower fault-current / field-weakening intercept. Those are geometry and cooling choices, not just a grade swap.

Constraint 2 — Thermal payback from reduced magnet volume.

A higher-Br grade allows a thinner magnet or a smaller rotor OD. Thinner magnets also put loss into a smaller thermal mass and can raise the equilibrium temperature, which erodes the coercivity margin the higher Br was supposed to buy. That circular dependence is why electromagnetic and thermal models have to iterate. A single-point Br/Hcj comparison on a 20 °C datasheet does not close the design.

Current Production Programs Cluster Around a Few Practical Design Windows

The windows below are typical industry patterns, not a census of named OEM specifications. Cooling method, pole count, and duty cycle can move a program from one row to another.

• High-performance (about 200 kW peak and above): Sintered NdFeB, usually GBD, SH or UH thermal class. Room-temperature Br often specified in the upper 1.2 T range or above. Continuous rotor-magnet design point commonly 150–160 °C with stator liquid cooling and, on some platforms, oil on the rotor. Direct oil cooling does not remove the need for a fault-case knee check.
• Mid-range (about 80–150 kW peak): GBD NdFeB or hybrid NdFeB + ferrite. This band is where low-Dy and reduced-Dy bills of material are pursued most actively, because magnet cost is visible and packaging is not yet unconstrained.
• A-segment / urban EV (about 30–70 kW peak): Ferrite-dominant spoke-type IPM, or a reluctance-based machine with little or no magnet. The packaging penalty is easier to accept in this vehicle class. Some programs use bonded NdFeB when a modest torque-density gain is worth more than sintered-magnet process constraints.
• Auxiliary motors (pumps, fans, compressors, roughly 0.5–5 kW): Bonded NdFeB or ferrite, chosen first on cost, net shape, and assembly — not on traction-level torque density.

These windows assume the magnet sees the temperatures and demagnetizing fields used to size it. They fail as soon as the boundaries in the next section are crossed.

These Material Choices Fail Outside Clearly Defined Thermal, Magnetic, and Environmental Boundaries

No magnet family is universal. The following inapplicability modes are well established and should be checked explicitly.

1. Thermal excursion beyond the knee. If the rotor hot-spot temperature during a three-phase short, locked-rotor event, or repeated launch exceeds the grade's knee-point temperature — even briefly — irreversible flux loss occurs. Ferrite is thermally forgiving on coercivity (positive $H_{cj}$ temperature coefficient) but magnetically weak. NdFeB is magnetically strong and thermally unforgiving. Neither is "safe" without a quantified hot-spot and fault-field analysis.
2. Deep flux-weakening at elevated temperature. Large negative d-axis current at high speed and high magnet temperature is a more common field demagnetization cause than bulk overheating alone. The operating point is driven toward or through the knee while $H_{cj}(T)$ is already reduced. Aggressive field-weakening maps must be overlaid on the hot $J$–$H$ curve, not on the 20 °C datasheet.
3. Corrosion, edge attack, and hydrogen-assisted degradation. Sintered NdFeB is chemically reactive. Incomplete coating at edges and corners of segmented magnets opens grain-boundary corrosion and progressive flux loss over vehicle life. Oil-cooled (ATF) rotors add chemical-compatibility and coating-adhesion requirements that air-cooled coating specs do not automatically satisfy.
4. Eddy-current heating in unsegmented blocks. Loss scales with electrical frequency, magnet conductivity, and the square of the dimension normal to the changing flux. There is no single frequency trip point. For unsegmented sintered blocks, eddy-current heating often becomes a design issue in the several-hundred-hertz range and is routinely addressed by segmentation or, less commonly, higher-resistivity / grain-isolated microstructures. Segmentation adds cutting yield loss, coating area, assembly tolerance stack-up, and magnetization complexity.
5. Manufacturing scatter that the magnetic margin did not budget. Dimensional tolerance on magnet thickness and width, Br and $H_{cj}$ lot scatter, magnetization incomplete at corners, and rotor-pocket distortion after shrink-fit or overmold all move the local load line. A design that is only marginally above the knee on nominal geometry will not survive PPAP-level capability.
6. Supply-chain single-source lock-in. Putting 100 % of magnet volume on one geography or one melt source creates program risk that rotor redesign cannot fix after tooling. Dual-source qualification with agreed Br, $H_{cj}$, and dimensional windows is a design-phase activity.

AIC Engineering Supports Traction-Motor Programs Across These Design Windows

AIC Engineering works with traction-motor programs from concept through volume production across the material families above:

• Magnetic circuit and rotor-assembly design. The AIC Engineering team works with motor designers on magnet geometry, segmentation, and magnetization patterns so that torque targets and demagnetization margin are satisfied together. Support includes multi-pole rings and, where the electromagnetic architecture actually benefits, Halbach-type arrays — the latter are not a default for mainstream interior permanent-magnet traction rotors.
• Prototype magnets on a short loop. When designers need physical magnets for dyno correlation, AIC Engineering can supply prototype sets with production-representative properties on a short working-day cycle (commonly in the 3–7 day range, depending on grade, coating, and magnetization). That loop only helps if the prototype grade, coating, and segmentation match the intended production construction.
• Lot characterization for automotive documentation. Lots can be characterized for Br, $H_{cj}$, flux uniformity, dimensional tolerance, and coating integrity. Data packages are prepared to support customer DFMEA, DVP&R, and PPAP submissions. Inspection scope should be written into the specification rather than assumed.
• Supply across NdFeB, ferrite, and bonded families. Logistics can be aligned to major EV manufacturing regions. Dual-source windows still have to be defined by the motor program; a second ship point does not by itself equal a second qualified melt and finish route.

A Design-Review Checklist Helps Freeze Magnet Specifications Before Tooling Commitments

Before freezing the magnet specification, verify the following:

1. Worst-case rotor-magnet temperature is confirmed by thermal analysis that includes fault events (three-phase short, locked rotor, repeated launch) — not only the continuous rating point.
2. Knee margin is checked at that temperature and at the peak demagnetizing field, using $K_{\mathrm{d}\mathrm{e}\mathrm{m}\mathrm{a}\mathrm{g}}$ only as a screen and temperature-dependent FEA plus recoil data as the close-out.
3. Coating specification matches the real coolant and atmosphere (air, water-glycol, ATF), including edge and corner coverage on segmented pieces.
4. Segmentation (or another eddy-current control) is decided from frequency, block size, and calculated magnet loss — not from a single frequency cutoff.
5. Dual-source qualification is started in design, with agreed tolerance bands for Br, $H_{cj}$, and dimensional scatter.
6. End-of-line magnetization and flux verification are agreed among the magnet supplier, the rotor assembler, and the motor plant.
7. Irreversible loss over life is budgeted in the torque margin. A few percent over vehicle life is a common NdFeB allowance when coating and hot-spot control are competent; it is not a material constant. Ferrite aging is usually negligible if the magnet remains mechanically intact.
8. Manufacturability is closed: pocket fill, chamfer/chip allowance, magnetize-in-assembly versus pre-magnetized insertion, and residual unbalance after magnetization.

The Next Steps for Your Program Follow From the Windows Above

1. Map the thermal and performance envelope onto the design windows above and narrow the material family before opening a full supplier RFQ.
2. Request application-specific magnet samples and use them to correlate electromagnetic-thermal predictions on the actual prototype motor, including a fault-case flux check.
3. Qualify dual-source supply early — property windows belong in the specification, not in a deviation report during SOP ramp.
4. Talk to the AIC Engineering team about traction-motor magnet requirements, prototype sets, and supply options matched to platform timing and regional delivery. Start at https://www.aicmagnetics.com.

References

1. Gutfleisch, O. et al., "Magnetic Materials and Devices for the 21st Century: Stronger, Lighter, and More Energy Efficient," Advanced Materials, vol. 23, no. 7, 2011, pp. 821–842.
2. Widmer, J. D., Martin, R., and Kimiabeigi, M., "Electric Vehicle Traction Motors Without Rare Earth Magnets," Sustainable Materials and Technologies, vol. 3, 2015, pp. 7–13.
3. Hirota, K., Nakamura, H., Minowa, T., and Honshima, M., "Coercivity Enhancement by the Grain Boundary Diffusion Process to Nd–Fe–B Sintered Magnets," IEEE Transactions on Magnetics, vol. 42, no. 10, 2006, pp. 2909–2911.
4. Bianchi, N., Bolognani, S., Bon, D., and Dai Pré, M., "Rotor Flux-Barrier Design for Torque Ripple Reduction in Synchronous Reluctance and PM-Assisted Synchronous Reluctance Motors," IEEE Transactions on Industry Applications, vol. 45, no. 3, 2009, pp. 921–928.
5. Hendershot, J. R. and Miller, T. J. E., Design of Brushless Permanent-Magnet Machines, Motor Design Books LLC, 2nd ed., 2010.

<!-- JSON-LD: {&quot;@context&quot;: &quot;https://schema.org&quot;, &quot;@type&quot;: &quot;TechArticle&quot;, &quot;headline&quot;: &quot;What Magnetic Materials Are Commonly Used in New Energy Vehicle Motors? A Complete Analysis of Drive Motor Magnet Applications&quot;, &quot;inLanguage&quot;: [&quot;en&quot;, &quot;zh&quot;], &quot;keywords&quot;: &quot;magnetic materials for EV traction motors, NdFeB vs ferrite in electric vehicle motors, rare earth magnet trends in new energy vehicles, EV drive motor magnet selection, permanent magnet motor material comparison&quot;, &quot;about&quot;: [{&quot;@type&quot;: &quot;Thing&quot;, &quot;name&quot;: &quot;&quot;}, {&quot;@type&quot;: &quot;Thing&quot;, &quot;name&quot;: &quot;&quot;}], &quot;author&quot;: {&quot;@type&quot;: &quot;Organization&quot;, &quot;name&quot;: &quot;AIC Engineering&quot;, &quot;alternateName&quot;: &quot;骏材磁应用团队&quot;}, &quot;publisher&quot;: {&quot;@type&quot;: &quot;Organization&quot;, &quot;name&quot;: &quot;AIC Engineering&quot;, &quot;url&quot;: &quot;https://www.aicmagnetics.com&quot;}} -->