Iron-nitride / FeN / α″-Fe₁₆N₂ rare-earth-free permanent magnets — technology deep-dive
Access date used throughout: 2026-04-26
Scope note: The central technology owner, Niron Magnetics, is private and does not file 10-K or S-1 documents with the U.S. SEC. Company-reported performance and adoption claims are therefore marked as Tier 2 self-reported unless separately supported by a Tier 1 patent, peer-reviewed paper, or regulator document.
Summary and why this matters
Iron-nitride permanent magnets, specifically the metastable ordered phase α″-Fe₁₆N₂, are a candidate class of “rare-earth-free” permanent magnets. The strategic reason is straightforward: high-performance neodymium-iron-boron (NdFeB) permanent magnets depend on neodymium, praseodymium, dysprosium, and terbium, all of which are subject to supply concentration, export controls, and price volatility. FeN magnets consist only of iron and nitrogen, with no rare-earth elements.
The technical claim that makes FeN interesting is a very high saturation magnetic polarization, often reported near 2.8–3.0 T for α″-Fe₁₆N₂. That is materially above the roughly 2.15 T of pure iron and above the roughly 1.6 T of NdFeB. If that magnetization could be paired with enough coercivity, the theoretical maximum energy product would be enormous — on the order of 200 MGOe or more by the standard magnetic energy-density bound. However, laboratory FeN magnets have so far demonstrated only about 20 MGOe in free-standing foils, which is below the roughly 50 MGOe already available in commercial NdFeB. The gap between theoretical FeN performance and measured FeN performance is the core technical risk.
Technical basis
2.1 Phase identity and crystal structure
The relevant phase is α″-Fe₁₆N₂, an ordered body-centered tetragonal, interstitial iron nitride. Its formation and properties are documented in the peer-reviewed literature:
“Synthesis of α″-Fe₁₆N₂ with high saturation magnetization,” Y. Jiang, M. A. Mehedi, J.-P. Wang, Scientific Reports 5, 15073 (2015), DOI: 10.1038/srep15073. Accessed 2026-04-26.
This paper reports synthesis of α″-Fe₁₆N₂ and reports high saturation magnetization for the phase; the exact value used in this note’s calculation is ≈2.9 T, sourced to that paper. The phase is not a stable equilibrium iron-nitrogen compound across ordinary processing conditions; it is metastable, which creates manufacturing and thermal-stability challenges.
A granted U.S. patent also discloses iron-nitride magnet formation methods:
U.S. Patent US9997279B2, “Iron nitride permanent magnet and technique for forming iron nitride permanent magnet,” granted 2018-06-12, assignee Regents of the University of Minnesota. Accessed 2026-04-26.
The patent is Tier 1 evidence of recorded claims to an iron-nitride permanent magnet and to techniques for forming it. It is not evidence of commercial yield, coercivity, thermal performance, or cost.
2.2 Why FeN has a high magnetization
The scientific rationale is that nitrogen insertion into the iron lattice changes the interatomic spacing and electronic structure, increasing the magnetic moment per iron atom relative to ordinary α-Fe. This is why thin-film and foil studies going back decades have reported anomalously high moments for α″-Fe₁₆N₂.
An independent review that places FeN in the broader rare-earth-free permanent-magnet landscape is:
J. M. D. Coey, “Permanent magnets: Plugging the gap,” Scripta Materialia 67, 524–529 (2012), DOI: 10.1016/j.scriptamat.2012.04.036. Accessed 2026-04-26.
Coey’s review discusses the gap between ferrites and NdFeB and identifies candidates such as Fe₁₆N₂. It also provides the context for why a permanent magnet needs both high saturation magnetization and high coercivity/anisotropy: a large magnetization alone does not make a useful permanent magnet.
2.3 Metastability and processing constraints
Because α″-Fe₁₆N₂ is metastable, it can decompose into α-Fe and γ′-Fe₄N if processed at excessive temperature. The patent and the peer-reviewed work describe low-temperature or kinetically controlled processing methods such as nitriding, nitrogen implantation, and foil processing. The precise thermal stability window varies by processing route, but this note treats any specific temperature ceiling as needs primary source validation unless sourced directly to a verified experimental dataset.
The no-rare-earth composition claim is repeated by Niron Magnetics:
Niron Magnetics technology page, self-reported technical description, accessed 2026-04-26, Tier 2 company document.
Niron states that FeN magnets use only iron and nitrogen and avoid rare-earth inputs. This is a company statement, not a regulator-verified claim, and it is commercially plausible because the phase formula contains no rare earth.
Patent landscape and IP evidence
3.1 Granted patent: US9997279B2
A granted U.S. patent was identified and verified:
- Patent number: US9997279B2
- Title: “Iron nitride permanent magnet and technique for forming iron nitride permanent magnet”
- Grant date: 2018-06-12
- Assignee: Regents of the University of Minnesota
- Document URL: patents.google.com
- Access date: 2026-04-26
- Tier: Tier 1 for the existence of recorded claims; not proof of commercial performance.
This is a granted patent, not an application. That is important for “moat” claims: a granted patent is stronger evidence of recorded claims than a pending application, but it does not show that any claim has survived post-grant review, licensing dispute, or that the claimed process is economical.
Niron Magnetics describes itself as commercializing technology based on University of Minnesota research. The company’s technology page links its work to that research lineage:
Niron Magnetics technology page, accessed 2026-04-26, Tier 2.
The license terms, exclusivity, royalty rate, and field scope are not disclosed in any public filing identified in this deep-dive. Mark: needs primary source validation.
3.2 Distinction: patents are not performance data
This deep-dive treats US9997279B2 only as evidence that someone has recorded claims on FeN magnet formation. It does not support the claim that FeN magnets are commercially ready, that they have high coercivity at scale, that they survive automotive temperature cycles, or that they are cost-competitive. Any performance or economic claim must come from peer-reviewed measurement or a regulator/standards-body document.
Peer-reviewed experimental evidence
4.1 High saturation magnetization
The principal Tier 1 paper used here is the Scientific Reports paper:
Y. Jiang, M. A. Mehedi, J.-P. Wang, “Synthesis of α″-Fe₁₆N₂ with high saturation magnetization,” Scientific Reports 5, 15073 (2015).
DOI: 10.1038/srep15073
URL: doi.org
Accessed 2026-04-26.
This paper reports synthesis of α″-Fe₁₆N₂ and reports a saturation magnetic polarization of approximately 2.9 T. This value is used in the worked calculation below. The paper is peer-reviewed and open-access, satisfying the Tier 1 requirement for technical claims.
Cross-verification caveat: The reported 2.9 T value was produced by the University of Minnesota/Niron-linked group. This deep-dive did not identify an independent institution reproducing that exact foil result. Therefore the saturation-magnetization performance claim is tagged [SINGLE-SOURCE] for independent-institution replication. The independent review by Coey (2012) supports the general scientific interest in Fe₁₆N₂, but it does not provide an independent measurement of the same foils.
4.2 Demonstrated energy product
The highest commonly cited near-commercial FeN energy product is 20 MGOe in free-standing foils:
Y. Jiang et al., “Synthesis of Fe₁₆N₂ Compound Free-Standing Foils with 20 MGOe Energy Product,” IEEE Magnetics Letters 7, 1303304 (2016).
DOI: 10.1109/LMAG.2016.2521572
URL: doi.org
Accessed 2026-04-26.
This is a peer-reviewed IEEE journal publication and qualifies as Tier 1 for the measured 20 MGOe result. The result is important because it demonstrates an energy product above ferrite and many bonded magnets, but still below high-end NdFeB.
Cross-verification caveat: The 20 MGOe demonstration is also from the same research lineage as the Scientific Reports paper and the patent. It is tagged [SINGLE-SOURCE] for independent reproduction. No independent lab result showing the same 20 MGOe in a separately manufactured FeN free-standing foil was identified in this deep-dive.
4.3 What this means for feasibility
FeN has an unusually high theoretical ceiling because of its magnetization, but the demonstrated coercivity and energy product are far below that ceiling. The technical feasibility question is therefore not whether FeN can be made in the lab — it has been made in foil form — but whether a high-coercivity, thermally stable, bulk FeN magnet can be manufactured at industrial scale. The patent and papers do not resolve that manufacturing question.
Worked calculation: theoretical energy-product ceiling and sensitivity
5.1 Formula
For a magnetic material, the maximum energy product \((BH)_{\max}\) is bounded by:
\[
(BH)_{\max} \le \frac{B_s^2}{4\mu_0}
\]
where:
- \(B_s = \mu_0 M_s\) is the saturation magnetic polarization in tesla,
- \(\mu_0 = 4\pi \times 10^{-7}\ \text{T·m/A}\).
This standard bound assumes an ideal rectangular loop and sufficient uniaxial anisotropy to achieve full squareness. It is a theoretical ceiling, not a predicted manufacturable value. The formula is standard magnetic-materials physics; Coey 2012 is used here as a Tier 1 source for magnetic-material fundamentals and the concept that anisotropy, not magnetization alone, controls real permanent-magnet performance.
5.2 Inputs and output
Input: \(B_s = 2.9\ \text{T}\) from the 2015 Scientific Reports paper.
\[
B_s^2 = 2.9^2 = 8.41\ \text{T}^2
\]
\[
4\mu_0 = 4 \times (4\pi \times 10^{-7}) = 5.026548 \times 10^{-6}\ \text{T·m/A}
\]
\[
(BH)_{\max} = \frac{8.41}{5.026548 \times 10^{-6}}
= 1.673 \times 10^6\ \text{J/m}^3
\]
Convert to MGOe:
\[
1\ \text{MGOe} = 7957.75\ \text{J/m}^3
\]
\[
(BH)_{\max} = \frac{1.673 \times 10^6}{7957.75} \approx 210\ \text{MGOe}
\]
So the theoretical energy-product ceiling for a hypothetical ideal FeN magnet with \(B_s = 2.9\ \text{T}\) is ≈210 MGOe.
5.3 Sensitivity table
| \(B_s\) (T) | Theoretical max energy product (MGOe) |
|---|---|
| 2.2 | ≈121 |
| 2.8 | ≈196 |
| 2.9 | ≈210 |
| 3.0 | ≈225 |
| 1.6 (NdFeB-class magnetization) | ≈64 |
This sensitivity table shows that the theoretical FeN ceiling is extremely sensitive to the actual \(B_s\). If the 2.9 T value is not reproduced at scale, the theoretical ceiling falls quickly. Even at \(B_s=2.2\) T, the theoretical ceiling is still large, but real magnets are limited by coercivity and loop squareness, not by saturation alone.
5.4 Why the real number is much lower
The 20 MGOe measured result is only about 9.5% of the 210 MGOe theoretical ceiling. The missing factor is coercivity/anisotropy. FeN’s demonstrated coercivity is not sufficient to produce a square hysteresis loop at that magnetization. Until FeN can be made with high coercivity at bulk scale, the theoretical ceiling does not translate into commercial magnets.
Parameter table
| Parameter | Value | Source (URL + ID + date) | Tier | Measured / derived / assumed / inferred | Cross-verified? |
|---|---|---|---|---|---|
| Saturation magnetic polarization \(B_s\) of α″-Fe₁₆N₂ | ≈2.9 T | Sci Rep 5, 15073 (2015), DOI 10.1038/srep15073, accessed 2026-04-26 | Tier 1 | Measured | Single-source at this exact value; independent review interest in FeN is in Coey 2012 |
| Demonstrated energy product | 20 MGOe | IEEE Magnetics Letters 7, 1303304 (2016), DOI 10.1109/LMAG.2016.2521572, accessed 2026-04-26 | Tier 1 | Measured | Single-source |
| Theoretical energy-product ceiling from \(B_s=2.9\) T | ≈210 MGOe | Calculation in this note; standard formula; Coey 2012 for magnetic background, DOI 10.1016/j.scriptamat.2012.04.036 | Tier 1 | Derived | Yes, formula standard |
| Commercial NdFeB benchmark | ≈50 MGOe | Coey 2012, DOI 10.1016/j.scriptamat.2012.04.036, accessed 2026-04-26 | Tier 1 | Published review | Single review source, but widely accepted |
| Phase identity | Ordered bct α″-Fe₁₆N₂ | Sci Rep DOI 10.1038/srep15073; US9997279B2, granted 2018-06-12 | Tier 1 | Measured / patent-disclosed | Single-source for indexing; patent records claims |
| Rare-earth-free composition | Fe + N only | Niron technology page, accessed 2026-04-26 | Tier 2 self-reported | Company claim | Single-source |
| Coercivity / thermal stability at commercial scale | Not publicly filed | needs primary source validation | n/a | Unverified | No |
| FeN production capacity / revenue | Not publicly filed | needs primary source validation | n/a | Unverified | No |
Company claims vs. filed disclosures
Niron Magnetics is private. There is no S-1, 10-K, or 20-F to cross-check its marketing statement that FeN magnets are a scalable, low-cost, rare-earth-free magnet platform. The company’s technology page provides performance intent, not independently audited numbers:
Niron Magnetics technology page, accessed 2026-04-26, Tier 2.
The grant date on the patent (2018) plus the peer-reviewed energy-product result (2016) shows that the FeN platform has been in development for roughly a decade, but the commercial gap remains: no public filing quantifies revenue, magnet shipments, customer offtake, or production capacity. Any statement that FeN is “ready for commercialization” should be treated as a company claim unless and until a customer contract, production capacity milestone, or audited financial disclosure appears.
A Volvo Cars Tech Fund investment is documented in a Tier 2 company press release:
Volvo Cars Tech Fund invests in Niron Magnetics, Volvo Cars global media press release, dated 2022-06-07, accessed 2026-04-26.
This shows strategic automotive interest, but an investment is not a purchase agreement, an offtake contract, or proof of magnet qualification. That distinction is important for interpreting catalysts: automotive capital entering the cap table is a signal, not a revenue event.
Unit economics / business model
The monetization structure is not fully disclosed because the primary technology owner is private.
Who monetizes:
- Niron Magnetics would monetize by manufacturing and selling FeN magnet powder or finished magnets, or by licensing production to magnet fabricators and motor manufacturers. Niron’s public statements support a manufacturing-plus-sales model, but no public price list, margin, or gross-profit figure is available. Any specific margin figure is needs primary source validation.
- Regents of the University of Minnesota holds the granted patent US9997279B2 and therefore participates in value capture through licensing or assignment. The patent assignee listed on the patent document is Regents of the University of Minnesota, not Niron, which implies a license relationship:
US9997279B2, granted 2018-06-12, assignee Regents of the University of Minnesota.
- Magnet fabricators, motor OEMs, and automotive suppliers would capture value if FeN magnets reduce rare-earth input costs and improve supply-chain resilience. There is no SEC-filed FeN-specific revenue or margin figure to cite here.
Typical margins: Not verifiable for FeN specifically. No public company reports FeN magnet margins. The correct evidentiary statement is: needs primary source validation.
How value is captured: The most likely value-capture path is a material-substitution margin: FeN magnets would compete on total vehicle/motor system cost by eliminating volatile NdPr/Dy/Tb inputs, not necessarily by beating NdFeB on energy product alone. That is an economic hypothesis, not yet proven by filed financials.
Disconfirming Evidence
The FeN investment thesis would be materially weakened or invalidated if:
- An independent lab reproduces the FeN foil synthesis and finds an energy product materially below the claimed 20 MGOe, or cannot reproduce the ≈2.9 T saturation magnetization reported by the University of Minnesota-linked group. This would invalidate the core performance premise.
- FeN magnets cannot achieve coercivity sufficient for electric-motor operating temperatures at bulk scale. A near-210 MGOe theoretical ceiling is irrelevant without practical coercivity; if FeN coercivity remains far below that required for EV traction motors, the commercial market would be limited to small, low-temperature, or non-motor applications.
- NdFeB prices decline or alternative low-rare-earth magnet technologies improve faster than FeN scales, narrowing the cost or supply-security advantage.
- Niron fails to convert the Volvo Cars Tech Fund investment or the Allison venture investment into an actual commercial supply agreement, customer qualification, or audited production capacity milestone. An investment on the cap table is not the same as a purchased magnet.
- The granted patent US9997279B2 is challenged, narrowed, or found not to cover the commercial FeN production route, weakening the licensing/value-capture story.
Recent developments
_Checked 2026-10-26 by the technology material-change monitor. Sources are the company's own press releases (primary) unless marked news._
- 2026-08-07 — $150M conditional U.S. Department of War loan commitment. Niron Magnetics announced a conditional commitment from the Department of War's Office of Strategic Capital (OSC) for a direct loan of up to $150 million, 20-year term, to fund construction and equipment for its Sartell, Minnesota plant. Niron states the plant is to be operational in 2027 and create up to 175 full-time jobs. This is the first U.S. federal debt commitment to iron-nitride magnet manufacturing and materially de-risks the capital plan for Plant 1. Source: nironmagnetics.com (DoW release referenced in the PR: war.gov OSC conditional loan commitment)
- 2026-09-09 — Honda Motor Co. investment. Niron announced an investment from Honda Motor Co., Ltd. through Honda Xcelerator Ventures, described as support for continued commercialization of rare-earth-free magnetic materials. OEM balance-sheet participation — not merely a supply MoU — is a stronger validation signal than the earlier MATTER (Jan 2026) and ASPINA (Jul 2026) collaborations. Source: nironmagnetics.com
- 2026-07-13 — ASPINA supply agreement. Niron and ASPINA (precision motors/actuators) entered a supply agreement to develop rare-earth-free motor solutions across global end markets — the first named motor-maker supply agreement. Source: nironmagnetics.com
- 2026-10-02 — Moog + Niron motor demo at AUSA 2026. Moog and Niron announced they will showcase rare-earth-free motor technology at AUSA 2026 (Oct 12–14) — a defense-market channel proof point, still a showcase rather than a disclosed production order. Source: nironmagnetics.com
- 2026-10-22 — second-plant site selection nearing decision (news). Axios Twin Cities reported Niron is "closing in on" the site for its proposed $1.8B high-volume plant (the site-selection process announced 2026-03-16). No site, incentive package, or financing has been disclosed; treat as unconfirmed until a company release. Source: axios.com
Read-through: the constraint on this technology has shifted from "is the phase manufacturable" toward "does financed capacity arrive on schedule." Federal debt (conditional), an OEM strategic investor, and named motor-maker supply agreements all point the same way, but none of it yet constitutes shipped volume or disclosed magnet performance at customer-qualified spec. Conditional-commitment language is not closed financing.