Microscopic mechanism of anomalously slow magnetization switching and nonmonotonic temperature-dependent conductance switching of Mn12 single-molecule magnets on graphene quantum dots

A 2024 experiment (preprint arXiv:2407.21156; published as Carbon 248, 121093, 2026) achieved gate-free electrical detection of magnetization switching of the archetypal single-molecule magnet Mn12-CHCl2 deposited on graphene quantum dots, at temperatures up to 70 K, and extracted both the molecule-graphene exchange interaction (about 0.1 meV) and an antiferromagnetic quantum-dot-mediated intermolecular interaction (0.5-1.4 meV) directly from the electrical response. The authors explicitly leave two mechanistic puzzles open: the magnetization switching of the deposited molecules is anomalously slow compared with Mn12-acetate crystals and aligned powders, with the responsible mechanism unidentified; and the electrically detected conductance switching disappears between 30-40 K and reappears near 50 K, plausibly involving thermally populated excited spin manifolds (e.g. the S=9 manifold near 40 K) but without an established microscopic pathway. This problem asks for a predictive microscopic account of the switching dynamics and proximity exchange in this SMM-on-graphene platform: identify the mechanism of the slowed relaxation and establish the pathway of the nonmonotonic temperature dependence.

Background

Single-molecule magnets (SMMs) are metal-organic clusters in which a large molecular spin combined with magnetic anisotropy produces a barrier against reversal of the magnetization, so that each molecule behaves as a tiny magnet with memory. The archetypal example is Mn12-acetate: a mixed-valence manganese cluster with spin ground state $S=10$ whose dominant easy-axis anisotropy (the preference of the spin for one molecular axis, parameterized by $D$ in the spin Hamiltonian $H = D S_z^2$) creates a barrier of roughly 70 K between the two lowest states $m_S=\pm 10$ (the two extreme projections of the spin along the easy axis). The magnetization can nevertheless reverse by quantum tunneling or by thermally assisted tunneling through excited $m_S$ levels; in bulk crystals and powders this produces stepped hysteresis loops and a characteristic relaxation-rate temperature dependence established since the mid-1990s (Friedman et al. 1996; Novak et al. 1995). At low enough temperature (the blocking temperature $T_B$, about 3 K for Mn12-CHCl2) relaxation freezes entirely.

Applications of SMMs to quantum computing, high-density memory, and spintronics require few or isolated molecules on device substrates rather than bulk crystals. The supramolecular spin-valve geometry (Urdampilleta et al. 2011; Candini et al. 2011) is one route: molecules adsorbed on a carbon nanotube or graphene quantum dot (GQD) — a nanoscale island of graphene whose discrete electronic states are probed by transport — modify the dot's conductance depending on their magnetic alignment, so the molecular magnetic state can be read out electrically. In such devices the molecule-graphene proximity exchange interaction spin-splits the dot states, and a quantum-dot-mediated intermolecular interaction can arise when molecules divide one dot into coupled sub-dots (Krainov et al. 2017).

The source work (Alqahtani et al., arXiv:2407.21156; Carbon 248, 121093 (2026)) fabricated GQDs patterned from epitaxial graphene on silicon carbide (SiC) — a single carbon monolayer grown directly on a SiC crystal — dry-deposited powder of the Mn12 derivative Mn12-CHCl2 onto them, and observed discrete jumps between two conductance states from a few kelvin up to about 70 K. From the ratio of the two conductance values they extracted, without fitting parameters or a gate electrode, a temperature-independent molecule-graphene exchange splitting of about $\Delta E \approx 0.1$ meV (consistent in order of magnitude with their density-functional-theory estimate of up to 0.07 meV for a molecule 3.3 Å above graphene), and from field-cooling experiments (cooling the device in an applied magnetic field so that the molecular moments align) they bounded an antiferromagnetic dot-mediated intermolecular interaction $0.5\,\mathrm{meV} < \Delta E_{\mathrm{IM}} < 1.4\,\mathrm{meV}$.

Three observations in this experiment remain unexplained. First, the rate of magnetization switching inferred from the conductance jumps is anomalously slow compared with previous measurements on Mn12-acetate crystals and aligned powders; the authors note as the only hint that the antiferromagnetic intermolecular interaction mediated by the graphene might affect the relaxation rate, by analogy with the slow relaxation found in antiferromagnetically coupled Mn4 assemblies (Lecren et al. 2005), and flag the question as an avenue for future investigation. Second, because the molecules are deposited as powder without positional or orientational control, the extracted exchange interaction is a device-dependent average; mapping its spatial and orientational dependence, and correlating microscopic molecule-graphene geometry with the transport signatures, requires controlled single-molecule placement or systematic geometry control. Third, the conductance jumps disappear between 30 K and 40 K and reappear near 50 K; the authors suggest this could be related to thermal population of spin manifolds other than the $S=10$ ground state — the $S=10 \to S=9$ transition is known to lie near 40 K from electron paramagnetic resonance (EPR, a microwave spectroscopy that resolves molecular spin states) and theoretical studies (Park et al. 2004; Petukhov et al. 2004) — but the microscopic transport pathways and the role of higher-spin manifolds are not established. Note also that deposition on graphene is known to change the molecules' charge state (Zhu et al. 2020 documented charge transfer from Mn12 to graphene), so an altered molecular environment at the surface is a further candidate cause of the slowed relaxation.

Problem Statement

For Mn12-CHCl2 single-molecule magnets deposited on graphene quantum dots — the platform of Alqahtani et al. (Carbon 248, 121093, 2026) — determine, with a quantitative and predictive microscopic account:

1. The mechanism responsible for the anomalously slow rate of magnetization switching of the deposited molecules compared with previous measurements on Mn12-acetate crystals and aligned powders. The source identifies one candidate cause: the antiferromagnetic intermolecular interaction mediated by the graphene quantum dot (bounded in the source at 0.5 meV < $\Delta E_{\mathrm{IM}}$ < 1.4 meV). An altered molecular environment at the graphene surface (e.g. substrate-induced charge transfer, documented for Mn12 on graphene by Zhu et al. 2020, or modified anisotropy) is a further candidate drawn from the adjacent literature rather than from the source.

2. The microscopic pathway of the nonmonotonic temperature dependence of the electrically detected conductance switching — jumps present at low temperature, disappearing between 30 K and 40 K, and reappearing near 50 K — including whether and how thermally populated excited spin manifolds (such as the $S=9$ manifold lying near 40 K) produce or gate the observed conductance jumps.

The common scientific objective is a validated microscopic understanding of how proximity to graphene and dot-mediated intermolecular coupling govern the magnetization switching dynamics of single-molecule magnets electrically read out on this platform; the two parts jointly serve that objective, addressing the switching-rate anomaly and the temperature profile of the electrically detected switching.

A completed account must satisfy the following completion criterion. It must be quantitative and predictive in the specific sense of reproducing both observables reported by the source: the temperature dependence of the magnetization switching rate, and the temperature profile of the electrically detected conductance switching (jumps present at low temperature, absent between 30 K and 40 K, and reappearing near 50 K). Agreement is judged at the order-of-magnitude level, against the energy and temperature scales stated above (the molecule-graphene exchange splitting of about 0.1 meV, the intermolecular coupling 0.5 meV < $\Delta E_{\mathrm{IM}}$ < 1.4 meV, the anisotropy barrier of roughly 70 K, and the $S=9$ manifold near 40 K); an exact numerical fit to the source's measured rate curve or switching window is not required, since those data are not part of this task package. Each numbered part must additionally specify an explicit microscopic mechanism rather than only naming candidate causes. Both numbered parts are mandatory and are assessed independently, while jointly serving the common scientific objective.

References

1. Alqahtani, A.; Henry, D.; Havlíček, L.; St. Marie, L.; Hrubý, J.; Sojka, A.; Hale, M.; Felsenfeld, S.; El Fatimy, A.; Myers-Ward, R. L.; Gaskill, D. K.; Nemec, I.; Neugebauer, P.; Liu, A. Y.; Barbara, P. 'Electrical Detection of Magnetization Switching in Single-Molecule Magnets', Carbon 248, 121093 (2026). DOI: 10.1016/j.carbon.2025.121093; arXiv:2407.21156. https://arxiv.org/abs/2407.21156
2. Friedman, J. R.; Sarachik, M. P.; Tejada, J.; Maciejewski, J.; Ziolo, R. 'Steps in the hysteresis loops of a high-spin molecule', Journal of Applied Physics 79, 6031-6033 (1996). DOI: 10.1063/1.361837. https://doi.org/10.1063/1.361837
3. Novak, M. A.; Sessoli, R.; Caneschi, A.; Gatteschi, D. 'Magnetic properties of a Mn cluster organic compound', Journal of Magnetism and Magnetic Materials 146, 211-213 (1995). DOI: 10.1016/0304-8853(94)00860-4. https://doi.org/10.1016/0304-8853(94)00860-4
4. Urdampilleta, M.; Klyatskaya, S.; Cleuziou, J.-P.; Ruben, M.; Wernsdorfer, W. 'Supramolecular spin valves', Nature Materials 10, 502-506 (2011). DOI: 10.1038/nmat3050. https://doi.org/10.1038/nmat3050
5. Candini, A.; Klyatskaya, S.; Ruben, M.; Wernsdorfer, W.; Affronte, M. 'Graphene Spintronic Devices with Molecular Nanomagnets', Nano Letters 11, 2634-2639 (2011). DOI: 10.1021/nl2006142. https://doi.org/10.1021/nl2006142
6. Krainov, I. V.; Klier, J.; Dmitriev, A. P.; Klyatskaya, S.; Ruben, M.; Wernsdorfer, W.; Gornyi, I. V. 'Giant Magnetoresistance in Carbon Nanotubes with Single-Molecule Magnets TbPc$_2$', ACS Nano 11, 6868-6880 (2017). DOI: 10.1021/acsnano.7b02014. https://doi.org/10.1021/acsnano.7b02014
7. Zhu, X. C.; Hale, A.; Christou, G.; Hebard, A. F. 'Electronegative ligands enhance charge transfer to Mn12 single-molecule magnets deposited on graphene', Journal of Applied Physics 127, 064303 (2020). DOI: 10.1063/1.5128329. https://doi.org/10.1063/1.5128329
8. Park, K.; Pederson, M. R.; Hellberg, C. S. 'Properties of low-lying excited manifolds in Mn12 acetate', Physical Review B 69, 014416 (2004). DOI: 10.1103/PhysRevB.69.014416. https://doi.org/10.1103/PhysRevB.69.014416
9. Petukhov, K.; Hill, S.; Chakov, N. E.; Abboud, K. A.; Christou, G. 'Evidence for the S=9 excited state in Mn12-bromoacetate measured by electron paramagnetic resonance', Physical Review B 70, 054426 (2004). DOI: 10.1103/PhysRevB.70.054426. https://doi.org/10.1103/PhysRevB.70.054426
10. Lecren, L.; Roubeau, O.; Coulon, C.; Li, Y.; Le Goff, X.; Wernsdorfer, W.; Miyasaka, H.; Clérac, R. 'Slow Relaxation in a One-Dimensional Rational Assembly of Antiferromagnetically Coupled [Mn4] Single-Molecule Magnets', Journal of the American Chemical Society 127, 17353-17363 (2005). DOI: 10.1021/ja0551685. https://doi.org/10.1021/ja0551685
11. Bunker, C.; Hoffman, S.; Liu, S.; Zhang, X.-G.; Cheng, H.-P. 'Using Near-Flat-Band Electrons for Read-Out of Molecular Spin Qubit Entangled States', ACS Nano 20, 5602-5612 (2026). DOI: 10.1021/acsnano.5c16530; arXiv:2602.21322. https://arxiv.org/abs/2602.21322
12. Meunier, V.; Ania, C.; Barbara, P.; Bianco, A.; Cao, M.-S.; Chen, Y.; Ewels, C.; Girao, E. C.; Kim, Y. A.; Liu, C.; Lv, R.; Morallón, E.; Nishihara, H.; Terrones, M.; Yu, W. 'Carbon science perspective in 2026: Current research and future challenges', Carbon 258, 121795 (2026). DOI: 10.1016/j.carbon.2026.121795. https://doi.org/10.1016/j.carbon.2026.121795