A 2021 first-principles study (Li, Frauenheim, and He, ACS Appl. Mater. Interfaces 13, 36098) predicted that single-molecule junctions built on the two-dimensional van der Waals ferromagnet Fe3GeTe2 - a benzene molecule physisorbed on the surface or a benzenedithiol molecule bridging two electrodes - exhibit effectively perfect spin filtering and a giant magnetoresistance of about 6075% at zero bias, decreasing only to about 2000% across a computed +/-100 meV bias window, and that these effects are robust to adsorption site, a single Te-vacancy defect, and electrode stacking order. Whether these predictions describe a real, robust device platform remains untested: no calculation has extended beyond the +/-100 meV window or beyond the two molecules studied, no treatment of thermal fluctuations, broader contact disorder, or device variability exists, and no mixed-dimensional molecule-two-dimensional-magnet transport junction has ever been fabricated or measured. This problem asks whether perfect spin filtering and giant magnetoresistance in Fe3GeTe2-based molecular spin valves survive realistic bias, molecular, and disorder conditions, and whether such junctions can be experimentally realized and measured.

Background

Magnetoresistance (MR) is the change of a device's electrical resistance with the relative orientation of two magnetizations; a spin valve is a two-terminal device whose resistance switches between a low state, when two ferromagnetic electrodes are magnetized in parallel (P), and a high state when they are antiparallel (AP), conventionally quantified as MR=(GP−GAP)/GAPMR=(G_P-G_AP)/G_AP using the conductances (or the analogous current ratio). In a magnetic tunnel junction (MTJ) the electrodes are separated by an insulating or molecular spacer, and the MR arises from spin-dependent tunneling. A single-molecule junction is a device in which one molecule bridges two electrodes - fabricated, for example, with a scanning tunneling microscope (STM) tip contacting a molecule on a surface, or with a mechanically controllable break junction (MCBJ) - so that charge transport proceeds by tunneling or resonant transport through discrete molecular orbitals. Spin filtering means the transmission is almost entirely carried by one spin channel, quantified by a spin polarization SP=(Iup−Idown)/(Iup+Idown)SP=(I_{up}-I_{down})/(I_{up}+I_{down}) close to 100%. First-principles spin-transport predictions combine density-functional theory (DFT) with the nonequilibrium Green's function (NEGF) method, which computes the spin-resolved Landauer transmission T(E)T(E) at zero bias and the spin-resolved current at finite bias; the natural conductance unit is G0=e2/hG_0=e^2/h per spin channel.

Molecular spin valves built on conventional ferromagnetic metals (Ni, Co, Fe) have long been limited to MR of a few tens of percent, because their conduction is dominated by nearly unpolarized s states and their performance is strongly contact-sensitive. Fe3GeTe2 (FGT) is a layered itinerant ferromagnet with perpendicular magnetic anisotropy that can be exfoliated into few-layer flakes; since its isolation it has become a workhorse two-dimensional (2D) van der Waals (vdW) magnet, and all-inorganic vdW spin valves with hexagonal boron nitride (hBN), molybdenum disulfide (MoS2), or tellurene (a two-dimensional form of elemental tellurium) spacers between FGT electrodes have already been demonstrated experimentally. Li, Frauenheim, and He (2021) asked what happens when the spacer is instead a single organic molecule - a mixed-dimensional junction combining a zero-dimensional (0D) molecule with 2D magnetic electrodes. Using DFT+NEGF they studied a benzene molecule physisorbed on FGT (STM-like geometry) and a benzenedithiol (BDT, a benzene carrying two thiol anchor groups) bridging two FGT electrodes in a break-junction-like setup. They predicted effectively perfect spin filtering, and a zero-bias MR of about 6075% for the FGT/BDT/FGT spin valve - more than two orders of magnitude above the corresponding conventional-metal junctions - decreasing to roughly 2000% but persisting over the computed bias window Vb∈[−100,+100]V_b\in[-100,+100] meV, with spin polarization near 100% throughout. They traced the effect to the intrinsic spin-polarized band structure of the FGT electrodes rather than to molecule-specific chemistry (the molecular choice mainly sets the absolute conductance through the majority-spin hybridization strength at the interface), and found the spin selectivity robust against three tested S adsorption sites, a single missing-Te interface defect, and different electrode stacking orders. What the paper did not do defines the open problem: it did not go beyond +/-100 meV of bias, did not treat thermal fluctuations, phonons, contact disorder beyond the tested cases, or device-to-device variability, examined only benzene and BDT, and involved no experiment.

Problem Statement

The source paper's predictions define one umbrella objective: establish whether perfect spin filtering and giant magnetoresistance are genuine, robust properties of the mixed-dimensional Fe3GeTe2-molecule spin-valve platform, rather than artifacts of an idealized low-bias, two-molecule, zero-temperature calculation. Four linked questions define the objective; a solution must resolve at least one of them completely, meeting the per-question completeness requirements stated after the list, and state which it resolves.

(1) Bias extension: does the near-100% spin polarization and the giant (order 10310^3%) MR of Fe3GeTe2/molecule/Fe3GeTe2 junctions persist at bias voltages beyond the finite-bias window computed in the source (reported as Vb∈[−100,+100]V_b\in[-100,+100] meV), where molecular resonances, inelastic channels, and bias-dependent electrode band structure enter?

(2) Molecular design space: beyond benzene and BDT, which molecules and anchor chemistries can tune the absolute conductance (set by the majority-spin hybridization strength at the molecule-electrode interface) without degrading the near-perfect spin selectivity?

(3) Realism: do the predicted effects survive the conditions of actual devices - thermal fluctuations and electron-phonon coupling at operating temperature, contact disorder beyond the specific adsorption sites, single Te-vacancy defect, and stacking orders tested in the source, and device-to-device variability?

(4) Experimental realization: can a mixed-dimensional molecular spin valve with a 2D vdW ferromagnet electrode (Fe3GeTe2-molecule-Fe3GeTe2, or the Fe3GeTe2-molecule-metal STM-like geometries also computed in the source) be fabricated and measured - for example by STM, break-junction, or transferred-flake techniques - so that the spin-dependent conductance and its dependence on the magnetization configuration are directly observed?

A complete answer meets the following per-question requirements for whichever question it resolves: (1) reports the spin polarization and magnetoresistance at bias magnitudes beyond the source window, the mechanisms named there being considered rather than each requiring a separate calculation; (2) reports the conductance and spin polarization for a finite, explicitly enumerated set of molecules and anchor chemistries beyond benzene and BDT; (3) reports the spin polarization and magnetoresistance for each of the listed perturbations treated individually, the listed conditions forming a conjunctive set rather than an indicative scope; and (4) reports the spin-dependent conductance of a fabricated or measured junction in both magnetization configurations. Throughout, an effect counts as surviving when the spin polarization remains near unity and the magnetoresistance remains giant - operationally, spin polarization of at least about 90% and magnetoresistance of at least about 10310^3%, the scale of the source's pristine junctions - rather than degrading to the few-tens-of-percent level of conventional-metal molecular spin valves.

The common objective is advanced by establishing the junction's spin-filtering and MR behavior across realistic bias and molecular choices with quantified robustness, and/or by experimentally realizing such a junction - which the source notes is experimentally feasible, for example via STM or break-junction setups - and demonstrating spin-dependent magnetotransport through a molecule coupled to 2D vdW ferromagnetic electrodes. Answers to questions (1)-(3) may be given in theory, in experiment, or both, whereas question (4) requires experiment.

References

  1. Dongzhe Li, Thomas Frauenheim, Junjie He, 'Robust Giant Magnetoresistance in 2D Van der Waals Molecular Magnetic Tunnel Junctions', ACS Applied Materials & Interfaces 13, 36098-36105 (2021). DOI: 10.1021/acsami.1c10673, https://doi.org/10.1021/acsami.1c10673
  2. Zhe Wang, Deepak Sapkota, Takashi Taniguchi, Kenji Watanabe, David Mandrus, Alberto F. Morpurgo, 'Tunneling Spin Valves Based on Fe3GeTe2/hBN/Fe3GeTe2 van der Waals Heterostructures', Nano Letters 18, 4303-4308 (2018). DOI: 10.1021/acs.nanolett.8b01278, https://doi.org/10.1021/acs.nanolett.8b01278
  3. Hailong Lin, Faguang Yan, Ce Hu, Quanshan Lv, Wenkai Zhu, Ziao Wang, Zhongming Wei, Kai Chang, Kaiyou Wang, 'Spin-Valve Effect in Fe3GeTe2/MoS2/Fe3GeTe2 van der Waals Heterostructures', ACS Applied Materials & Interfaces 12, 43921-43926 (2020). DOI: 10.1021/acsami.0c12483, https://doi.org/10.1021/acsami.0c12483
  4. Jing Zeng, 'Spin filtering and magnetoresistance effects in a dithiane ultra-short single-molecule insulator', Physical Chemistry Chemical Physics 24, 27804-27811 (2022). DOI: 10.1039/d2cp03729g, https://doi.org/10.1039/d2cp03729g
  5. Jun-Chen Ou, Jing Zeng, 'Nearly perfect spin-filtering effect with tunable spin-polarization direction and huge magnetoresistance in molecular spin valves', Physics Letters A 470, 128784 (2023). DOI: 10.1016/j.physleta.2023.128784, https://doi.org/10.1016/j.physleta.2023.128784
  6. Jing Zeng, 'Large magnetoresistance and spin negative differential resistance in photoisomers-based molecular spin valves', Physics Letters A 525, 129935 (2024). DOI: 10.1016/j.physleta.2024.129935, https://doi.org/10.1016/j.physleta.2024.129935
  7. Anita Halder, Declan Nell, Antik Sihi, Akash Bajaj, Stefano Sanvito, Andrea Droghetti, 'Half-Metallic Transport and Spin-Polarized Tunneling through the van der Waals Ferromagnet Fe4GeTe2', Nano Letters 24, 9221-9228 (2024). DOI: 10.1021/acs.nanolett.4c01479, https://doi.org/10.1021/acs.nanolett.4c01479
  8. Masoumeh Davoudiniya, Biplab Sanyal, 'Efficient spin filtering through Fe4GeTe2-based van der Waals heterostructures', Nanoscale Advances 6, 6278-6289 (2024). DOI: 10.1039/d4na00639a, https://doi.org/10.1039/d4na00639a
  9. Xiangyu Zeng, Liang Zhang, Yang Zhang, Fazhi Yang, Liqin Zhou, Yong Wang, Cizhe Fang, Xiaoxi Li, Siying Zheng, Yang Liu, Yan Liu, Xiaozhi Wang, Yue Hao, Genquan Han, 'Spin valve effect in the van der Waals heterojunction of Fe3GeTe2/tellurene/Fe3GeTe2', Applied Physics Letters 125, 092406 (2024). DOI: 10.1063/5.0215304, https://doi.org/10.1063/5.0215304
  10. Zhongzhong Luo, Zhihao Yu, Xiangqian Lu, Wei Niu, Yao Yu, Yu Yao, Fuguo Tian, Chee Leong Tan, Huabin Sun, Li Gao, Wei Qin, Yong Xu, Qiang Zhao, Xiang-Xiang Song, 'Van der Waals Magnetic Electrode Transfer for Two-Dimensional Spintronic Devices', Nano Letters 24, 6183-6191 (2024). DOI: 10.1021/acs.nanolett.4c01885, https://doi.org/10.1021/acs.nanolett.4c01885
  11. Mayank Sharma, Junhyeon Jo, Garen Avedissian, Bertug Bayindir, Jun Kang, Hasan Sahin, Felix Casanova, Marco Gobbi, Luis E. Hueso, 'Molecular Engineering of Exchange Bias in Fe3GeTe2/Molecule Heterostructures', ACS Applied Electronic Materials 7, 9204-9211 (2025). DOI: 10.1021/acsaelm.5c01598, https://doi.org/10.1021/acsaelm.5c01598
  12. Anita Halder, Declan Nell, Akash Bajaj, Stefano Sanvito, Andrea Droghetti, 'Spin-dependent transport in Fe3GaTe2 and Fe_nGeTe_2 (n = 3-5) van der Waals ferromagnets for magnetic tunnel junctions', 2D Materials 13, 025002 (2026). DOI: 10.1088/2053-1583/ae4040, https://doi.org/10.1088/2053-1583/ae4040