The perovskite-type oxynitride LaTa1−xCox(O,N)3−δ (x = 0.01–0.05) has been reported as a possible bulk dilute ferromagnetic semiconductor whose magnetization persists to at least 600 K, far above room temperature, with no evidence of metallic Co clusters. The microscopic mechanism behind this long-range ferromagnetic order — how dilute intermediate-spin/mixed-valence Co ions, anionic (oxygen/nitrogen) vacancies, and a magnetically polarizable La–Ta–(O,N) host couple to produce it — is unknown, the intrinsic Curie temperature is undetermined because the material loses nitrogen near 600 K, and the element-resolved magnetic structure is beyond current neutron sensitivity. The problem is to establish a predictive microscopic account of this ferromagnetism: identify the Co spin/valence distribution and its correlation with vacancies, the exchange pathways and host polarization, the intrinsic Curie temperature of a thermally stable composition, and the resulting design rules for other hosts and dopants. The report exists only as an uncited arXiv preprint, so independent reproduction or refutation is itself part of the open question.

Background

A diluted magnetic semiconductor (DMS) is a non-magnetic semiconductor doped with a few percent or less of magnetic transition-metal ions, sought since 2000 as a material for spintronics — electronics that exploits electron spin. Dietl and co-workers predicted in that year that p-type ZnO- or GaN-based DMS compounds should order ferromagnetically above room temperature, i.e. with a Curie temperature TCT_C (the temperature above which spontaneous magnetization disappears) well above 300 K. Two decades of attempts failed: homogeneously dissolved 3d ions mostly remained paramagnetic, and many reported above-room-temperature signals were traced to extrinsic artifacts such as metallic Co clusters, while other unexpected ferromagnetism appeared in undoped, defect-rich oxides ('d0 magnetism', where the magnetism arises not from partially filled d shells but from defects such as vacancies and grain boundaries). No verified bulk semiconducting ferromagnet with TCT_C above room temperature existed.

The relevant host here is the perovskite-type oxynitride LaTa(O,N)3. Perovskites have the formula ABX3 with a small B-site cation in corner-sharing octahedra of anions; in oxynitrides some oxide anions are replaced by nitride anions, which are less electronegative and make the bonding more covalent. LaTa(O,N)3 was previously studied mainly as a red photocatalyst pigment, synthesized by ammonolysis (high-temperature reaction of an oxide precursor in flowing NH3), and is diamagnetic (non-magnetic). Bubeck and co-workers had previously shown that the precursor microstructure controls the tantalum oxidation state in the product oxynitride.

In the source work, Bubeck et al. (arXiv:2003.14259, four versions 2020–2022, apparently never journal-published) synthesized LaTa1−xCox(O,N)3−δ (x = 0.01, 0.03, 0.05) by ammonolysis of nanocrystalline Co-substituted LaTaO4 precursors. The products are red semiconductors (optical gap 1.7–1.9 eV), single-phase with space-group type Imma, and contain significant anion vacancies (marked □ in the refined composition, e.g. LaTa0.95Co0.05O1.12(6)N1.84(1)□0.03(3)). Conventional SQUID magnetometry (a superconducting quantum interference device measures sample magnetization) shows nearly temperature-independent saturating ferromagnetism from 1.8 K to 350 K, with moments rising from 0.6 to 1.53 µB per Co as x increases, strong enough that powder is visibly pulled by permanent magnets. In vacuum SQUID runs up to 950 K the magnetization persists and even increases by about 12% at 600 K, which the authors attribute to partial nitrogen loss creating further vacancies — from this they infer TC>600KT_C > 600 K while conceding the intrinsic TCT_C cannot be measured because the material decomposes below it. X-ray magnetic circular dichroism (XMCD, the difference in X-ray absorption for opposite light helicities, measured at the Co L2,3 edges in transmission) proves the magnetism is carried by oxidized, non-metallic Co — the spectral shape and energy position differ decisively from metallic Co — and indicates intermediate-spin Co3+ (a Co3+ ion, d6, whose spin state lies between the low-spin and high-spin configurations); applying XMCD sum rules (formulae that extract spin and orbital moments from integrated XMCD spectra, which require assuming the number nH of 3d holes) with the upper-limit hole count nH = 4 yields Co moments about 25% below the SQUID values, implying that the non-magnetic host matrix itself carries part of the magnetization (is 'magnetically polarized'). Extensive controls exclude extrinsic origins: transmission electron microscopy (TEM) with energy-dispersive X-ray spectroscopy (EDX) element mapping shows homogeneous Co below the 4 nm cluster-detection limit with no Co-rich regions; the magnetization lacks the strong temperature dependence expected for superparamagnetic clusters; in-plane/out-of-plane anisotropy of pressed pellets scales with the demagnetization factor of a volume magnetization; and a hypothetical 10 at% ferromagnetic secondary phase would need an impossible moment of 11.4 µB per Co. Neutron diffraction at 10 K and 300 K, however, could not detect any magnetic scattering — at ≤1 at% Co it falls below the detection limit — so the spatial magnetic structure (where the moments sit, which exchange paths connect them, and whether the host polarization is localized near Co/vacancies or extended) is unresolved. A comparison system, LaTa0.95Ni0.05(O,N)3−δ, has a much lower Ni solubility (≤0.6 at%), shows secondary Ni3N, and is dominated by paramagnetism, suggesting Co is currently the only dopant that works. The authors' own preliminary interpretation likens the effect to the vacancy-dominated 'ferromagnetic foam' proposed for grain boundaries in nanograined ZnO, but note that the LaTa(O,N)3 case must be more complicated, involving a complex interplay between the Co ions, the anion vacancies, the O/N distribution, and the polarizable host.

Problem Statement

Establish a predictive microscopic understanding of the reported above-room-temperature dilute ferromagnetism in the perovskite-type oxynitride LaTa1−xCox(O,N)3−δ (x = 0.01–0.05). All requested results jointly serve this single objective — a validated mechanism — and are the components the source work itself identifies as unresolved:

  1. Mechanism and spin/valence chemistry. Identify how the dilute Co ions (the distribution of intermediate-spin Co3+ versus Co2+, and their surface-to-bulk gradient), the anionic (O/N) vacancies and their spatial correlation with Co, and the O/N-disordered La–Ta–(O,N) host electronic structure couple to produce the reported robust long-range ferromagnetic order, whose magnetization persists to at least 600 K — including whether, and by what pathway, the magnetically polarized host mediates exchange between Co ions separated by large ionic distances, and what accounts quantitatively for the observed ~25% excess of SQUID-derived over XMCD-derived moments per Co.
  2. Intrinsic Curie temperature. Determine the equilibrium TCT_C of a composition or measurement protocol that avoids the partial nitrogen loss that currently masks it (e.g. thermally stable compositions, inert or high-pressure environments).
  3. Magnetic structure. Determine the element- and site-resolved magnetic structure — the spatial distribution of Co moments and of the inferred host polarization (localized near Co/vacancies versus extended), and the dominant exchange paths — beyond the current neutron-diffraction detection limit.
  4. Stoichiometry control and design rules. Establish how anionic stoichiometry and nitrogen/vacancy content tune the magnetic exchange, magnetic volume fraction, and TCT_C, and derive from the mechanism which host matrices and dopants (with what synthesis windows) can reproduce the behavior — explaining why Co succeeds while Ni fails.

An answer may take the form of a validated microscopic model (e.g. first-principles electronic-structure and exchange calculations over realistic dilute-disordered, vacancy-containing configurations), decisive new experiments (element-specific magnetometry, resonant or polarized neutron/X-ray scattering, muon spin rotation, inert-atmosphere thermomagnetometry), or a quantitative demonstration that the reported signal is extrinsic — provided, in every case, that every published observable (saturation moments per Co of 0.6–1.53 µB, hysteretic behavior to 950 K with the 12% step at 600 K, XMCD spectral shape distinct from Co metal, anisotropic demagnetization behavior, phase purity) is accounted for. These are alternative evidential forms for the same single objective, not alternatives among items 1–4: whichever form is taken, the answer must identify the microscopic origin of the reported magnetization and thereby settle items 1–4 to the extent that the established origin makes them well posed. In the extrinsic branch, item 1 is answered by identifying the actual source of the observed signal, and items 2–4, which presuppose intrinsic ferromagnetism, are resolved by that demonstration rather than by separate results.

References

  1. Cora Bubeck, Eberhard Goering, Robert Lawitzki, Kathrin Küster, Wilfried Sigle, Marc Widenmeyer, Ulrich Starke, Clemens Ritter, Gabriel J. Cuello, Peter Nagel, Michael Merz, Stefan Schuppler, Gisela Schütz, Anke Weidenkaff, 'Observation of a possible diluted ferromagnetism above room temperature in cobalt-substituted LaTa(O,N)3−δ', arXiv:2003.14259 (v1 2020, v4 2022; preprint, no journal version found), doi:10.48550/arXiv.2003.14259, https://arxiv.org/abs/2003.14259
  2. Cora Bubeck, Marc Widenmeyer, Gunther Richter, Mauro Coduri, Eberhard Goering, Songhak Yoon, Anke Weidenkaff, 'Tailoring of an unusual oxidation state in a lanthanum tantalum(IV) oxynitride via precursor microstructure design', Communications Chemistry 2, 134 (2019), doi:10.1038/s42004-019-0237-x, https://doi.org/10.1038/s42004-019-0237-x
  3. T. Dietl, H. Ohno, F. Matsukura, J. Cibert, D. Ferrand, 'Zener Model Description of Ferromagnetism in Zinc-Blende Magnetic Semiconductors', Science 287, 1019–1022 (2000), doi:10.1126/science.287.5455.1019, https://doi.org/10.1126/science.287.5455.1019
  4. J. M. D. Coey, M. Venkatesan, C. B. Fitzgerald, 'Donor impurity band exchange in dilute ferromagnetic oxides', Nature Materials 4, 173–179 (2005), doi:10.1038/nmat1310, https://doi.org/10.1038/nmat1310
  5. B. B. Straumal, A. A. Mazilkin, S. G. Protasova, A. A. Myatiev, P. B. Straumal, G. Schütz, P. A. van Aken, E. Goering, B. Baretzky, 'Magnetization study of nanograined pure and Mn-doped ZnO films: Formation of a ferromagnetic grain-boundary foam', Physical Review B 79, 205206 (2009), doi:10.1103/PhysRevB.79.205206, https://doi.org/10.1103/PhysRevB.79.205206
  6. J. M. D. Coey, P. Stamenov, R. D. Gunning, M. Venkatesan, K. Paul, 'Ferromagnetism in defect-ridden oxides and related materials', New Journal of Physics 12, 053025 (2010), doi:10.1088/1367-2630/12/5/053025, https://doi.org/10.1088/1367-2630/12/5/053025
  7. L. M. C. Pereira, 'Experimentally evaluating the origin of dilute magnetism in nanomaterials', Journal of Physics D: Applied Physics 50, 393002 (2017), doi:10.1088/1361-6463/aa801f, https://doi.org/10.1088/1361-6463/aa801f
  8. Shishin Mo, Yuji Kurauchi, Tsukasa Katayama, Yasushi Hirose, Tetsuya Hasegawa, 'Theoretical Investigation of the Role of the Nitride Ion in the Magnetism of Oxynitride MnTaO2N', Journal of Physical Chemistry C 123, 25379–25384 (2019), doi:10.1021/acs.jpcc.9b07446, https://doi.org/10.1021/acs.jpcc.9b07446
  9. Jhonatan R. Guarín, Carlos Frontera, Judith Oró-Solé, Jaume Gàzquez, Clemens Ritter, Josep Fontcuberta, Amparo Fuertes, 'High-Temperature Synthesis of Ferromagnetic Eu3Ta3(O,N)9 with a Triple Perovskite Structure', Inorganic Chemistry 62, 17362–17370 (2023), doi:10.1021/acs.inorgchem.3c02691, https://doi.org/10.1021/acs.inorgchem.3c02691
  10. Margarida Barroso, Mian Dai, Cora Bubeck, Marco Scavini, Gabriel J. Cuello, Hongbin Zhang, Anke Weidenkaff, Marc Widenmeyer, 'Investigation of O/N Ordering in Perovskite-Type Oxynitrides La1−xYxTa(O,N)3 on Long Range and Short Scale', Inorganics 12, 90 (2024), doi:10.3390/inorganics12030090, https://doi.org/10.3390/inorganics12030090
  11. Sara Alvarez-Sosa, J. Manuel Mora-Hernandez, Leticia M. Torres-Martínez, 'LaTaON2 insights, from synthesis to photocatalytic and photoelectrochemical applications; a review', Materials Science in Semiconductor Processing 197, 109720 (2025), doi:10.1016/j.mssp.2025.109720, https://doi.org/10.1016/j.mssp.2025.109720