Unresolved core–shell mechanism of $T_1$ relaxivity in dopant-engineered ultrasmall ferrite nanoparticles: atomic structure of the disordered surface shell, the apparent surface Mn-substitution ceiling, and quantitative core–shell magnetic coupling

Ultrasmall ferrite nanoparticles (diameter below roughly 5 nm) are leading safer alternatives to gadolinium-based T1 MRI contrast agents, and their longitudinal relaxivity $r_1$ is widely believed to arise cooperatively from a crystalline, magnetically ordered core and a structurally and magnetically disordered surface shell. A 2021 Nano Letters study (Miao et al.) built the decisive controlled model system — 3.8 nm $Zn_xFe_{3-x}O_4$ cores with cation-exchanged $Zn_xMn_yFe_{3-x-y}O_4$ surface shells under identical surface chemistry — and showed that Zn doping of the core (raising the saturation magnetization) and Mn substitution in the shell jointly raise $r_1$ to $20.22\ \mathrm{mM^{-1}\,s^{-1}}$, but it left the underlying mechanism unresolved at three linked levels: the atomic arrangement of the disordered shell has never been directly determined; the cation-exchange Mn substitution in the co-doped series reaches only a reported grid edge near $y \approx 0.24$, and whether that reflects a true saturation limit, and what sets it, is unknown; and the magnetic coupling between core and shell ions that is claimed to cooperatively enhance $r_1$ has never been quantitatively measured or predicted from first principles. The problem asks for a quantitative, structure-resolved account of how core magnetization and surface-ion electronic-relaxation parameters (electron spin, electronic relaxation time, and bound-water residence time) jointly determine $r_1$, comprising direct atomic or near-atomic structural determination of the disordered shell, a mechanistic account of the apparent substitution ceiling, and measured or ab initio-predicted coupling constants that quantitatively account for $r_1$ in the dopant-engineered particles.

Background

Magnetic resonance imaging (MRI) contrast agents shorten the relaxation times of nearby water protons. The longitudinal relaxation time $T_1$ (spin–lattice relaxation, which governs signal brightness in T1-weighted imaging) is shortened by paramagnetic centers; an agent's effectiveness is quantified by its longitudinal relaxivity $r_1$, the slope of the water proton relaxation rate $1/T_1$ versus agent concentration (units $\mathrm{mM^{-1}\,s^{-1}}$ of metal ions). Clinical T1 agents are gadolinium chelates such as Gd-DTPA (a linear complex of the $Gd^{3+}$ ion), whose safety concerns in renally impaired patients motivate iron-oxide-based alternatives. Ultrasmall ferrite nanoparticles (UFNPs, diameter below 5 nm) act as T1 probes with pseudoparamagnetic behavior: their small size leaves a shrunken crystalline core and a proportionally large surface layer whose ions have broken exchange bonds and defective coordination, making that layer structurally disordered and its spins magnetically frustrated — the long-established phenomenon of surface spin disorder documented for ferrite nanoparticles by Mössbauer spectroscopy and neutron scattering.

Ferrites such as magnetite $Fe_3O_4$ and maghemite $\gamma$-$Fe_2O_3$ adopt the spinel structure, in which cations occupy tetrahedral (A) and octahedral (B) sites of a close-packed oxygen lattice with antiparallel sublattice magnetizations. Diamagnetic $Zn^{2+}$ preferentially enters tetrahedral sites, diluting one sublattice and raising the net saturation magnetization $M_s$ (the high-field magnetization, in $\mathrm{emu\,g^{-1}}$) up to an optimum doping level, beyond which dilution of magnetic ions dominates. This Zn-doping strategy for ultrasmall ferrites was developed by the same group in 2017 (Zhang et al., ACS Nano, via a "dynamic simultaneous thermal decomposition" synthesis), which also noted that the conventional outer-sphere relaxation picture alone could not fully account for the observed $r_1$.

Miao et al. (Nano Letters 2021) then built the controlled model system at the center of this problem: 3.8 nm particles denoted $Zn_xF@Zn_xMn_yF$, consisting of a crystalline $Zn_xFe_{3-x}O_4$ core, a surface disordered shell of nominal composition $Zn_xMn_yFe_{3-x-y}O_4$ produced by cation exchange (partial replacement of surface Fe ions by $Mn^{2+}$ from a precursor solution), and a uniform 3,4-dihydroxyhydrocinnamic acid (DHCA) ligand coating that holds surface chemistry identical across all samples. Element-specific local-structure probes (X-ray absorption spectroscopy; specifically Zn K-edge and Mn K-edge EXAFS, extended X-ray absorption fine structure) showed Zn occupying tetrahedral spinel sites of the crystalline core, while Mn lacked the spinel-lattice spectral signatures and thus resides in the disordered surface layer. With size and surface chemistry fixed, the experiments established: (i) in the $Zn_xF$ series, $M_s$ and $r_1$ (measured at 3.0 T) rise together with $x$ up to a joint maximum at $x = 0.4$ ($M_s = 51.13\ \mathrm{emu\,g^{-1}}$, $r_1 = 13.77\ \mathrm{mM^{-1}\,s^{-1}}$) and fall at higher $x$; (ii) substituting Mn into the shell of undoped $\gamma$-$Fe_2O_3$ raises $r_1$ from 3.29 to $9.5\ \mathrm{mM^{-1}\,s^{-1}}$ at essentially unchanged $M_s$; (iii) co-optimization gives $r_1 = 20.22\ \mathrm{mM^{-1}\,s^{-1}}$ for $Zn_{0.4}F@Zn_{0.4}Mn_{0.2}F$ — 5.2-fold the pristine ultrasmall ferrite and 6.5-fold Gd-DTPA under the same conditions — with $Gd^{3+}$ and $Cu^{2+}$ shell substitutions giving 22.99 and $14.61\ \mathrm{mM^{-1}\,s^{-1}}$. The interpretation rests on the standard theory of paramagnetic relaxation: inner-sphere relaxation, from water molecules transiently bound at surface metal ions, governed by the ion's electron spin quantum number $S$, its electronic relaxation time $T_e$, and the mean residence lifetime of bound water $\tau_m$; and outer-sphere relaxation, from water diffusing near the particle's magnetic moment, governed by the core magnetization. Numerical simulations of $r_1$ versus core magnetization and $\tau_m$ (Solomon–Bloembergen–Morgan-type modeling, the classical equations combining these two channels) motivated the dopant choices, and the authors concluded that a cooperative, putatively synergistic coupling between core and shell raises $r_1$ beyond what either component achieves alone.

Three gaps were left open, and together they define the unresolved mechanism. First, the disordered shell itself has never been directly observed: in transmission electron microscopy the surface disordered structure could not be detected against the amorphous carbon grid background at this particle size, so the shell's existence rests on indirect inference (size-dependent scaling of the disordered fraction, the absence of spinel EXAFS signatures for Mn, and magnetization deficits). Second, the reach of the cation-exchange substitution is unresolved: the co-doped $Zn_xF@Zn_xMn_yF$ composition grid extends only to $y \approx 0.24$ — reportedly without further increase at higher Mn precursor concentration — whereas the parallel Mn-only surface-substitution series reached $y = 0.26$–$0.43$ by ICP-AES; whether the co-doped value reflects a true saturation limit, and whether its origin would be thermodynamic (site availability, solubility), defect- or vacancy-mediated, kinetic, or specific to the co-doped cation-exchange route, is unknown. Third, the coupling is unquantified: no measurement or ab initio calculation determines the magnetic coupling strength between specific core-site occupations and specific surface-ion electronic-relaxation parameters that would quantitatively link $M_s$, $T_e$, and $\tau_m$ to the measured $r_1$; moreover, whether the joint enhancement is non-additive cannot be read off the reported numbers — the best co-doped $r_1$ (20.22) approximately equals the sum of the separate core contribution (13.77) and the shell-only enhancement ($9.5 - 3.29 = 6.2$), yet the published 5.2-fold enhancement over the "original UFNPs" implies a pristine baseline of $\approx 3.9\ \mathrm{mM^{-1}\,s^{-1}}$ rather than the 3.29 undoped $\gamma$-$Fe_2O_3$ value, and the concentration normalization across the series is not pinned down — so even the non-additivity of the claimed synergy has not been experimentally established.

Problem Statement

Establish a quantitative, structure-resolved account of the cooperative core–shell mechanism of T1 relaxivity in dopant-engineered ultrasmall ferrite nanoparticles — taking as the model system the approximately 3.8 nm DHCA-coated $Zn_xFe_{3-x}O_4@Zn_xMn_yFe_{3-x-y}O_4$ particles — by resolving three linked questions that remain open on the record of the source study:

1. Shell structure. Determine, at atomic or near-atomic scale, the structure of the magnetically disordered surface shell: its thickness, atomic arrangement including dopant positions, composition profile, and the nature of its magnetic disorder (the character and correlations of the canted or frozen surface spins) — given the intrinsic disorder and the ensemble-averaged probes, as an ensemble-averaged structure or an appropriate distribution thereof — by direct imaging (for example aberration-corrected scanning transmission electron microscopy with element-resolved spectroscopy) or by an equivalent quantitative structure determination (for example constrained refinement of total-scattering and EXAFS data that separates core and shell contributions). In either case the determination must identify a shell structure discriminated against competing core–shell structural models, not one that several incompatible shell models reproduce equally well.

2. Substitution ceiling. Establish whether the cation-exchange Mn substitution in the co-doped $Zn_xF@Zn_xMn_yF$ series is genuinely limited — the reported co-doped composition grid extends only to $y \approx 0.24$, whereas the parallel Mn-only surface-substitution series reached $y = 0.26$–$0.43$ — and, if a limit exists, its mechanistic origin: thermodynamic (site availability, solubility), defect- or vacancy-mediated, kinetic, or specific to the co-doped cation-exchange route rather than intrinsic to it.

3. Coupling constants. Quantitatively determine the magnetic coupling between the crystalline core and the surface-shell ions: infer from measurements or predict ab initio the coupling parameters linking core-site occupation (through the core saturation magnetization $M_s$) and surface-ion electronic-relaxation parameters (electron spin $S$, electronic relaxation time $T_e$, and bound-water residence time $\tau_m$) to the longitudinal relaxivity $r_1$ — stating explicitly the parameterization adopted (for example a core–shell exchange coupling, or effective coefficients in a Solomon–Bloembergen–Morgan-type relaxivity expression combining the inner- and outer-sphere channels) — at a level that quantitatively accounts for the measured relaxivities of the dopant-engineered core–shell particles.

These three deliverables jointly support one objective: converting the qualitative, correlation-based core–shell picture into a predictive microscopic mechanism. The structural determination (1) supplies the physical basis on which any coupling model must rest; the substitution-ceiling question (2) concerns the accessible shell-composition space over which the mechanism must hold and the origin of its boundaries; and the coupling constants (3) are the quantitative link that makes the account predictive. An answer that resolves only one facet, or that restates the observed correlations without quantitative structural or coupling determination, does not complete the problem.

References

1. Miao, Y.; Zhang, H.; Cai, J.; Chen, Y.; Ma, H.; Zhang, S.; Yi, J. B.; Liu, X.; Bay, B.-H.; Guo, Y.; Zhou, X.; Gu, N.; Fan, H. Structure–Relaxivity Mechanism of an Ultrasmall Ferrite Nanoparticle $T_1$ MR Contrast Agent: The Impact of Dopants Controlled Crystalline Core and Surface Disordered Shell. Nano Letters 2021, 21, 1115–1123. DOI: 10.1021/acs.nanolett.0c04574. https://doi.org/10.1021/acs.nanolett.0c04574
2. Zhang, H.; Liu, M.; Zhou, Z.; Ye, P.; Wen, L.; Feng, Q.; Song, N.; Wang, Y.; Ma, G.; Sun, R.; Zhang, K.; Wu, Q.; Zhang, M.; Wang, Y.; Guo, Y.; Fan, H. Trace Mg-Doped Ultrasmall Iron Oxide Nanoparticles with Boosted $T_1$ Relaxivity for Highly Sensitive MRI of Subsegmental Pulmonary Embolism. ACS Nano 2026 (online August 19, 2026; volume and pages not yet assigned). DOI: 10.1021/acsnano.6c09779. https://doi.org/10.1021/acsnano.6c09779
3. Liu, M.; Feng, Q.; Zhang, H.; Guo, Y.; Fan, H. Progress in ultrasmall ferrite nanoparticles enhanced $T_1$ magnetic resonance angiography. Journal of Materials Chemistry B 2024, 12, 6521–6531. DOI: 10.1039/d4tb00803k. https://doi.org/10.1039/d4tb00803k
4. Zhang, H.; Liu, X. L.; Fan, H. M. Advances in magnetic nanoparticle-based magnetic resonance imaging contrast agents. Nano Research 2023, 16, 12531–12542. DOI: 10.1007/s12274-023-6214-9. https://doi.org/10.1007/s12274-023-6214-9
5. Komogortsev, S. V.; Stolyar, S. V.; Mokhov, A. A.; Fel'k, V. A.; Velikanov, D. A.; Iskhakov, R. S. Effect of the Core–Shell Exchange Coupling on the Approach to Magnetic Saturation in a Ferrimagnetic Nanoparticle. Magnetochemistry 2024, 10, 47. DOI: 10.3390/magnetochemistry10070047. https://doi.org/10.3390/magnetochemistry10070047
6. Che Dji, L. V.; Girardet, T.; Fleutot, S.; Bouguet-Bonnet, S. Characterization of Ultrasmall Superparamagnetic Iron Oxide via Fast Field-Cycling NMR Relaxometry at Two Temperatures. Magnetic Resonance in Chemistry 2026, 64, 312–318 (published online December 13, 2025). DOI: 10.1002/mrc.70070. https://doi.org/10.1002/mrc.70070
7. Benaissa, M.; Al Shami, A.; Labrim, H. Prediction of Transverse Relaxivity of $MnFe_2O_4$ Nanoparticles as a Function of Their Sizes and Cation Distribution. Journal of Superconductivity and Novel Magnetism 2025, 38, 242. DOI: 10.1007/s10948-025-07079-3. https://doi.org/10.1007/s10948-025-07079-3
8. Cooper, S. R.; Candler, R. O.; Cosby, A. G.; Johnson, D. W.; Jensen, K. M. Ø.; Hutchison, J. E. Evolution of Atomic-Level Structure in Sub-10 Nanometer Iron Oxide Nanocrystals: Influence on Cation Occupancy and Growth Rates. ACS Nano 2020, 14, 5480–5490. DOI: 10.1021/acsnano.9b09551. https://doi.org/10.1021/acsnano.9b09551
9. Sun, X.; Tayal, A.; Ullrich, A.; Petracic, O.; Haas, S. Phase Composition of Iron Oxide Nanoparticles Studied Using Hard X-ray Absorption Spectroscopy. The Journal of Physical Chemistry C 2023, 127, 12077–12083. DOI: 10.1021/acs.jpcc.3c01023. https://doi.org/10.1021/acs.jpcc.3c01023
10. Zhao, Z.; Sun, C.; Bao, J.; Yang, L.; Wei, R.; Cheng, J.; Lin, H.; Gao, J. Surface manganese substitution in magnetite nanocrystals enhances $T_1$ contrast ability by increasing electron spin relaxation. Journal of Materials Chemistry B 2018, 6, 401–413. DOI: 10.1039/c7tb02954c. https://doi.org/10.1039/c7tb02954c
11. Zhang, H.; Li, L.; Liu, X. L.; Jiao, J.; Ng, C.-T.; Yi, J. B.; Luo, Y. E.; Bay, B.-H.; Zhao, L. Y.; Peng, M. L.; Gu, N.; Fan, H. M. Ultrasmall Ferrite Nanoparticles Synthesized via Dynamic Simultaneous Thermal Decomposition for High-Performance and Multifunctional $T_1$ Magnetic Resonance Imaging Contrast Agent. ACS Nano 2017, 11, 3614–3631. DOI: 10.1021/acsnano.6b07684. https://doi.org/10.1021/acsnano.6b07684
12. Lentijo-Mozo, S.; Deiana, D.; Sogne, E.; Casu, A.; Falqui, A. Unexpected Insights about Cation-Exchange on Metal Oxide Nanoparticles and Its Effect on Their Magnetic Behavior. Chemistry of Materials 2018, 30, 8099–8112. DOI: 10.1021/acs.chemmater.8b04331. https://doi.org/10.1021/acs.chemmater.8b04331