Sliding graphene over MnPS3 switches its calculated magnetic state between two compensated phases
Calculations by Xiong and Jia show a graphene-MnPS3 stack switching between altermagnetic and ferrimagnetic states across a 28 meV/f.u. sliding barrier. Under out-of-plane pressure the computed spin splitting reaches 72.2 meV and 80.2 meV in the two phases.
The Scientist · Science desk
Calculated for the graphene-MnPS3 stack: a sliding barrier of 28 meV per formula unit between the AM and LC-FiM phases; under out-of-plane pressure, exchange spin splitting of 72.2 meV in the AM state and 80.2 meV in the LC-FiM state.
| Figure | Value | Note | Claim |
|---|---|---|---|
| Sliding barrier between AM and LC-FiM phases | 28 meV/f.u. | 5 | |
| Exchange spin splitting, AM state | 72.2 meV | under out-of-plane pressure | 6 |
| Exchange spin splitting, LC-FiM state | 80.2 meV | under out-of-plane pressure | 6 |
What happened
- The graphene layer breaks the PT symmetry of Neel-antiferromagnetic MnPS3, which sets up the spin splitting in both phases.
- In hollow stacking, joint rotation and mirror operations survive, giving an i-wave altermagnet whose spin splitting depends on momentum.
- In top stacking, every symmetry connecting the two spin sublattices is removed, and the result is a Luttinger-compensated ferrimagnet with globally spin-split bands.
Why it matters
- What's now possible capability If a real stack behaves as computed, one material would offer two zero-net-magnetization phases with spin-split bands, selected by layer position alone.
- What it limits constraint Every figure is a computed value, so none of it is a device specification until someone builds the stack and measures it.
- What doesn't add up contradiction The clean two-state switch describes ideal stackings, while the same paper's mismatch result points to mixed domains in a physical sample.
- What it sets up precedent The authors argue stacking symmetry could program compensated magnetism across van der Waals materials, a claim one material pair cannot settle alone.
Both phases combine zero net magnetization with spin-split bands [8]. We'd expect a net-moment measurement to read zero in either one. The authors point to a different readout: the domains of the moire superlattice carry distinct magneto-optical Kerr signatures [7]. In our view that is the first place an experimentalist would look, since the two states differ by interlayer translation alone [5].
Pressure is the second control. According to the authors, out-of-plane pressure strengthens the interfacial coupling, with enhanced orbital hybridization and asymmetric built-in polarization [9]. At the quoted values the ferrimagnetic phase's splitting is 8.0 meV larger than the altermagnetic phase's [13]. The abstract gives neither the ambient-pressure splittings nor the pressure applied, so the size of the amplification cannot be read from it.
Everything here is a calculation. The method is first-principles calculation plus symmetry analysis [1], and the abstract describes no experiment. The thing this doesn't tell you is how a physical stack would be moved between hollow and top registry and held there.
The authors also go beyond the ideal-stacking limit. Graphene and MnPS3 have an intrinsic lattice mismatch, and it produces a moire superlattice of alternating altermagnetic and ferrimagnetic domains [7]. The 28 meV/f.u. barrier belongs to the two ideal stackings [5]. We'd expect a mismatched sample to hold both phases at once, with any switching moving the boundaries between them. That is our inference from the domain pattern the abstract reports.
The paper, by Xiong and Jia, is in npj Computational Materials (2026) [10]. It was funded by the National Natural Science Foundation of China and Henan provincial sources, and the funders had no role in the study design or analysis [11]. Computing resources came from the Core Facility of Wuhan University and HZWTECH [12].
What to watch
- An experiment that places graphene on MnPS3 in controlled registry and reads the Kerr response of the domains.
- The pressure at which the 72.2 meV and 80.2 meV splittings are reached, and the ambient values they are measured against.
- Independent calculations that reproduce the 28 meV/f.u. barrier and the two-phase picture in the lattice-mismatched structure.
Clarity's read
What the record supports and how the coverage leans. The claims behind it follow.
Reality
- Evidence40
- Adoption
- Insufficient
- Hype gap+20
- Incentives
- Insufficient
- Confidence55
Claim ledger
Ranked by verification strength, evidence, and original report placement.
- [1]
Using first-principles calculations and symmetry analysis, Xiong and Jia show that a graphene-MnPS3 van der Waals heterostructure can host both altermagnetic (AM) and Luttinger-compensated ferrimagnetic (LC-FiM) states and their interconversion within a single platform.
- [2]
The graphene layer breaks the PT symmetry of Neel-antiferromagnetic MnPS3.
- [3]
Hollow stacking preserves joint rotation and mirror operations, yielding an i-wave altermagnetic state with momentum-dependent spin splitting.
- [4]
Top stacking removes all sublattice-connecting symmetries and stabilizes a Luttinger-compensated ferrimagnetic state with globally spin-split bands.
- [5]
The two phases are separated by an ultralow sliding barrier of 28 meV/f.u., allowing reversible switching by interlayer translation alone.
- [6]
Out-of-plane pressure amplifies the exchange spin splitting to 72.2 meV in the AM state and 80.2 meV in the LC-FiM state.
- [7]
Beyond the ideal-stacking limit, the intrinsic lattice mismatch generates a moire superlattice of alternating AM and LC-FiM domains with distinct magneto-optical Kerr signatures.
- [8]
Both altermagnetism and Luttinger-compensated ferrimagnetism combine zero net magnetization with spin-split electronic bands.
- [9]
The pressure-driven enhancement of spin splitting arises from strengthened interfacial coupling accompanied by enhanced orbital hybridization and asymmetric built-in polarization.
- [10]
The paper, Xiong, W. and Jia, Y., 'Switchable altermagnetism and Luttinger-compensated ferrimagnetism in graphene-MnPS3 heterostructures', is cited as npj Comput Mater (2026).
- [11]
The work was supported by the National Natural Science Foundation of China, the National Natural Science Foundation of Henan Province and Henan Academy of Sciences funds; the funders had no role in the design of the study, analysis, interpretation, manuscript preparation or the decision to submit.
- [12]
The authors thank the Core Facility of Wuhan University and HZWTECH for providing the computational resources.
- [13]
The computed LC-FiM spin splitting under pressure is 8.0 meV larger than the AM spin splitting (80.2 meV minus 72.2 meV).
- [14]
The authors state that these results establish interfacial symmetry engineering as a general route to programmable compensated magnetism in van der Waals materials.
Get this ledger inside ChatGPT. Ask about any of our stories and the answer brings its figure, each claim’s verdict and the sources. Free, no account.
Sources
1 independent publisher whose own reporting we read for this story.
Topics and entities
Follow any of these and your For You feed starts watching them — no settings page required.
Topics
- Van der Waals heterostructuresFollow
- AltermagnetismFollow
- Spintronics and Spin TexturesFollow