Interfacial origin of exchange bias in γ-Fe2O3/CoO and inverted CoO/γ-Fe2O3 binary nanoparticles: separating uncompensated-spin from compensated-interface contributions

In chemically synthesized binary nanoparticles pairing ferrimagnetic maghemite (γ-Fe2O3) with antiferromagnetic CoO, field cooling produces both a horizontal hysteresis-loop shift (the exchange-bias field $H_{\mathrm{ex}}$) and a vertical loop offset $M_{\mathrm{off}}$ attributed to frozen uncompensated interfacial spins. The 2009 study that established both the normal (γ-Fe2O3 core/CoO shell) and reverse (CoO core/γ-Fe2O3 shell) morphologies concluded that the unidirectional anisotropy originates from both magnetically compensated and uncompensated interfaces. A quantitative separation of the two contributions remains open, along with the linked aspects: how to separate $H_{\mathrm{ex}}$ from the uncompensated-moment contribution in the reverse structure where the switching-field distributions of the two phases overlap; the physical origin of the observed linear $M_{\mathrm{off}}$–$H_{\mathrm{ex}}$ correlation and whether it extends to the reverse structure; how particle size sets the blocking temperature of the exchange bias; and the atomic structure of the interface together with the relative weights of compensated and uncompensated interfacial spin configurations. Later literature — element-specific measurements on thin-film bilayers, frozen-spin analyses of Fe/Fe3O4 nanoparticles, granular CoO/γ-Fe2O3 studies, Monte Carlo simulations, and electron-magnetic-dichroism interface imaging on other core/shell systems — has constrained pieces of this question but performed no quantitative compensated/uncompensated decomposition for the γ-Fe2O3/CoO system. The problem is to provide that quantitative, interface-resolved account of exchange bias in these binary nanoparticles, or to refute the premises of the question (e.g., to show the $M_{\mathrm{off}}$–$H_{\mathrm{ex}}$ correlation is a measurement artifact).

Background

Exchange bias (EB) is the shift of a magnetic hysteresis loop along the field axis that appears when a ferromagnet (FM) — in nanoparticles, typically a ferrimagnet (FiM) — is cooled in a static magnetic field through the ordering temperature of an adjacent antiferromagnet (AFM). The cooling field orients the FM/FiM magnetization while the AFM orders; exchange coupling across the interface then imprints a unidirectional anisotropy, so that after field cooling (FC) the loop is displaced by the exchange-bias field $H_{\mathrm{ex}}$ and usually broadened, while after zero-field cooling (ZFC) it stays centered. In core/shell nanoparticles this coupling arises across the curved interface between the two phases; the temperature below which the EB appears is called its blocking temperature, and in nanoparticles it typically lies far below the bulk AFM ordering (Néel) temperature because the AFM volume is only a few nanometres across. Maghemite (γ-Fe2O3) is the ferrimagnetic iron oxide with the spinel crystal structure (a cubic oxide lattice shared by many ferrites) and cation-vacancy order; CoO is a rock-salt antiferromagnet whose bulk Néel temperature is about 291 K but whose nanoscale ordering is strongly size-dependent. Two quantities are customarily extracted from a field-cooled loop: the exchange-bias field $H_{\mathrm{ex}}$, taken as minus the mean of the two branch coercive fields, and the vertical loop offset $M_{\mathrm{off}}$, the displacement of the loop along the magnetization axis. A vertical offset signals a net moment of interfacial spins that are frozen by the cooling field and do not reverse in the measurement field — uncompensated spins, i.e., interfacial AFM (or FiM) moments whose antiparallel sublattice partners do not cancel locally. When a vertical offset is present, the coercive intercepts must be read on the horizontal line at $M_{\mathrm{off}}$; this bookkeeping convention was taken from the source paper into later nanoparticle work. By contrast, a magnetically compensated interface (equal and opposite interfacial moments) can transmit exchange coupling while carrying no net moment. Whether the EB of a given nanoparticle system is dominated by uncompensated frozen spins, by compensated-interface coupling, or by a mixture — and how that balance depends on which phase is the core, on particle size, and on the atomic structure of the interface — is a long-standing open issue; recent reviews of EB in nanostructures state that the origin of the effect is still not fully understood. In extended thin-film bilayers the uncompensated/pinned interfacial spins have been quantified (e.g., by polarized neutron reflectometry — which resolves the magnetization depth profile layer by layer — and X-ray magnetic circular dichroism, XMCD — an element-specific probe of the projection of local moments along the beam) and are known to correlate linearly with the EB field (Takano et al. 1997; Ohldag et al. 2003; Tsunoda et al. 2010; Guo et al. 2021). In chemically synthesized core/shell nanoparticles no such direct decomposition exists. The source study (Panagiotopoulos et al. 2009) prepared, by a two-step seed-mediated thermal decomposition of iron and cobalt acetylacetonates in hot oleylamine, both the normal γ-Fe2O3/CoO (FiM core, AFM shell) and the reverse CoO/γ-Fe2O3 (AFM core, FiM shell) binary nanoparticles of roughly 5 nm size, and measured FC/ZFC hysteresis loops at 10 K in fields up to 50 kOe. Both structures showed horizontal shifts (of order 1–2 kOe) together with vertical offsets; from the coexistence of the two the authors concluded that the unidirectional anisotropy originates from both magnetically compensated and uncompensated interfaces. A known methodological hazard frames the whole question: vertical loop shifts can also arise as minor-loop artifacts when the measured loop fails to reach saturation (Geshev 2008), and the source loops themselves did not close at the largest applied field, so any valid answer must control for this.

Problem Statement

For chemically synthesized γ-Fe2O3/CoO (normal: ferrimagnetic core, antiferromagnetic shell) and reverse CoO/γ-Fe2O3 binary nanoparticles, establish quantitatively the interfacial origin of the exchange bias — one overarching objective grounded in the source study's conclusion that the unidirectional anisotropy originates from both magnetically compensated and uncompensated interfaces, and comprising four linked open aspects identified from that conclusion and the subsequent literature: (1) In the reverse CoO/γ-Fe2O3 structure, where the switching-field distributions of the two phases overlap, disentangle the exchange-bias field $H_{\mathrm{ex}}$ from the contribution of the uncompensated moment (the vertical loop offset $M_{\mathrm{off}}$), which bulk hysteresis loops alone cannot do; this calls for probes that resolve the phases (e.g., element-specific magnetometry such as X-ray magnetic circular dichroism (XMCD), polarized neutron scattering, or equivalent field/temperature protocols) together with an explicit exclusion of minor-loop artifacts. (2) Determine whether the linear correlation between $M_{\mathrm{off}}$ and $H_{\mathrm{ex}}$ reported for the normal structure also holds in the reverse structure, and identify the physical mechanism that produces this correlation — why the frozen uncompensated interfacial moment should scale linearly with the exchange-bias field. (3) Determine how particle size, for samples with controlled size distributions, sets the blocking temperature of the exchange bias. (4) Determine the atomic structure of the ferrimagnet/antiferromagnet interface and the relative contributions of magnetically compensated and uncompensated interfacial spin configurations to the observed exchange bias. A complete answer takes one of two routes. Route (A): it addresses all four linked aspects above and quantitatively determines, for each configuration, how the measured $H_{\mathrm{ex}}$ partitions between uncompensated interfacial spins and compensated-interface coupling. Route (B): it refutes one of the two premises on which those aspects rest — (i) the unidirectional anisotropy has both magnetically compensated and uncompensated interface contributions, or (ii) the vertical loop offset $M_{\mathrm{off}}$ is a genuine frozen-uncompensated-spin signal (equivalently, the linear $M_{\mathrm{off}}$–$H_{\mathrm{ex}}$ correlation is physical). Refuting either premise alone completes the task and replaces, rather than supplements, route (A): for instance, demonstrating that $M_{\mathrm{off}}$, and hence the $M_{\mathrm{off}}$–$H_{\mathrm{ex}}$ correlation, is an artifact of unsaturated (minor) loops refutes premise (ii) and resolves the question without further work on the four aspects.

References

1. I. Panagiotopoulos, G. Basina, V. Alexandrakis, E. Devlin, G. Hadjipanayis, L. Çolak, D. Niarchos, V. Tzitzios, "Synthesis and Exchange Bias in γ-Fe2O3/CoO and Reverse CoO/γ-Fe2O3 Binary Nanoparticles", J. Phys. Chem. C 113(33), 14609–14614 (2009). DOI: 10.1021/jp8085446, https://doi.org/10.1021/jp8085446
2. T. Błachowicz, A. Ehrmann, M. Wortmann, "Exchange Bias in Nanostructures: An Update", Nanomaterials 13(17), 2418 (2023). DOI: 10.3390/nano13172418, https://doi.org/10.3390/nano13172418
3. M.-H. Phan, J. Alonso, H. Khurshid, P. Lampen-Kelley, S. Chandra, K. Stojak Repa, Z. Nemati, R. Das, Ó. Iglesias, H. Srikanth, "Exchange Bias Effects in Iron Oxide-Based Nanoparticle Systems", Nanomaterials 6(11), 221 (2016). DOI: 10.3390/nano6110221, https://doi.org/10.3390/nano6110221
4. K. Takano, R. H. Kodama, A. E. Berkowitz, W. Cao, G. Thomas, "Interfacial Uncompensated Antiferromagnetic Spins: Role in Unidirectional Anisotropy in Polycrystalline Ni81Fe19/CoO Bilayers", Phys. Rev. Lett. 79, 1130 (1997). DOI: 10.1103/PhysRevLett.79.1130, https://doi.org/10.1103/PhysRevLett.79.1130
5. H. Ohldag, A. Scholl, F. Nolting, E. Arenholz, S. Maat, A. T. Young, M. J. Carey, J. Stöhr, "Correlation between Exchange Bias and Pinned Interfacial Spins", Phys. Rev. Lett. 91, 017203 (2003). DOI: 10.1103/PhysRevLett.91.017203, https://doi.org/10.1103/PhysRevLett.91.017203
6. M. Tsunoda, H. Takahashi, T. Nakamura, C. Mitsumata, S. Isogami, M. Takahashi, "Linear correlation between uncompensated antiferromagnetic spins and exchange bias in Mn–Ir/Co100−xFex bilayers", Appl. Phys. Lett. 97, 072501 (2010). DOI: 10.1063/1.3479500, https://doi.org/10.1063/1.3479500
7. J. Guo, X. Zhao, Z. Lu, P. Shi, Y. Tian, Y. Chen, S.-s. Yan, L. Bai, M. Harder, "Evidence for linear dependence of exchange bias on pinned uncompensated spins in an Fe/FeO bilayer", Phys. Rev. B 103, 054413 (2021). DOI: 10.1103/PhysRevB.103.054413, https://doi.org/10.1103/PhysRevB.103.054413
8. Q. K. Ong, A. Wei, X.-M. Lin, "Exchange bias in Fe/Fe3O4 core-shell magnetic nanoparticles mediated by frozen interfacial spins", Phys. Rev. B 80, 134418 (2009). DOI: 10.1103/PhysRevB.80.134418, https://doi.org/10.1103/PhysRevB.80.134418
9. J. Geshev, "Comment on: 'Exchange bias and vertical shift in nanoparticles' [J. Magn. Magn. Mater. 313 (2007) 266]", J. Magn. Magn. Mater. 320, 600 (2008). DOI: 10.1016/j.jmmm.2007.05.026, https://doi.org/10.1016/j.jmmm.2007.05.026
10. Ó. Iglesias, A. Labarta, X. Batlle, "Exchange Bias Phenomenology and Models of Core/Shell Nanoparticles", J. Nanosci. Nanotechnol. 8, 2761 (2008). DOI: 10.1166/jnn.2008.18306, https://doi.org/10.1166/jnn.2008.18306
11. Ó. Iglesias, X. Batlle, A. Labarta, "Microscopic origin of exchange bias in core/shell nanoparticles", Phys. Rev. B 72, 212401 (2005). DOI: 10.1103/PhysRevB.72.212401, https://doi.org/10.1103/PhysRevB.72.212401
12. E. Eftaxias, K. N. Trohidou, "Numerical study of the exchange bias effects in magnetic nanoparticles with core/shell morphology", Phys. Rev. B 71, 134406 (2005). DOI: 10.1103/PhysRevB.71.134406, https://doi.org/10.1103/PhysRevB.71.134406
13. M. Vasilakaki, K. N. Trohidou, "Numerical study of the exchange-bias effect in nanoparticles with ferromagnetic core/ferrimagnetic disordered shell morphology", Phys. Rev. B 79, 144402 (2009). DOI: 10.1103/PhysRevB.79.144402, https://doi.org/10.1103/PhysRevB.79.144402
14. M. Vasilakaki, K. N. Trohidou, J. Nogués, "Enhanced Magnetic Properties in Antiferromagnetic-Core/Ferrimagnetic-Shell Nanoparticles", Sci. Rep. 5, 9609 (2015). DOI: 10.1038/srep09609, https://doi.org/10.1038/srep09609
15. G. Salazar-Alvarez, J. Sort, S. Suriñach, M. D. Baró, J. Nogués, "Synthesis and Size-Dependent Exchange Bias in Inverted Core−Shell MnO/Mn3O4 Nanoparticles", J. Am. Chem. Soc. 129, 9102 (2007). DOI: 10.1021/ja0714282, https://doi.org/10.1021/ja0714282
16. C. Liu, J. Cui, X. He, H. Shi, "Large exchange bias with remarkable thermostability in an inverted quasi core/shell CoO/γ-Fe2O3 granular system", J. Nanopart. Res. 16, 2320 (2014). DOI: 10.1007/s11051-014-2320-7, https://doi.org/10.1007/s11051-014-2320-7
17. X. Liu, B. P. Pichon, C. Ulhaq, C. Lefèvre, J.-M. Grenèche, D. Bégin, S. Bégin-Colin, "Systematic Study of Exchange Coupling in Core–Shell Fe3−δO4@CoO Nanoparticles", Chem. Mater. 27, 4073 (2015). DOI: 10.1021/acs.chemmater.5b01103, https://doi.org/10.1021/acs.chemmater.5b01103
18. H. Khurshid, M.-H. Phan, P. Mukherjee, H. Srikanth, "Tuning exchange bias in Fe/γ-Fe2O3 core-shell nanoparticles: Impacts of interface and surface spins", Appl. Phys. Lett. 104, 072407 (2014). DOI: 10.1063/1.4865904, https://doi.org/10.1063/1.4865904
19. D. del-Pozo-Bueno, M. Varela, M. Estrader, A. López-Ortega, A. G. Roca, J. Nogués, F. Peiró, S. Estradé, "Direct Evidence of a Graded Magnetic Interface in Bimagnetic Core/Shell Nanoparticles Using Electron Magnetic Circular Dichroism (EMCD)", Nano Lett. 21, 6923 (2021). DOI: 10.1021/acs.nanolett.1c02089, https://doi.org/10.1021/acs.nanolett.1c02089
20. A. López-Ortega, B. Muzzi, C. de Julián Fernández, C. Sangregorio, "Competitive Size Effects in Antiferromagnetic|Ferrimagnetic Core|Shell Nanoparticles for Large Exchange Bias", ACS Appl. Nano Mater. 7, 27489 (2024). DOI: 10.1021/acsanm.4c05505, https://doi.org/10.1021/acsanm.4c05505
21. X.-M. He, C.-W. Zhang, F.-F. Guo, S.-M. Yan, Y.-T. Li, L.-Q. Liu, H.-G. Zhang, Y.-W. Du, W. Zhong, "Exchange-biased hybrid γ-Fe2O3/NiO core–shell nanostructures: three-step synthesis, microstructure, and magnetic properties", Phys. Chem. Chem. Phys. 21, 11967 (2019). DOI: 10.1039/C9CP01265F, https://doi.org/10.1039/C9CP01265F
22. P. Maltoni, R. López-Martín, E. Sánchez, P. S. Normile, M. Vasilakaki, S. Lee, B. Burgos, E. López del Castillo, D. Peddis, C. Binns, K. N. Trohidou, R. Mathieu, J. Nogués, J. De Toro, "Non-Exchange Bias in Binary Nanoparticle Systems", Research Square preprint, posted 12 July 2023. DOI: 10.21203/rs.3.rs-3125651/v1, https://doi.org/10.21203/rs.3.rs-3125651/v1