The four commercially dominant permanent magnet families, NdFeB, SmCo, alnico, and ceramic ferrite, cover roughly the full operating envelope from consumer-grade holding tasks to 350 °C aerospace actuators, and the right pick is decided by Br, intrinsic coercivity HcJ, maximum service temperature, corrosion resistance, and cost per unit of energy product [S1][S2][S3].
For most industrial buyers, the selection collapses to a small set of questions: required pull force or flux density, peak and continuous operating temperature, exposure to humidity or chemicals, dimensional envelope, geometry complexity, and budget per part [S1][S2].
Decision Criteria: The Five Numbers That Drive Material Choice
Eclipse Magnetics frames the selection around four questions (required pull/lift, environment, temperature, exposure to demagnetising fields), and each question maps to a measurable parameter on the magnet supplier's data sheet [S1].
Dura Magnetics recommends starting from end-use (holding, lifting, sensing, motor commutation), environment, temperature range, and required holding force, then narrowing material and grade from those constraints [S2].
First4Magnets ranks neodymium as the strongest commercially available material, with ceramic ferrite (strontium or barium ferrite) sitting at the opposite end of the strength/cost curve as the cheapest and most corrosion-tolerant option for low-to-medium performance duties [S3].
Jobmaster Magnets adds two design-side constraints that buyers often miss: hysteresis-loop shape (square-loop rare earths versus curved alnico) and the reluctance of the surrounding magnetic circuit, both of which decide whether a small rare-earth part will outperform a larger alnico or ferrite block in the same assembly [S4].
The Four Mainstream Families: Where Each One Wins
Neodymium-iron-boron (NdFeB) delivers the highest energy product (BH)max in the commercial catalogue, with common grades spanning N35 to N52 and a practical maximum service temperature that depends on grade: standard N-series parts are typically rated for 80 °C continuous service, while UH (80–180 °C) and EH/AH grades extend continuous operation toward 200 °C and above [S2][S3].
Samarium cobalt (SmCo) trades some peak energy for thermal headroom and corrosion resistance: SmCo parts retain useful flux at continuous service temperatures of 250–350 °C and are essentially immune to oxidation, which is why aerospace, medical, and downhole tooling routinely specify it where neodymium would irreversibly lose flux [S2][S3].
Alnico (Al-Ni-Co, typically grades 2 through 9) is the legacy high-temperature option, with a maximum service temperature approaching 500–550 °C and strong corrosion resistance, but at roughly one-fifth the energy product of samarium cobalt, so the same duty requires a substantially larger magnet volume [S2][S4].
Ceramic ferrite (hard ferrite, SrFe12O19 or BaFe12O19) is the lowest-cost, most corrosion-resistant family, widely used in motors, speakers, refrigerator magnets, and craft applications, but it is brittle, requires diamond tooling for tight-tolerance machining, and is restricted to relatively simple geometries [S2][S3].
Side-by-Side Comparison on the Four Decision Criteria
Pulled directly from the four reference guides, the trade-off matrix looks like this on the criteria that drive 80% of industrial selections [S1][S2][S3][S4]:
Strength (energy product, BH)max: NdFeB highest, typically 30–55 MGOe for common grades; SmCo second at roughly 16–32 MGOe; alnico around 4–10 MGOe; ceramic ferrite lowest at roughly 1–5 MGOe. Maximum continuous service temperature: alnico best at 500 °C and up; SmCo next at 250–350 °C; NdFeB 80–200 °C depending on grade; ceramic ferrite typically 250 °C but with low flux density. Corrosion resistance: SmCo and ceramic ferrite essentially immune without coating; alnico also corrosion-tolerant; NdFeB the weakest, almost always requiring nickel, zinc, epoxy, or parylene plating in humid or salt-spray service. Relative cost: ceramic ferrite cheapest per kg; NdFeB medium and dominated by rare-earth price cycles; alnico moderate; SmCo the most expensive, driven by cobalt content and tighter supply [S2][S3].
Geometry flexibility follows the same pattern: alnico can be cast into complex shapes including horseshoes and rotor assemblies, SmCo and NdFeB are sintered and limited to simpler shapes, and ceramic ferrite requires diamond grinding for tight tolerances [S2][S4].
Who Each Material Is For, and Who Should Avoid It
Pick NdFeB when strength-to-volume is the dominant constraint: brushless DC motors, e-mobility traction rotors, magnetic couplings, magnetic separators, and any compact consumer device. Buyers in marine, medical-implant, or high-humidity outdoor service should plan for plating or a sealed housing, since uncoated NdFeB corrodes rapidly and the corrosion products damage the magnetic structure [S2][S3].
Pick SmCo when the duty includes continuous exposure above 200 °C, operation in corrosive chemicals, or tight stability over decades (aerospace actuators, travelling-wave tubes, military hardware, certain medical devices). Buyers on a consumer-electronics BOM should not pick SmCo: the unit cost premium versus NdFeB is rarely justified outside thermal or corrosion niches [S2].
Pick alnico when the magnet must survive 500 °C+ peaks, when a cast complex shape is needed (traditional horseshoe, rotor, or sensor housing), or when long-term flux stability under mild temperature reversals matters more than raw strength. Pick against alnico for any miniaturised consumer assembly: the volume penalty versus NdFeB is the deal-breaker in phones, headphones, and similar products [S2].
Operating Limits, Failure Modes, and Sourcing Pitfalls
Three failure modes account for most field returns. Thermal demagnetisation hits NdFeB first: a standard N-grade part near its Curie temperature, or even a hot spot from a localised heat source, can lose a permanent fraction of its Br and the loss is not recoverable by remagnetising through the part. The same failure shows up in SmCo and alnico only at much higher temperatures, which is the practical reason the high-temperature bracket exists [S1][S4].
Corrosion is the second mode and is essentially an NdFeB-only problem in commercial assemblies; once the sintered grain boundary oxidises, the magnet can crack from internal pressure and shed flux. Plating (Ni, Ni-Cu-Ni, Zn, epoxy, parylene) is standard, not optional, for any NdFeB part exposed to humidity cycling or salt spray [S2].
On the supply side, neodymium and samarium prices have been notably volatile through 2024–2025 because both are tied to Chinese separation capacity and to by-product supply from other rare-earth mining; Jobmaster Magnets explicitly flags heavy rare-earth additives (Dy, Tb) used to raise HcJ in high-temperature NdFeB grades as labour-intensive to extract, which is the structural reason small rare-earth parts carry the cost premium they do [S4].
Selection Workflow and Shortlist Logic
A repeatable five-step shortlist keeps the choice from drifting. Step 1, set the required pull force or air-gap flux density and the available envelope; step 2, lock the maximum continuous and peak service temperature including any nearby heat sources; step 3, write down the corrosion environment (indoor dry, outdoor, salt spray, chemical bath) and decide whether plating or a sealed housing is acceptable; step 4, fix the geometry method (sintered, cast, bonded, machined) and the tolerance class; step 5, apply unit-cost and volume constraints to eliminate candidates [S1][S2].
With those five answers the shortlist usually reduces to one or two families. For example, a 25 N holding force in a 12 mm diameter, 5 mm thick disc at room temperature indoors almost always lands on NdFeB grade N35 or N42. A 25 N holding force at 250 °C continuous service in a corrosive outdoor enclosure lands on SmCo grade 2:17. A 5 N holding force in a low-cost consumer fridge magnet almost always lands on anisotropic ceramic ferrite. A horseshoe-shaped sensor magnet cast to near-net shape at 300 °C service lands on alnico 5 [S1][S2][S3].
For a deeper dive on how material choice cascades into the rest of the assembly, the magnetic material properties reference covers Br, HcJ, and (BH)max definitions that anchor every data sheet on the four families above. Where the magnet is the field source inside a larger instrument, the magnetic sensor specification reference explains how temperature coefficient, linearity, and hysteresis of the source magnet drive sensor accuracy. In fluid-handling applications where the magnet is sealed inside a moving coupling, the magnetic drive pump duty map shows how SmCo's temperature headroom versus NdFeB changes the allowed process fluid temperature.
Two trackable signals to watch through the rest of 2026: NdFeB pricing through Chinese rare-earth separation output, which directly shifts the NdFeB-versus-SmCo crossover on cost-driven designs, and the rollout of higher-HcJ grades (N54 and above, plus Dy-lean UH/EH grades) that keep pushing the maximum usable temperature of NdFeB upward and erode the SmCo thermal niche on a part-by-part basis [S2][S3][S4].
For related coverage, see Power Grid Raw Material Sourcing Guide: 2026 Spec Reference for Copper, Aluminium.